kuoy81347
359 Pages

kuoy81347

Course Number: KUOY 81347, Fall 2009

College/University: University of Texas

Word Count: 49433

Rating:

Document Preview

Copyright by Yao-chen Kuo 2004 The Dissertation Committee for Yao-chen Kuo Certifies that this is the approved version of the following dissertation: HIGHWAY EARTHWORK AND PAVEMENT PRODUCTION RATES FOR CONSTRUCTION TIME ESTIMATION Committee: James T. O'Connor, Supervisor Carl T. Haas John D. Borcherding Ellen M. Rathje Daniel Powers HIGHWAY EARTHWORK AND PAVEMENT PRODUCTION RATES FOR CONSTRUCTION TIME...

Unformatted Document Excerpt
Coursehero >> Texas >> University of Texas >> KUOY 81347

Course Hero has millions of student submitted documents similar to the one
below including study guides, practice problems, reference materials, practice exams, textbook help and tutor support.

Course Hero has millions of student submitted documents similar to the one below including study guides, practice problems, reference materials, practice exams, textbook help and tutor support.

by Copyright Yao-chen Kuo 2004 The Dissertation Committee for Yao-chen Kuo Certifies that this is the approved version of the following dissertation: HIGHWAY EARTHWORK AND PAVEMENT PRODUCTION RATES FOR CONSTRUCTION TIME ESTIMATION Committee: James T. O'Connor, Supervisor Carl T. Haas John D. Borcherding Ellen M. Rathje Daniel Powers HIGHWAY EARTHWORK AND PAVEMENT PRODUCTION RATES FOR CONSTRUCTION TIME ESTIMATION by Yao-chen Kuo, B.S., M.S. DISSERTATION Presented to the Faculty of the Graduate School of The University of Texas at Austin in Partial Fulfillment of the Requirements for the Degree of DOCTOR OF PHILOSOPHY The University of Texas at Austin August, 2004 To my wife, Lichen Huang, my daughter, Megan Kuo, and my son, Rick Kuo Acknowledgements I would like to express my deepest appreciation to my supervisor, Dr. James T. O'Connor for his solid support, valuable guidance and patience. Without his My valuable guidance and support, this accomplishment can not be achieved. special thanks are extended to my advisory committee members, Dr. John D. Borcherding, Dr. Carl T. Haas, Dr. Ellen M. Rathje, and Dr. Daniel Powers, for their valuable advice. In addition, I would like to thank the project monitoring committee members of the TxDOT research project 0-4416 for their assistance on this study. I would like to thank my parents for their support. Finally, I would like to express my deepest appreciation to my wife, Lichen Huang, for her love, encouragement and support throughout this study. v Highway Earthwork and Pavement Production Rates for Construction Time Estimation Publication No. ______________________ Yaochen Kuo, Ph.D. The University of Texas at Austin, 2004 Supervisor: James T. O'Connor In recent decades, the complexity and size of highway construction projects have increased dramatically. Because of this change, Contract Time estimates for most construction projects have been based on the critical path method (CPM). However, with the use of the CPM, many problems associated with unrealistic contract timing are encountered. In order to solve these problems, many transportation agencies have attempted to establish a standard process to estimate Contract Time with the belief that reasonable Contract Time estimation should rely on realistic Production Rates. Personal experience, historical records, and existing standards are usually used for Production Rates estimation. These sources are often unreliable because they do not include the effects of important drivers on Production Rates. vi Many studies on construction productivity have been conducted. However, most of them focus on cost management rather than construction time estimation. Little information is available on Production Rates for construction time estimation. This study is intended to be a reference tool for the highway construction industry to schedule and plan construction time. The purpose of this research study was to investigate the Production Rates of seven major Work Items in Earthwork- and Pavement-related construction. In addition, drivers that are known at the design stage and have a significant impact on Production Rates were identified and the effects of those drivers were quantified. vii Table of Contents LIST OF TABLES ................................................................................................ xv LIST OF FIGURES.............................................................................................. xix CHAPTER I: INTRODUCTION ............................................................................ 1 1.1 Research Background and Motivation ..................................................... 1 1.2 Research Objectives ................................................................................. 4 1.3 Research Scope Limitations ..................................................................... 5 1.4 Structure of Dissertation........................................................................... 5 CHAPTER II: LITERATURE REVIEW................................................................ 7 2.1 Definitions of Productivity....................................................................... 8 2.2 Approaches for Measuring Productivity ................................................ 10 2.3 Sources of Production Rates .................................................................. 11 2.3.1 RS Means ................................................................................... 13 2.3.2 Contract Time Determination System (CTDS) .......................... 14 2.3.3 Historical Records ...................................................................... 17 2.4 General Factors Affecting Productivity ................................................. 17 2.4.1 Weather ...................................................................................... 18 2.4.2 Scheduled Overtime ................................................................... 20 2.4.3 Disruptions ................................................................................. 22 2.4.4 Congestion and Accessibility ..................................................... 23 2.4.5 Region ........................................................................................ 24 2.4.6 Learning curve............................................................................ 25 2.4.7 Other factors............................................................................... 25 2.5 Productivity Studies of Earthwork and Pavement.................................. 26 2.5.1 Earthmoving Production Rates and Match Factor ..................... 26 2.5.2 Truck Payload ............................................................................ 29 viii 2.5.3 Rainfall ....................................................................................... 29 2.5.4 Advance of Technology ............................................................. 30 2.5.5 Traffic......................................................................................... 31 2.5.6 Construction Productivity Associated with Concrete Pavement .................................................................................... 32 2.6 Methods of Productivity Analysis.......................................................... 34 2.7 Advancing to Present Research.............................................................. 35 CHAPTER III: RESEARCH METHODOLOGY................................................. 36 3.1 Overview of Research Methodology...................................................... 36 3.2 Research Formulation ............................................................................ 37 3.3 Survey to Select Targeted Work Items .................................................. 38 3.4 Planning for Data Collection.................................................................. 39 3.4.1 Data Collection Process ............................................................. 40 3.5 Data Collection....................................................................................... 43 3.6 Data Analysis ......................................................................................... 45 3.6.1 Descriptive Statistics and Box Plots........................................... 45 3.6.2 Test of the Difference of Mean Observed Production Rates and Average CTDS Production Rates ........................................ 48 3.6.3 Driver Analysis .......................................................................... 48 CHAPTER IV: DATA COLLECTION PLAN AND EXECUTION ................... 56 4.1 Research Hypotheses.............................................................................. 56 4.2 Candidate Drivers of Targeted Work Items ........................................... 57 4.2.1 Candidate Drivers for Excavation .............................................. 57 4.2.2 Candidate Drivers for Embankment........................................... 61 4.2.3 Candidate Drivers for Lime-Treated Sub-grade......................... 65 4.2.4 Candidate Drivers for Aggregate Base Course .......................... 69 4.2.5 Candidate Drivers for Hot Mix Asphalt Pavement .................... 72 4.2.6 Candidate Drivers for Slip-form Concrete Pavement ................ 77 ix 4.2.7 Candidate Drivers for Conventional Form Concrete Pavement .................................................................................... 82 4.3 Data Collection Tools............................................................................. 88 4.3.1 Project-Level Data Collection Tool ........................................... 88 4.3.2 Work Zone-level and Work Item-level ...................................... 89 4.3.3 Work Item Sheets ....................................................................... 96 4.4 Pilot Data Collection .............................................................................. 96 4.5 Data Collection....................................................................................... 97 4.5.1 Excavation.................................................................................. 97 4.5.2 Embankment............................................................................... 99 4.5.3 Lime-Treated Sub-grade .......................................................... 100 4.5.4 Aggregate Base Course ............................................................ 102 4.5.5 Hot Mix Asphalt Pavement ...................................................... 104 4.5.6 Slip-form Concrete Pavement .................................................. 105 4.5.7 Conventional Form Concrete Pavement .................................. 107 4.6 Summary of Study Districts and Projects ............................................ 107 4.7 Summary of Production Rate Data....................................................... 109 CHAPTER V: DESCRIPTIVE STATISTICS OF OBSERVED PRODUCTION RATES ............................................................................ 111 5.1 Excavation................................................................................... 112 5.2 Embankment................................................................................ 113 5.3 Lime-Treated Sub-grade ............................................................. 114 5.4 Aggregate Base Course ............................................................... 115 5.5 Hot Mix Asphalt Pavement ......................................................... 116 5.6 Slip-form Concrete Pavement ..................................................... 117 5.7 Conventional Form Concrete Pavement ..................................... 118 CHAPTER VI: DATA ANALYSIS AND HYPOTHESIS TESTING FOR EARTHWORK-RELATED WORK ITEMS............................................. 120 6.1 Test Difference in Mean Production Rates .......................................... 121 x 6.2 Analysis of Drivers of Production Rates.............................................. 122 6.2.1 Excavation................................................................................ 122 6.2.2 Embankment............................................................................. 129 6.2.3 Lime-Treated Sub-grade .......................................................... 139 6.2.4 Aggregate Base Course ............................................................ 149 6.3 Correlations Testing of Drivers............................................................ 166 6.4 Effects of Multiple Drivers on Production Rates ................................. 169 6.4.1 Embankment: Production Rates by Logarithmic Transformation of Work Area Quantity and Work Zone Congestion................................................................................ 169 6.5 Summary of Findings on Driver Analyses........................................... 171 CHAPTER VII: DATA ANALYSIS AND HYPOTHESIS TESTS FOR PAVEMENT-RELATED WORK ITEMS ................................................ 174 7.1 Test Difference in Mean Production Rates .......................................... 174 7.2 Analysis of Drivers of Production Rates.............................................. 175 7.2.1 Hot Mix Asphalt Pavement ...................................................... 175 7.2.2 Slip-form Concrete Pavement .................................................. 183 7.2.3 Analysis of Drivers of Production Rates for Conventional Form Concrete Pavement ......................................................... 192 7. 3 Correlations Testing of Drivers........................................................... 199 7.4 Effects on Multiple Drivers on Production Rates ................................ 201 7.4.1 Hot Mix Asphalt Pavement: Production Rates by Logarithmic Transformation of Work Area Quantity and Course Type ............................................................................. 201 7.5 Summary of Findings on Driver Analyses........................................... 203 CHAPTER VIII: CONCLUSIONS OF THIS RESEARCH............................... 206 8.1 Conclusions .......................................................................................... 206 8.2 Recommendations ................................................................................ 210 xi APPENDICES..................................................................................................... 211 Appendix A. Questionnaire for Selecting Work Items for the Study ................. 212 Appendix B. Results of the Survey for Selecting Work Items to be Tracked..... 216 Appendix C. Project-Level Data Collection Tool............................................... 217 Appendix D. Work Zone & Work Item -Level Data Collection Tool ................ 219 Appendix E. Work Item Sheets........................................................................... 224 Appendix F. Safety Protocol ............................................................................... 231 Appendix G. General Information of Investigated Projects................................ 233 Appendix H. Production Rates Data for this Study ............................................ 234 Appendix I-1. Excavation: Scatter Plots of Observed Production Rates vs. Candidate Drivers....................................................................................... 241 Appendix I-2. Excavation: Scatter Plots of Observed Production Rates (adjusted by crew size) vs. Candidate Drivers ........................................... 247 Appendix J. Results of Testing Assumptions of the Regression Analysis for Excavation: Production Rates vs. Work Area Quantity............................. 253 Appendix K. Embankment: Scatter Plots of Observed Production Rates vs. Candidate Drivers....................................................................................... 255 Appendix L-1. Results of Testing Assumptions of the Regression Analysis for Embankment: Production Rates and Work Area Quantity ........................ 261 Appendix L-2. Results of Testing Normality of Variables for Embankment: Production Rates by Work Zone Congestion............................................. 263 Appendix M. Lime-Treated Sub-grade: Scatter Plots of Observed Production Rates vs. Candidate Drivers ....................................................................... 264 Appendix N-1. Results of Testing Assumptions of the Regression Analysis for Lime-Treated Sub-grade: Production Rates and Work Area Quantity ...... 271 xii Appendix N-2. Results of Testing Assumptions of the Regression Analysis for Lime-Treated Sub-grade: Production Rates vs. Length of Work Area...... 273 Appendix O. Cement-Treated Base: Scatter Plots of Observed Production Rates vs. Candidate Drivers ....................................................................... 275 Appendix P. Flexible Base: Scatter Plots of Observed Production Rates vs. Candidate Drivers....................................................................................... 281 Appendix Q-1. Results of Testing Assumptions of the Regression Analysis for Cement-Treated Base: Production Rates vs. Work Area Quantity ............ 287 Appendix Q-2. Results of Testing Assumptions of the Regression Analysis for Cement-Treated Base: Production Rates vs. Lift-Length of Work Area ... 289 Appendix R-1. Results of Testing Assumptions of the Regression Analysis for Flexible Base: Production Rates vs. Work Area Quantity ......................... 291 Appendix R-2. Results of Testing Assumptions of the Regression Analysis for Flexible Base: Production Rates vs. Lift-Length of Work Area ................ 293 Appendix S. Results of Testing Assumptions of the Multiple Regression Analysis for Embankment:......................................................................... 295 Appendix T. Hot Mix Asphalt Pavement: Scatter Plots of Observed Production Rates vs. Candidate Drivers..................................................... 296 Appendix U-1. Results of Testing Assumptions of the Regression Analysis for Hot Mix Asphalt Pavement: Production Rates vs. Work Area Quantity ... 303 Appendix U-2. Results of Testing Normality of Variables for Hot Mix Asphalt Pavement: Production Rates by Course Types .......................................... 305 Appendix V. Slip-form Concrete Pavement: Scatter Plots of Observed Production Rates vs. Candidate Drivers..................................................... 306 Appendix W-1. Results of Testing Assumptions of the Regression Analysis for Slip-form Concrete Pavement Construction: Production Rates vs. Work Area Quantity ................................................................................... 313 Appendix W-2. Results of Testing Assumptions of the Regression Analysis for Slip-form Concrete Pavement Construction: Production Rates vs. Length of Work Area ................................................................................. 315 xiii Appendix X. Conventional Form Concrete Pavement: Scatter Plots of Observed Production Rates vs. Candidate Drivers .................................... 317 Appendix Y-1. Results of Testing Assumptions of the Regression Analysis for Conventional Form Concrete Pavement Construction: Production Rates vs. Work Area Quantity ............................................................................. 323 Appendix Y-2. Results of Testing Normality for Conventional Form Concrete Pavement Construction: Observed Production Rates by Configuration of Concrete Pavement..................................................................................... 325 Appendix Z. Results of Testing Assumptions of the Multiple Regression Analysis for Hot Mix Asphalt Pavement: .................................................. 326 GLOSSARY........................................................................................................ 327 BIBLIOGRAPHY ............................................................................................... 329 VITA ................................................................................................................... 335 xiv List of Tables Table 2.1 Definitions of Production Rates for the Targeted Work Items ............. 10 Table 2.2 Finalized Production Rates Database for CTDS ................................... 15 Table 2.3 Adjustment Values for CTDS job Factors (Hancher et al. 1992) ......... 16 Table 2.4 CTDS Base Production Rates and Sensitivity Factors (Hancher et al. 1992).................................................................................................... 16 Table 2.5 Definition of Congestion and Accessibility (Ovararin and Popescu 2001).................................................................................................... 24 Table 3.1 Selected Districts vs. Area .................................................................... 43 Table 3.2 Rational for Production Rate Computation........................................... 45 Table 3.3 General Information of the Sources of Historical Records ................... 47 Table 3.4 Sample Sizes Required to Test the Hypothesis that the Population Multiple Correlation Equals Zero with a Power of 0.80 and of 0.05 (adopted from Green 1991) ................................................................. 52 Table 4.1 Candidate Drivers vs. Seven Targeted Work Items .............................. 86 Table 4.2 Work Item Level Candidate Drivers and Data Attributes..................... 87 Table 4.3 Scope of Excavation for Data Collection.............................................. 98 Table 4.4 Scope of Embankment for Data Collection ........................................ 100 Table 4.5 Scope of Lime-Treated Sub-grade for Data Collection ...................... 102 Table 4.6 Scope of Aggregate Base for Data Collection .................................... 104 Table 4.7 Scope of Hot Mix Asphalt Pavement for Data Collection .................. 105 Table 4.8 Scope of Slip-form Concrete Pavement for Data Collection .............. 106 Table 4.9 Scope of Conventional Form Concrete Pavement for Data Collection ........................................................................................................... 107 Table 4.10 Dates of Collecting Data, Number of Investigated Projects and Number of Observed Data by Visited Districts ................................ 108 Table 4.11 Number of Projects by Prime Contractor I.D.................................... 109 xv Table 4.12 Number of Projects by Contract Amount.......................................... 109 Table 4.13 Sources for Data and Observed Quantity for Seven Work Items ..... 110 Table 5.1 Range Data for Seven Work Items...................................................... 111 Table 6.1 Average Production Rates of the CTDS and Mean Observed Production Rates .................................................................................................. 122 Table 6.2 Logarithmic Model for Excavation: Production Rates (CY/Crew Day) by Work Area Quantity (CY)............................................................ 126 Table 6.3 Daily Production Rates (CY/Day) of Different Loading Machine (Adopted from Heavy Construction Cost Data of RS Means 2002). 128 Table 6.4 Logarithmic Model for Embankment: Production Rates (CY/Crew Day) by Work Area Quantity (CY)............................................................ 135 Table 6.5 Results of Group Variances Test and ANOVA Test for Embankment: Work Zone Accessibility................................................................... 137 Table 6.6 Numbers of Observed Data Points and Mean Production Rate for Embankment: Work Zone Accessibility ........................................... 137 Table 6.7 Results of Group Variances Test and ANOVA Test for Embankment: Work Zone Congestion ..................................................................... 138 Table 6.8 Numbers of Observed Data Points and Mean Production Rate for Embankment: Work Zone Congestion.............................................. 139 Table 6.9 Logarithmic Model for Lime-Treated Sub-grade: Observed Production Rates (SY/Crew Day) by Work Area Quantity (SY) ........................ 144 Table 6.10 Logarithmic Model for Lime-Treated Sub-grade: Production Rates (SY/Crew Day) by Length of Work Area (LF)................................. 147 Table 6.11 Results of Group Variances Test and t-test for Lime-Treated Subgrade: Project Location (Rural and Urban) ....................................... 149 Table 6.12 Linear Model for Cement-Treated Base: Production Rates (LiftSY/Crew Day) by Work Area Quantity (Lift-SY) ............................ 156 xvi Table 6.13 Linear Model for Cement-Treated Base: Production Rates (LiftSY/Crew Day) by Lift-Length of Work Area (LF)........................... 159 Table 6.14 Logarithmic Model for Flexible Base: Production Rates (Lift-SY/Crew Day) by Work Area Quantity (Lift-SY) ............................................ 162 Table 6.15 Logarithmic Model for Flexible Base: Production Rates (Lift-SY/Crew Day) by Lift-Length of Work Area (LF)........................................... 165 Table 6.16 Correlations Test for Work Area Quantity and Work Zone Congestion of Embankment Construction............................................................ 167 Table 6.17 Correlations Test for Work Area Quantity and Length of Work Area of Lime-Treated Sub-grade Construction.............................................. 168 Table 6.18 Correlations Test for Work Area Quantity and Lift-Length of Work Area of Cement-Treated Base Construction ..................................... 168 Table 6.19 Correlations Test for Work Area Quantity and Lift-Length of Work Area of Flexible Base Construction .................................................. 169 Table 6.20 Multiple Regression Model for Embankment................................... 170 Table 6.21 Summary of Results of Driver Analyses........................................... 172 Table 6.22 Summary of Identified Production Rate Drivers .............................. 173 Table 7.1 Average CTDS Production Rates and Mean Observed Production Rates for HMAP and Slip-form Concrete Pavement .................................. 175 Table 7.2 Logarithmic Model for Hot Mix Asphalt Pavement: Production Rates (Ton/Crew Day) by Work Area Quantity (Ton) ............................... 180 Table 7.3 Results of Group Variances Test and ANOVA Test for Hot Mix Asphalt Pavement: Course Type ....................................................... 182 Table 7.4 Hot Mix Asphalt Pavement: Numbers of Data Points and Mean Production Rate ................................................................................. 182 Table 7.5 Logarithmic Model for Slip-form Concrete Pavement: Production Rates (SY/Crew Day) by Work Area Quantity (SY) .................................. 187 xvii Table 7.6 Logarithmic Model for Slip-form Concrete Pavement: Production Rates (SY/Crew Day) by Work Area Quantity (SY) .................................. 191 Table 7.7 Linear Model for Conventional Form Concrete Pavement: Production Rates (SY/Crew Day) by Work Area Quantity (SY) ........................ 196 Table 7.8 Results of Group Variances Test and T Test for Conventional Form Concrete Pavement: Configuration ................................................... 198 Table 7.9 Numbers of Data Points and Mean Production Rate for Conventional Form Concrete Pavement: Configuration ......................................... 199 Table 7.10 Correlations Test for Work Area Quantity and Course Type of Hot Mix Asphalt Pavement Construction ................................................ 199 Table 7.11 Correlations Test for Work Area Quantity and Length of Work Area of Slip-form Concrete Pavement Construction...................................... 200 Table 7.12 Correlations Test for Work Area Quantity and Configuration of Conventional Form Concrete Pavement Construction...................... 200 Table 7.13 Multiple Regression Model for Hot Mix Asphalt Pavement ............ 202 Table 7.14 Summary of Results of Driver Analyses........................................... 204 Table 7.15 Summary of Identified Production Rate Drivers .............................. 205 Table 8.1 Work Items vs. Significant Drivers of this Research and the CTDS .. 208 xviii List of Figures Figure 2.1 Sources of Production Rates Used for Contract Time Determination adopted from (Hancher et al. 1992)..................................................... 12 Figure 2.2 Factor Model (adopted from Thomas and Yiakoumis 1987) .............. 18 Figure 2.3 Effective return from working 50 or 60 hours a week for various numbers of weeks (Source: Business Roundtable Cost Effectiveness Study Report C-3, November 1980.) .................................................. 21 Figure 3.1 Research Methodology ........................................................................ 37 Figure 3.2 Data Collection Process ....................................................................... 40 Figure 3.3 Annotated Sketch of the Box Plot........................................................ 46 Figure 3.4 Flow Chart of Driver Analysis............................................................. 49 Figure 4.1 Influence Diagram of the Production Rate (CY/Crew Day) for Excavation........................................................................................... 60 Figure 4.2 Influence Diagram of the Production Rate (CY/Crew Day) for Embankment........................................................................................ 64 Figure 4.3 Influence Diagram of the Production Rate (SY/Crew Day) for LimeTreated Sub-grade ............................................................................... 68 Figure 4.4 Influence Diagram of the Production Rate (Lift-SY/Crew Day) for Aggregate Base Course ....................................................................... 71 Figure 4.5 Influence Diagram of the Production Rate (Ton/Crew Day) for Hot Mix Asphalt Pavement ........................................................................ 76 Figure 4.6 Influence Diagram of the Production Rate (SY/Crew Day) for Slipform Concrete Pavement ..................................................................... 81 Figure 4.7 Influence Diagram of the Production Rate (SY/Crew Day) for Conventional form Concrete Pavement .............................................. 85 Figure 5.1 Comparison of Excavation Production Rates from Different Sources ........................................................................................................... 112 xix Figure 5.2 Comparison of Embankment Production Rates from Different Sources ........................................................................................................... 113 Figure 5.3 Comparison of Lime-Treated Sub-grade Production Rates from Different Sources............................................................................... 114 Figure 5.4 Comparison of Aggregate Base Course Production Rates from Different Sources............................................................................... 116 Figure 5.5 Comparison of Hot Mix Asphalt Pavement Production Rates from Different Sources............................................................................... 117 Figure 5.6 Comparison of Slip-form Concrete Pavement Production Rates from Different Sources............................................................................... 118 Figure 5.7 Comparison of Conventional Form Concrete Pavement Production Rates from Different Sources ............................................................ 119 Figure 6.1 Scatter Plot for Excavation: Observed Production Rates (CY/Crew Day) and Work Area Quantity (CY) ................................................. 123 Figure 6.2 Excavation: Box Plot of Observed Production Rates (CY/Crew Day) ........................................................................................................... 124 Figure 6.3 Excavation: Box Plot of Logarithmic Transformation of Work Area Quantity (CY).................................................................................... 125 Figure 6.4 Scatter Plot and Fitted Logarithmic Model for Excavation: Observed Production Rates (CY/Crew Day) vs. Work Area Quantity (CY) .... 125 Figure 6.5 Scatter Plot and Fitted Linear Model for Excavation: Observed Work Area Quantity (CY) vs. Size of an Employed Loader Fleet.............. 129 Figure 6.6 Scatter Plot for Embankment: Observed Production Rate (CY/Crew Day) vs. Work Area Quantity (CY) .................................................. 131 Figure 6.7 Scatter Plot for Embankment: Observed Production Rate (CY/Crew Day) vs. Work Zone Accessibility .................................................... 131 Figure 6.8 Scatter Plot for Embankment: Observed Production Rate (CY/Crew Day) vs. Work Zone Congestion....................................................... 132 xx Figure 6.9 Embankment: Box Plot of Observed Production Rates (CY/Crew Day) ........................................................................................................... 133 Figure 6.10 Embankment: Box Plots of Log (Work Area Quantity (CY))......... 133 Figure 6.11 Scatter Plot and Fitted Logarithmic Model for Embankment: Observed Production Rates (CY/Crew Day) vs. Work Area Quantity (CY)................................................................................................... 134 Figure 6.12 Scatter Plot for Lime-Treated Sub-grade: Production Rate (SY/Crew Day) vs. Work Area Quantity (SY)................................................... 140 Figure 6.13 Scatter Plot for Lime-Treated Sub-grade: Production Rate (SY/Crew Day) vs. Length of Work Area (LF) ................................................. 141 Figure 6.14 Scatter Plot for Lime-Treated Sub-grade: Production Rate (SY/Crew Day) vs. Location .............................................................................. 141 Figure 6.15 Lime-Treated Sub-grade: Box Plot of Observed Production Rates (SY/Crew Day).................................................................................. 142 Figure 6.16 Lime-Treated Sub-grade: Box Plot of Log Transformation of Work Area Quantity (SY) ........................................................................... 143 Figure 6.17 Scatter Plot and Logarithmic Model for Lime-Treated Sub-grade: Observed Production Rates (SY/Crew Day) vs. Work Area Quantity (SY) ................................................................................................... 143 Figure 6.18 Lime-Treated Sub-grade: Box Plot of Log Transformation of Length of Work Area (LF) ............................................................................ 146 Figure 6.19 Scatter Plot and Logarithmic Model for Lime-Treated Sub-grade: Observed Production Rates (SY/Crew Day) vs. Length of Work Area Quantity (LF)..................................................................................... 147 Figure 6.20 Aggregate Base: Scatter Plot of Observed Production Rates (LiftSY/Crew Day) vs. Types of Aggregate Base Operations ................. 151 Figure 6.21 Cement-Treated Base: Scatter Plot of Observed Production Rates (Lift-SY/Crew Day) vs. Work Area Quantity (Lift-SY)................... 151 xxi Figure 6.22 Cement-Treated Base: Scatter Plot of Observed Production Rates (Lift-SY/Crew Day) vs. Lift-Length of Work Area (LF).................. 152 Figure 6.23 Flexible Base: Scatter Plot of Observed Production Rates (LiftSY/Crew Day) vs. Work Area Quantity (Lift-SY) ........................... 152 Figure 6.24 Flexible Base: Scatter Plot of Observed Production Rates (LiftSY/Crew Day) vs. Lift-Length of Work Area (LF) .......................... 153 Figure 6.25 Cement-Treated Base: Box Plot of Observed Production Rates (LiftSY/Crew Day) ................................................................................... 154 Figure 6.26 Cement-Treated Base: Box Plot of Log Transformation of Work Area Quantity (Lift-SY)............................................................................. 154 Figure 6.27 Scatter Plot and Linear Model for Cement-Treated Base: Production Rates (Lift-SY/Crew Day) vs. Work Area Quantity (Lift-SY) ......... 155 Figure 6.28 Cement-Treated Base: Box Plot of Lift-Length of Work Area (LF)157 Figure 6.29 Scatter Plot and Linear Model for Cement-Treated Base: Observed Production Rates (Lift-SY/Crew Day) vs. Lift-Length of Work Area (LF).................................................................................................... 158 Figure 6.30 Flexible Base: Box Plot of Observed Production Rates (Lift-SY/Crew Day) ................................................................................................... 160 Figure 6.31 Flexible Base: Box Plot of Logarithmic Transformation of Work Area Quantity (Lift-SY)............................................................................. 161 Figure 6.32 Scatter Plot and Linear Model for Flexible Base: Observed Production Rates (Lift-SY/Crew Day) vs. Work Area Quantity (LiftSY) .................................................................................................... 161 Figure 6.33 Flexible Base: Box Plot of Logarithmic Transformation of Work Area Quantity (Lift-SY)............................................................................. 164 Figure 6.34 Scatter Plot and Logarithmic Model for Flexible Base: Observed Production Rates (Lift-SY/Crew Day) vs. Lift-Length of Work Area (LF).................................................................................................... 165 xxii Figure 7.1 Hot Mix Asphalt Pavement: Scatter Plot of Observed Production Rates (Ton/Crew Day) vs. Work Area Quantity (Ton)............................... 176 Figure 7.2 Hot Mix Asphalt Pavement: Scatter Plot of Observed Production Rates (Ton/Crew Day) vs. Course Type ..................................................... 177 Figure 7.3 Hot Mix Asphalt Pavement: Box Plot of Observed Production Rates (Ton/Crew Day) ................................................................................ 178 Figure 7.4 Hot Mix Asphalt Pavement: Box Plot of Logarithmic Transformation of Work Area Quantity (Ton)............................................................ 178 Figure 7.5 Scatter Plot and Logarithmic Model for Hot Mix Asphalt Pavement: Observed Production Rates (Ton/Crew Day) vs. Work Area Quantity (Ton).................................................................................................. 179 Figure 7.6 Slip-form Concrete Pavement: Scatter Plot of Observed Production Rates (SY/Crew Day) vs. Work Area Quantity (SY)........................ 184 Figure 7.7 Slip-form Concrete Pavement: Scatter Plot of Observed Production Rates (SY/Crew Day) vs. Length of Work Area (LF) ...................... 184 Figure 7.8 Slip-form Concrete Pavement: Box Plot of Observed Production Rates (SY/Crew Day).................................................................................. 185 Figure 7.9 Slip-form Concrete Pavement: Box Plot of Logarithmic Transformation of Work Area Quantity (SY) ................................... 186 Figure 7.10 Scatter Plots and Logarithmic Model for Slip-form Concrete Pavement: Observed Production Rates (SY/Crew Day) vs. Work Area Quantity (SY) .................................................................................... 186 Figure 7.11 Slip-form Concrete Pavement: Box Plot of Logarithmic Transformation of Length of Work (LF)........................................... 189 Figure 7.12 Scatter Plot and Logarithmic Model for Slip-form Concrete Pavement: Observed Production Rates (SY/Crew Day) vs. Work Area Quantity (SY) .................................................................................... 190 xxiii Figure 7.13 Conventional Form Concrete Pavement: Scatter Plot of Observed Production Rates (SY/Crew Day) vs. Work Area Quantity (SY) ..... 193 Figure 7.14 Conventional Form Concrete Pavement: Scatter Plot of Observed Production Rates (SY/Crew Day) vs. Configuration ........................ 193 Figure 7.15 Conventional Form Concrete Pavement: Box Plot of Observed Production Rates (SY/Crew Day) ..................................................... 194 Figure 7.16 Conventional Form Concrete Pavement: Box Plot of Logarithmic Transformation of Work Area Quantity (SY) ................................... 195 Figure 7.17 Pavement: Observed Production Rates (SY/Crew Day) vs. Logarithmic Transformation of Work Area Quantity (SY) .............. 195 xxiv CHAPTER I: INTRODUCTION Construction projects have increased in both complexity and size in recent decades. This is in part, due to the pursuit of better construction efficiency and Because of safety, lower construction cost, and higher quality (Gidado 1996). this change, the more simplistic approaches of determining Contract Time using manual calculations and bar charts are not feasible. method (CPM) is usually employed. Instead, the critical path Consequently, today, an improved information system needs to be developed to support the application of CPM on Contract Time estimation. 1.1 RESEARCH BACKGROUND AND MOTIVATION The critical path method (CPM) is a generally accepted method to estimate project duration for construction projects. Four steps are necessary to develop a CPM scheduling network for a construction project. First, a work breakdown structure is established to define the activities in the project. relationships between activities are then established. activities are estimated. Second, the Third, the durations of the Finally, the activities are compiled to develop a CPM Because project scheduling network and the project duration is computed. duration is calculated from compiling all activities into a scheduling network, the accuracy of estimating project duration is determined by the accuracy of 1 estimating each activity-duration. The only way to obtain accurate activity duration is to use realistic Production Rates. The duration of a construction project is usually determined by the clients at the design stage and is then documented in the bid documents. Contractors are usually under an obligation to evaluate the feasibility of the project duration before a contract has been awarded. In reality, however, time pressures typically do not allow contractors to perform this analysis. Further complicating the matter, clients frequently use inaccurate Production Rates to estimate construction time. Therefore, many projects are developed using unrealistic Contract Time duration. Projects with overestimated project durations can cause unnecessary inconvenience to the traveling public and may reduce the project profits of the general contractors because of the increased project overhead. In addition these projects may allow contractors with lower productivity to bid, discourage the utilization of advanced techniques, or allow contractors to bid for additional work that otherwise they would not have been able to handle (NCHRP 1981; Herbsman and Ellis 1995). Underestimation of project duration also leads to many problems (NCHRP 1981; Herbsman and Ellis 1995). Bid prices may be higher due to the increased cost of accelerated construction productivity. Construction quality may be 2 reduced, and litigation may increase due to liquidated damages caused by project delay. Some qualified contractors may not even bid due to these concerns. Using inaccurate Production Rates to estimate construction time has been recognized as a major source of bias in Contract Time estimation. To prevent inaccurate Contract Time estimation, a process for obtaining reliable Production Rate data is needed. Recognizing this need, a reliable Production Rates database for construction time estimation will help the construction industry solve many problems associated with construction time. Clients may use the database to determine a General reasonable Contract Time and also to apply it to site management. contractors can use it to verify the Contract Time imposed on the bid documents and to monitor and manage the job. The Center for Transportation Research (CTR) at the University of Texas, Austin, received funding from the Texas Department of Transportation (TxDOT) to develop a reliable Production Rate database for highway construction time estimation. This research project was undertaken by a research team comprised of research supervisor Dr. James T. O'Connor, who is a professor at the University of Texas Austin, and three graduate research assistants. The project was monitored by the Project Monitoring Committee (PMC) that included seven professional engineers at TxDOT. 3 1.2 RESEARCH OBJECTIVES Many factors, such as weather conditions (Kahkonen 1991) and resource utilization (Proverbs et al. 1998) can cause large variances to Production Rates in highway construction. These factors can either speed up or slow down work production. The main purpose of this study was to investigate and examine field construction Production Rates in two Work Areas, namely Earthwork and Pavement construction. Many studies have explored the relationship between The intention of this research was to identify productivity and various factors. the relationships between Production Rates and their drivers, with an emphasis on examining those drivers which have a great impact during the construction process and can be determined at the design stage. Further analysis will be carried out to determine the exact relationship and intensity of these relationships on the Production Rates. The objectives of the research were to 1.) document Production Rate information for twenty-six major Work Items of highway construction projects from TxDOT's ongoing projects, 2.) identify the factors significantly influencing the Production Rate of each targeted Work Item and 3.) explore the relationships between daily Production Rates and identified Significant Drivers. A portion of the data from this TxDOT research project was used for the data analysis of this study. 4 1.3 RESEARCH SCOPE LIMITATIONS Productivity can be viewed by management from two perspectives. perspective is for cost management purposes. The first This type of productivity is usually used to measure the efficiency of labor-intensive activities. The main purposes of such productivity measurement are to discover the factors which lead to low productivity, and to quantify their impacts on productivity for further improvements (AbouRizk et al. 2001; Christian and Hachey 1995). The second perspective on productivity is for time management. of productivity is usually called Production Rate. This type The difference between the This study will focus two types of productivity is discussed further in Chapter 2. on the second type of productivity measurement and only concentrate on Production Rates associated with the major Work Items in highway Earthwork and Pavement construction. 1.4 STRUCTURE OF DISSERTATION This dissertation consists of eight chapters and twenty-six appendices which contain supporting information and results of the data collection and analysis. Chapter 2 elaborates on construction productivity with a comprehensive literature review. It begins with defining productivity and follows by focusing on quantification of the effects of factors on construction productivity, Earthwork and Pavement productivity, and the various methods used in analyzing 5 construction productivity. Chapter 3 discusses the research methodology Chapter 3 starts with an overview employed to achieve the research objectives. of the research methodology, followed by a brief description of the preparation and execution of data collection. The chapter ends with proposed statistical methods used to analyze Production Rates and to identify the drivers of Production Rates. Chapter 4 discusses the details of preparation and execution of data collection. At the beginning of this chapter, the research hypotheses of the study are introduced. Next follows a description of developing data collection tool, planning the data collection process and collecting the data. summarizes the results of data collection. statistics of observed Production Rates. This chapter also Chapter 5 presents the descriptive Chapter 6 presents the key findings of the hypotheses tests and the driver analyses for Earthwork-related Work Items. The relationships between Production Rates and their drivers are discussed in this Chapter. Items. Similar to Chapter 6, Chapter 7 focuses on Pavement-related Work Chapter 8 concludes this research study and provides suggestions for future research. 6 CHAPTER II: LITERATURE REVIEW In order to estimate Contract Time of construction projects in a more consistent fashion, many transportation agencies have attempted to establish a standard procedure to determine Contract Time. Hancher et al. (1992) and Werkmeister et al. (2000) suggested further study on exploring realistic highway construction Production Rates. In addition the Transportation Research Board conducted studies in 1981 and 1995 to investigate the system used to determine Contract Time for construction projects in most state transportation agencies (NCHRP 1981; Herbsman et al. 1995). They indicated that "realistic Production Rates are the key in determining reasonable Contract Times" (Herbsman et al. 1995). Productivity study has been an important and continuing area of interest in the construction industry. In this section, the definitions of construction productivity and Production Rates used in this study are presented. This chapter also reviews the methods for measuring productivity with different sources of Production Rates and the methods of quantifying the effects of factors on productivity. Furthermore, preceding productivity studies on Earthwork and Pavement construction, and methods of productivity analysis are presented. 7 2.1 DEFINITIONS OF PRODUCTIVITY Productivity has been defined in many ways for different applications. Productivity can be defined using an economic model, project-specific model, or an activity model (see Equations 1, 2 and 3 below). measured for different purposes (Thomas et al. 1990). Economic Model: Productivity = Output $/Input $ Each of these models is (Equation 1) Project-Specific Model: Productivity = Square Feet/Dollars or Physical Output/Dollars (Equation 2) Activity Model: Productivity = Output/Labor Cost or Output/Work Hours (Days) (Equation 3) In order to evaluate the impact of equipment technology on productivity, Goodrum and Hass (2002) modified the project-specific model to the partial factor model, as shown in Equation 4. They removed the material cost from the dollars in the project-specific model. Partial Factor Model: Productivity = Physical Output/(Labor Cost + Fixed Capital Cost) (Equation 4) 8 The most popular definition of productivity is the unit rate (Borcherding et al. 1986) shown in Equation 5. The output is taken as the completed quantity, and This definition is usually the input is the engaged manpower to produce output. used for cost management to identify the variability of required manpower for completing a unit of output. Unit Rate: Productivity = Input / Output (Equation 5) In this study, the activity model will be for Production Rates measurement. The duration of an activity is usually determined by multiplying the estimated Production Rates by the work quantity for an activity. Therefore, Production Rates used for construction time estimation will be measured as the completed quantity in a Work Area divided by the working days that a crew needs to complete an activity. The Work Area Quantity for each item is measured in the unit that is available for designers in order to facilitate the calculation of the activity duration. Table 2.1 indicates the units of measurement for the seven targeted Work Items. 9 Table 2.1 Definitions of Production Rates for the Targeted Work Items Work Item Excavation Embankment Lime-Treated Sub-grade Aggregate Base Course Hot Mix Asphalt Pavement Slip-form Concrete Pavement Conventional Form Concrete Pavement Unit of Measurement CY/Crew Day CY/Crew Day SY/Crew Day Lift-SY/Crew Day TON/Crew Day SY/Crew Day SY/Crew Day Remark CY: Bank Quantity CY: Compacted Quantity SY: Completed Area Lift-SY: Total Area of Completed Working Lifts TON: Placed Weight SY: Completed Area SY: Completed Area 2.2 APPROACHES FOR MEASURING PRODUCTIVITY Various approaches such as work sampling (Liou and Borcherding 1986; Thomas 1991), the craftsman questionnaire (Chang and Borcherding 1986), and the foreman delay survey (Tucker et al. 1982) have been employed to investigate the causes that lead to inefficiency in construction tasks. Work sampling has been utilized to evaluate workers' time utilization. Liou and Borcherding (1986) collected data from eleven nuclear power projects and four fossil fuel projects to study whether the unit rate productivity could be predicted using workers' time utilization data. This study concluded that there was a high correlation between them. 10 Thomas (1991) conducted a similar study to test whether a high direct work rate would lead to better labor productivity. It was reported that the direct work Winch and Carr (2001) rate was not directly correlated with labor productivity. also used work sampling to investigate what caused the difference in concrete productivity between France and the UK. Chang and Borcherding (1986) used craftsman questionnaire sampling that combined a craftsman questionnaire and work sampling as a new approach to identify the sources of delays. This method provided some useful solutions to problems impacting construction productivity. large nuclear power plant site. Tucker et al. (1982) developed a new approach, Foremen Delay Survey (FDS), to identify the sources of delay and to quantify time or dollar losses. This quantification method easily ranks the sources that caused delays according to their time and cost impacts. Subsequent application of the FDS on a job site This approach was tested at a can be used to evaluate the cost effectiveness of various solutions. 2.3 SOURCES OF PRODUCTION RATES Developing scheduling networks is a complicated and time consuming task if there is not a reliable Production Rates resource. This is true even for an experienced project engineer (Kahkonen 1991). According to Hancher et al. (1992), in their review on Rowing's (1992) study, several resources are in current use. The participants of thirty-six Departments of Transportation (DOT) 11 responded to a survey on resources used to estimate Production Rates for Contract Time determination. The results of the survey are shown in Figure 2.1. Forty- four percent of the respondents relied on personal experience to predict Production Rates. Thirty percent of the respondents used standard Production Rates and twenty-two percent used Production Rates from completed projects or historical records. Sources of Production Rates Used for Contract Time Determinatio n 50% 45% 40% 35% 30% 25% 20% 15% 10% 5% 0% Personal Experience Standard Production Rates Past Projects and Historical Records Figure 2.1 Sources of Production Rates Used for Contract Time Determination adopted from (Hancher et al. 1992) The Production Rates obtained from personal experience and historical records were fragmented and unevaluated. Vital information such as the factors This that significantly influence the variability of Production Rates was lacking. resulted in that an average Production Rate was used as the representative 12 Production Rate for an entire project. The neglect of the effects of various factors caused Production Rates estimation to be inaccurate and biased. Other sources of construction productivity are biased as well. The Construction Time Determination System (CTDS), a system developed by the Texas Department of Transportation (TxDOT) to guide designers in construction time estimation at the design stage, is inadequate as is the RS Means data. Although the RS Means Heavy Construction Cost Data publishes hourly and daily productivity and adjusts it regularly, those productivity data are intended primarily for cost management purposes. 2.3.1 RS Means The productivity data in the RS Means Heavy Construction Cost Data is measured in both daily and hourly output formats. It is based on an eight hour The rates have working day during daylight hours and in moderate temperature. to be adjusted if hours in a working day total more than eight or when the environment is considered adverse. Unfortunately, no further information on the methods to adjust the productivity data is provided in RS Means. However, the productivity data in RS Means is primarily intended for cost management, therefore, its use in construction time estimation may be limited. The differences between productivity for cost management and for construction time estimation is best demonstrated in the following example. For instance, an Operation for a 900 CY Excavation is performed by an excavator and 13 three trucks within a day. A total of 7 hours will be spent on this Operation and within the 7 hours, 1 hour may be unproductive because the excavator breaks down. For cost management purposes, the productivity should be measured by the completed quantity of 900 CY plus the 6 hours of normal work, and the cost of the idle time should be counted as a project overhead. So, productivity is 150 CY/Hour and 1,200 CY/Day for an 8-hour working day for cost management purposes. But for the purpose of construction time estimation, the daily productivity is 900 CY/Day. 2.3.2 Contract Time Determination System (CTDS) The Contract Time Determination System (CTDS) is "a conceptual estimating system for predicting Contract Time for highway construction projects and is not to be used for the detailed planning of actual construction activities for a project". (Hancher et al. 1992) This system is a product of research conducted by the Texas Transportation Institute and the Texas Department of Transportation (TxDOT) in 1992. A portion of this research was to explore Production Rates in highway A survey was employed to investigate the daily Production Rate construction. from twenty-five TxDOT districts. In the survey questionnaire, forty two Work Items were defined and the low, average and high Production Rates for each Work Item were asked to be estimated. From the forty three responses, the mean value of the low, average and high Production Rates for each item was computed. In addition, a request form was sent to all fifty state transportation agencies to 14 request Production Rates data. Rates data for the study. Twenty-four states provided their Production Finally, a Production Rates database was developed Table 2.2 lists the finalized Production Rate of the using the two sets of results. CTDS study for the Work Items associated with the targeted items in this study. (The Work Items in this study are listed in Table 2.1.) Table 2.2 Finalized Production Rates Database for CTDS MAJOR WORK ITEMS Earth Excavation Embankment Lime-Treated Sub-grade Flexible Base Course Cement Treated Base Hot Mix Asphalt Base Hot Mix Asphalt Surface Concrete Paving UNIT CY CY SY SY SY Ton Ton SY LOW 1,200 1,200 2,000 1,500 1,500 500 500 1,000 AVERAGE 3,400 3,500 4,000 3,000 3,000 1,200 1,200 3,000 HIGH 7,000 7,000 6,000 4,500 4,500 4,500 4,500 5,000 Five factors, namely, location, traffic conditions, complexity, soil conditions and quantity of work were analyzed and their effects on Production Rates were investigated via a survey so that the Production Rates could be adjusted to fit job conditions in the CTDS study. Table 2.3 displays the adjustment values for job factors for the eight related Work Items. Table 2.4 demonstrates the daily production base rates and the sensitivity factors determined from the surveys for the eight related Work Items in the CTDS study. 15 It was found that the eight related Work Items were very sensitive to quantity of work to be done according to the database in the CTDS. Earth Excavation, Embankment, Lime stabilization, and Cement-treated base material were all found to be influenced by soil conditions. Flexible base material, Hot mix asphalt base and Concrete paving were affected by location and Hot mix asphalt surface was affected by traffic. Table 2.3 Adjustment Values for CTDS job Factors (Hancher et al. 1992) Factors Location Traffic Condition Complexity Soil Conditions Quantity of Work Adjustment for Noted Conditions Rural = 1.0 Light = 1.0 Low = 1.0 Good = 1.00 Large = 1.00 Small City = 0.85 Moderate = 0.88 Medium = 0.85 Fair = 0.85 Medium = 0.88 Big City = 0.75 High = 0.70 High = 0.70 Poor = 0.65 Small = 0.75 Table 2.4 CTDS Base Production Rates and Sensitivity Factors (Hancher et al. 1992) DAILY PRODUCTION Sensitivity Factors BASE RATE Earth Excavation CY 4,200 l t c S Q Embankment CY 4,200 l t c S Q Lime Stablization SY 4,500 l t c S Q Flexible Base Material SY 3,400 L t c s Q Cement Treated Base Material SY 3,400 l t c S Q Hot Mix Asphalt Base Ton 1,400 L t c s Q Hot Mix Asphalt Surface Ton 1,400 l T c s Q Concrete Paving SY 3,400 L t c s Q L: Location; T: Traffic; C: Complexity; S: Soil Condition; Q: Quantity of Work MAJOR WORK ITEMS UNIT *Sensitivity Factors with capital letters indicate significant factors for individual work item 16 2.3.3 Historical Records For this study, some highway construction Production Rates data were collected from historical data recorded by contractors and were compared with observed Production Rates data. Most collected historical data were only There available for Excavation, Embankment and Hot mix asphalt pavement. was not sufficient information in these historical records to identify the factors that cause variability in daily Production Rates. Even for other Work Items such as Lime-treated sub-grade, Aggregate base course, Slip-form concrete pavement, and Conventional form concrete pavement, rates for some sub-activities such as remixing Lime-treated sub-grade, processing Flexible base, installing rebar for Concrete pavement, were not documented in the historical records. 2.4 GENERAL FACTORS AFFECTING PRODUCTIVITY Many productivity studies have identified productivity factors and measured their effects on productivity. Most of these were interested in the identification and quantification of factors that caused losses of construction productivity. Frequently cited factors from these studies include weather, scheduled overtime, disruption, congestion, and region (Halligan 1994; Koehn 2001). This section will review published studies associated with the identification and quantification of productivity factors related to this study. 17 Thomas and Yiakoumis (1987) employed the factor model to present relationships between labor productivity and productivity factors. The factor model displays the effects of learning curve and other factors on labor productivity, as shown in Figure 2.2. In the factor model, the ideal productivity curve presents a correlation between the cumulative man-hour per unit of work and the cumulative unit of work in an ideal condition of no disruption. The ideal productivity curve is varied with different crews. Their study indicated that losses in productivity are caused by numerous factors such as environmental factors, site factors, management factors, and design factors. Figure 2.2 Factor Model (adopted from Thomas and Yiakoumis 1987) 2.4.1 Weather Weather conditions at the construction site have a large impact on highway construction. Almost half of construction Operations are sensitive to weather 18 conditions (Oglesby et al. 1989). Precipitation, extremes of temperature, and humidity cause productivity loss (Borcherding 1991; Halligan 1994) and may even cause activities to be delayed. Hot temperature may increase the frequency of travel time for the workers in order to avoid heat and as such, productive time may be reduced as a result (Borcherding 1991). Cold temperature may increase the idle time of workers standing beside heat sources to warm themselves up (Borcherding 1991). Weather also has a huge impact on some work Operations, such as Lime-treated sub-grade, Concrete placement and Hot mix asphalt, as these Operations cannot be carried out in extreme weather (TxDOT 1993). Several studies have been conducted to quantify the effects of adverse weather on labor productivity. Grimm and Wagner (1974) conducted a study to measure the effects of temperature and humidity on masonry productivity. In their study, other factors influencing the loss of productivity were controlled to be constants in order to find the effects of temperature and humidity on labor productivity. It was reported that masonry productivity decreased with increasing deviation from 75 degrees of Fahrenheit or humidity of 60%. An experimental study (NECA 1974) conducted by the National Electrical Contractors Association measured the labor productivity of an electrician installing duplex receptacles in an environmental chamber where temperature and humidity was controlled. It was found that productivity decreased when the temperature was above 80F and below 40F, or when the relative humidity was 19 above 80%. Another study examined losses of labor productivity on steel It found that labor productivity was impacted erection due to cold temperature. by 32% due to cold temperature (Thomas et al. 1999). 2.4.2 Scheduled Overtime Scheduled overtime refers to "a planned decision by project management to accelerate the progress of the work by scheduling more than 40 work hours per week for an extended period of time for much of the craft work force" (Thomas and Raynar 1999). Scheduled overtime causes fatigue among workers and reduces motivation and indirectly causes labor productivity to deteriorate. Many studies have attempted to quantify the effects of such overtime on labor productivity. The 1980 Business Roundtable republished the findings of weekly productive returns from working 50 or 60 hours a week for various numbers of weeks. In the late 1960s, Weldon McGlaun reported these findings to members It was found that productivity during of the National Constructors Association. the first week of scheduled overtime fell dramatically and that productivity continued to go down week by week. After working for 50 hours per week continuously for seven weeks, the weekly output was similar to that when the workers actually worked 40 hours per week. For a 60 work-hour week, by the ninth week of scheduled overtime, the weekly output was similar to the output of working for only 40 hours a week. This is clearly shown in Figure 2.3. 20 However, conclusions from a study conducted by the Construction Industry Institute (1988) were inconsistent with previous findings. This study concluded that "productivity does not necessarily decrease with an overtime schedule" based on monitoring 25 crews on seven projects (three insulation crews, seven pipe crews, eleven electrical crews, one formwork crew, one rebar crew, and two concrete crews). Thomas and Raynar (1997) quantified the effects of scheduled overtime on productivity by studying the productivity of electrical and piping craftsmen on four active construction projects. Their study reported a loss of 10% ~ 15% efficiency for both scheduled overtime scenarios of 50 working hours and 60 working hours per week. Figure 2.3 Effective return from working 50 or 60 hours a week for various numbers of weeks (Source: Business Roundtable Cost Effectiveness Study Report C-3, November 1980.) 21 2.4.3 Disruptions Disruptions are considered to have a huge impact on construction productivity. Disruptions can be divided into two categories: short term A short term disruption leads to disruptions, and long term disruptions. productivity loss as extra work is needed to overcome obstacles causing disruptions. A long term disruption may even eradicate the productivity increases from learning curve effects (Halligan 1994). Thomas and Raynar (1997) classified disruptions into three categories, which are listed as follows: 1. Resources Material availability Tool availability Equipment availability Information availability 2. Rework Change Rework 3. Management Congestion Out-of-sequence work Supervisory 22 Miscellaneous In their study, each type of disruption was measured by frequency of occurrence during a working week. It was found that more working days per week were required when there was a higher frequency of disruption. Rework, tool availability, material availability, equipment availability, and congestion were found to have significant impact on performance. 2.4.4 Congestion and Accessibility Ovararin and Popescu (2001) conducted a study to quantify the effects of sixteen field factors on productivity loss in masonry construction. Fifty participants who were either owners or the chief estimators of the masonry contractors were randomly selected and a survey package was distributed to them. In their study, productivity losses due to levels of congestion and accessibility were quantified. The definition of levels of congestion and accessibility are The disruptions of an additional crew working in the same The results Levels of shown in Table 2.5. area were evaluated as the field condition on levels of congestion. reported that congestion caused 10 to 32 percent productivity loss. accessibility were evaluated by considering the convenience of accessing the Work Area and the distance between the Work Area and material storage. They found that disruptions associated with accessibility caused 13 to 35 percent productivity loss. 23 Table 2.5 Definition of Congestion and Accessibility (Ovararin and Popescu 2001) Field Factors Minor Standard Field Conditions Moderate Additional crews/contractors working in the same area 2~3 days/week High Congestion An additional crew/contractor working in the same area 1 day/week Additional crews/contractors working in the same area everyday Once/week, > 50 yards to materials storage Accessibility 2~3 days/week, 25~50 4 days/week, < 25 yards yards to materials to materials storage storage 2.4.5 Region The location of a construction project was found to be a factor influencing construction Production Rates. A productivity study was conducted by Koehn in 2001 to investigate the Production Rates in different regions in Bangladesh. Low production was found in rural areas. According to their investigation, lack of training and improper supervision was the major reasons for low production. Most big construction companies in Bangladesh are located in urban areas and only big construction companies provide training for the operation of sophisticated equipment. Moreover, low productivity can be due to workers' fatigue from long distance commuting (Borcherding and Alarcon 1991). The location of a project can affect both workers' motivation and the availability of advanced tools or equipment. Project location can also have an impact on the availability of Worker motivation (Borcherding 1980; 24 skilled labor (AbouRizk et al. 2001). Borcherding and Garner 1981) and the availability of skilled labor (Koehn and Brown 1985) have a huge impact on construction productivity. 2.4.6 Learning curve When performing repetitive tasks, productivity tends to increase as the number of cycles increase. This increased productivity is due to experience gained from previous tasks, improved resource allocation, better engineering support, better management and supervisions, and development of more efficient methods (Thomas et al. 1986). Thomas et al. (1986) conducted a study to evaluate the efficiency of various learning curve models on productivity estimation and to investigate the learning rates from four field studies. The learning rate is the rate of change of the cumulative average man-hours when production doubles. It was found that the learning rate was not constant and therefore a straight line model is not appropriate. Instead, the cubic power model was found to be the best learning curve model among five studied models. 2.4.7 Other factors Sanders and Thomas (1991 and 1993) conducted a series of research studies related to masonry productivity. The purpose of their 1991 study was to identify the major project-related factors that significantly influenced masonry productivity. They used ANOVA to examine whether the average masonry productivity in each category was statistically different for each of the factors 25 associated with masonry productivity. This study indicated that four of the investigated factors had significant impacts on masonry productivity: work type, building element, construction methods, and design requirements. In their 1993 study, a masonry prediction model was developed to estimate the masonry productivity based on different crew sizes as well as those factors identified in their 1991 study. analysis. This model was established by using the multiple regression The R square of their model was 0.411. 2.5 PRODUCTIVITY STUDIES OF EARTHWORK AND PAVEMENT 2.5.1 Earthmoving Production Rates and Match Factor The maximum possible Production Rate of an Excavation Operation is equivalent to the maximum Production Rate of the loading machine (Gransberg 1996; Smith 1999). This rate can only be achieved under ideal conditions with a sufficient number of trucks to allow the loading machine to maintain its maximum productivity. In reality, this condition seldom occurs due to concerns regarding cost-effectiveness of trucking and traffic conditions. Many studies have been conducted to estimate the Production Rate of Excavation Operations (Gransberg 1996; Peurifoy and Schexnayder 2002). The rates were determined based on the characteristics and numbers of loading machines and haul trucks, and characteristics of the haul road and excavated 26 materials. An example Production Rate calculation for an Excavation Operation is described in the following paragraph (Peurifoy and Schexnayder 2002). In this example, a loader is assumed to be used for the Excavation Operation. The loader cycle time is calculated according to the speed of a loading machine and the capacity of a truck. The truck cycle time is computed depending on the Furthermore, the optimum loading speed, traveling speed and unloading speed. number of trucks required for the Excavation Operation is calculated by dividing the truck cycle time by the loading cycle time. The optimum number is usually not an integer value. Therefore, the two integer values that are closest to the optimum number are established as the possible number of trucks for further Production Rate and unit cost analysis. If the number of trucks is more than the optimum value, the Production Rate of an Excavation Operation will be equivalent to the Production Rate of the loading machine, and is computed as shown in Equation 6. If the determined number of trucks is less than the optimum value, the Production Rate of an Excavation Operation will be equivalent to the Production Rate of the truck fleet, and is computed in Equation 7. Production Rate (lcy/hr) = Truck load (lcy) 60min Loader cycle time (min) (Equation 6) 27 Production Rate (lcy/hr) = Truck load (lcy) Number of Trucks 60min (Equation 7) Truck cycle time (min) Smith (1999) established a multiple regression model, with the R square of 90.6%, to predict earthmoving productivity. Four highway construction projects were involved in the study. From these four projects, a total of 141 earthmoving Operations were observed and analyzed. The factors included in the regression model were only the variables which could be determined or estimated in advance of earthmoving Operations. Six factors were identified as the major drivers from their earthmoving productivity model: number of trucks, number of buckets per load, volume per bucket, one-way haul length, match factor and travel time. The match factor (MF), as shown in Equation 8, is used to measure if there are sufficient haul units to reach the possible maximum Production Rate of an Excavation Operation. If the MF is greater than one, the possible maximum Production Rate of an Excavation Operation is equivalent to the maximum Production Rate of the loading machine. If the match factor is less than one, the possible maximum Production Rate is equivalent to the multiplication of the maximum Production Rate of the loading machine and the match factor. MF = Number of Haulers Loader Cycle Time Number of Loaders Hauler Cycle Time (Equation 8) 28 2.5.2 Truck Payload The truck payload is computed as the multiplication of the number of buckets per load and the volume per bucket. Schexnayder et al. (1999) conducted a study on the effect of truck payload weight on earthmoving productivity. The production of trucks was tracked for different haul distances and varying loading weights. It was found that the production of earthmoving increased with an increase in average payload of haul trucks but decreased when the average payload exceeded the rated gravimetric capacity. 2.5.3 Rainfall Rainfall has a great impact on highway construction productivity. El-Rays (2001) presented a decision support system that could quantify the impact of rainfall on productive day losses and estimate the duration for certain types of construction Operations in highway construction projects. A knowledge base of the effects of rainfall on productive day losses was acquired from interviews with experts in the highway construction industry. The experts indicated that three factors, namely the types of construction Operations, the intensity of rainfall and the drying conditions on site, are highly correlated to rainfall-related productivity losses. In addition they indicated that earthmoving, construction of the base course, construction of drainage layers and paving construction are the four tasks 29 in highway construction that are most sensitive to rainfall (El-Rays and Moselhi 2001). 2.5.4 Advance of Technology Technology advancements lead to improvement of construction productivity due to level of control, amplification of human energy, and information processing (Schexnayder and David 2002). Bhurisith and Touran (2002) conducted a case study with regard to obsolescence and equipment Production Rate. The ideal Production Rates of wheel-type loaders, track-type loaders, scrapers and crawler dozers were collected from the 1983, 1992 and 1998 Caterpillar Performance Handbooks. Production Rate changes according to change of technology were also examined. The results showed that Production Rates under ideal conditions have increased 1.58% on average per year due to technology advancements. Jonason et al. (2002) studied the productivity of Earthwork for different types of advanced positioning systems. In their study, the productivity of Earthwork for each advanced positioning system was estimated based upon site observation and interviews with field personnel. It was found that advanced positioning systems lead to improvements on schedule and cost performance of Earthwork construction due to saving time and reducing the cost of field surveying. However, there are still several shortcomings that inhibit the usage of these advanced positioning systems. The application of 2D and 3D guidance 30 technologies are limited to Work Areas with direct line-of-sight between the control station and the receiver on the equipment. signal noise can affect the accuracy of measurement. Goodrum and Hass (2002) studied the change of productivity and technology according to productivity data published by RS Means, Richardson, and Dodge between 1976 and 1998. They found a substantial improvement in partial factor Furthermore, GPS-related productivity among activities that have had significant improvements according to a technology index. The technology index was evaluated as a function of level of control, amplification of human energy, information processing, functional range, and ergonomics of equipment. It was found that site work has had the greatest improvement in mean partial factor productivity and technology index when compared to other work activities. Allmon et al. (2000) examined changes in construction productivity and unit cost for twenty Work Items according to productivity data published by RS Means between 1974 and 1996. It was found that the productivity of soil compaction and concrete placement increased by 260% and 55%, respectively. It was reported that new technology was the main driver of this improvement. 2.5.5 Traffic Jiang (2003) studied the effects of traffic flow on the construction productivity of Hot mix asphalt pavement. He observed 24-hour traffic flow at a cross-over Work Zone and used queuing theory to compute the cycle time of transporting 31 trucks in a hypothetical Hot mix asphalt Operation. According to the cycle time and an assumed number of transporting trucks, construction productivity of Hot mix asphalt pavement was computed. It was found that traffic delays increased the cycle time of transporting trucks. As a result of increasing cycle time, the construction productivity, in terms of tonnage per hour, decreased. However, adding more transport trucks could balance the negative effects of congested traffic flow. 2.5.6 Construction Productivity Associated with Concrete Pavement A constructability analysis tool was developed by Lee et al. (2000) to help the California Department of Transportation to examine the productivity performance and the traffic impacts of several strategies used on concrete pavement rehabilitation and construction in an urban area. A hypothetical concrete pavement construction (including the demolition of existing concrete pavement and base course, construction of Cement-treated base and construction of concrete pavement) was used to examine the variability of productivity performance with the variability of design profile, required curing time, working methods, paving strategies, truck capacity, and loading/discharging time. This hypothetical project involved the replacement of two outer lanes of a four-lane roadway during weekend closures from 10:00 p.m. Friday to 5:00 a.m. Monday. The process of pavement rehabilitation, lead-lag relationships between activities, constraints that limit construction productivity, approximate process productivity, and capacities 32 of equipments and facilities were gathered based on the previous urban freeway rehabilitation experience of a group of experienced California concrete paving contractors. Table 2.6 presents their findings in terms of percent reduction in ideal productivity (lane-km/a weekend closure) for different factors. Slab thickness was found to have the greatest impact on the productivity of concrete pavement rehabilitation because thicker slabs increase the quantity of demolition. The curing time of poured concrete varied with the usage of various types of concrete material. Because the construction time was limited to 55 working hours in a weekend closure and the constructed lanes had to open for traffic at the end of closure, more curing time lead to less construction time and output. The work method (that reflects the relative sequence of base construction and paving construction) also had a great impact on output. In addition, the paving lane (which refers to the working sequence of the two replaced lanes) and the end dump truck capacity and load/discharge time also had impacts on output. 33 Table 2.6 Percent Reduction in Production Capacity under Optimistic Conditions (adopted from Lee et al. 2000) Options Design Profile Comparison 203mm --> 254mm 203mm --> 305mm 254mm --> 305mm 4 hours --> 8 hours 8 hours --> 12 hours 4 hours --> 12 hours Concurrent --> Sequential Concurrent --> Sequential Double --> Single Double --> Single 22 Ton --> 15 Ton 3 Minutes --> 4 Minutes Reduction 40% 47% 12% 10% 11% 19% 29% 21% 17% 7% 15% 24% Curing Time Working Method 203-mm slab 254-or 305-mm slab 203-mm slab Paving Lane 254-or 305-mm slab End Dump Truck Capacity Load/Discharge Time 2.6 METHODS OF PRODUCTIVITY ANALYSIS Expert Systems are another technique to deal with relationships between productivity and driving factors. Hendrickson et al. (1987) developed an expert The system to predict the activity duration for masonry construction. productivity estimation, as a part of the activity duration estimation, included two steps. The first step was to estimate the maximum expected productivity and a subsequent step was to adjust the maximum rate to a reasonable rate according to the characteristics of the job or site. The information associated with productivity was established based on interviews with an experienced mason and a supporting laborer. Another Expert System was developed by Christian and Hachey (1995) to estimate the Production Rate of concrete pouring. After a 34 simple question-and-answer routine, the Expert System was able to estimate Production Rates of concrete pouring, depending on established decision rules. In addition, Neural Networks have been used by many researchers (Karshenas and Feng 1992; Lu et al. 2000; AbouRizk et al. 2001) to predict construction productivity. in data. A Neural Network has the capability of learning with an increase The greatest advantage of using Neural Networks to predict construction productivity is that it can include interactive effects of multiple factors in the productivity estimation if the network is trained using an adequate and representative data set. In reality, the size and quality of the training data set usually limits the effectiveness of Neural Networks due to lack of standards for collecting real productivity data. 2.7 ADVANCING TO PRESENT RESEARCH Although many studies have addressed construction productivity, few studies have been undertaken to study Production Rates for highway construction time estimation. The purpose of this study is to examine and determine the Production Rate in two Work Areas, namely Earthwork and Pavement construction for highway projects. Such information will help the Texas Department of Transportation (TxDOT) to improve the accuracy of highway construction time estimation and should lead to better project time management. 35 CHAPTER III: RESEARCH METHODOLOGY 3.1 OVERVIEW OF RESEARCH METHODOLOGY Figure 3.1 provides an overview of the research methodology. The research objectives and scope were determined first. A survey was conducted to select critical Work Items for the study while a comprehensive literature review helped in understanding relevant productivity factors and productivity measurement methods. A data collection process and associated tools were developed, Data associated with Production incorporating selected factors and methods. Rates were collected and analyzed. developed. Conclusions and recommendations were 36 Determining Research Objectives and Scope Survey to Select Work Items Develop a Questionnaire Identify Critical Work Items Select Work Items Literature Review Planning for Data Collection Identify Research Hypotheses Identify Candidate Drivers Determine Data Collection Process Develop Data Collection Tool Develop Site Visits Safety Protocol Conduct Pilot Data Collection Data Collection Select Job Site for Visiting Collect Data Data Analysis Summarize Production Rates Data Compare Mean Production Rate of Different Sources Identify Drivers of Production Rates Conclusions and Recommendations Figure 3.1 Research Methodology 3.2 RESEARCH FORMULATION The research objective is to develop improved information on Production Rates for highway Earthwork and Pavement construction so that Contract Time estimation can be enhanced. Production Rates as well as relevant factors were collected from a selection of TxDOT's on-going highway construction projects. Drivers of Production Rates for each Work Item were explored through statistical 37 analyses and their relationships with Production Rates were further quantified. In addition, CTDS Production Rates were compared with the field-collected rates to examine the differences. 3.3 SURVEY TO SELECT TARGETED WORK ITEMS A highway construction project usually includes hundreds of Work Items. Some of them are more critical to project schedule than others. A survey questionnaire, as shown in Appendix A, was used to identify the priority of Work Items in this study. The survey questionnaire includes a Work Item list, adapted from the TxDOT's Contract Time Determination System (CTDS), and the binary question (Yes or No) of tracking requirements for each Work Item. The Work Item list in the CTDS is a comprehensive list (Hancher et al. 1992) of Work Items for highway construction projects and it includes all major Work Items from the thirteen types of TxDOT construction projects. Survey participants were selected by the Project Monitoring Committee (PMC) of TxDOT research project 0-4416. The survey questionnaire was distributed to thirteen TxDOT engineers working in the Design and Construction divisions. The results of the survey are presented in Appendix B. Twenty five Work Items were indicated by more than seven engineers as the Work Items that should be tracked. construction. Eight of these belong to Earthwork and Pavement The results were presented and discussed in a PMC meeting. Consequently, seven Work Items related to Earthwork and Pavement construction 38 were established as priorities for this study. These include Excavation, Embankment, Lime-treated sub-grade, Aggregate base course, Hot mix asphalt pavement, Slip-form concrete pavement, and Conventional-form concrete pavement. 3.4 PLANNING FOR DATA COLLECTION A data collection process plan was developed in order to ensure the effectiveness and efficiency of data collection. Following that, data collection tools were developed. These tools include a project-level data collection tool (Appendix C), a Work Zone- and Work Item- level data collection tool (Appendix D), and individual Work Item sheets (Appendix E). The project-level data The collection tool was used to collect general information on selected projects. Work Zone- and Work Item-level data collection tool was used to document specific information regarding Work Items and Work Zone at the investigated site. Work Item sheets were developed for each targeted Work Item and were used as a guideline to ensure thoroughness and consistency in the data collection process. They were also used to collect information on specific Production Rate Factors. In addition, a site visit safety protocol (Appendix F) was developed to ensure safety during the data collection process. Upon finalization of the data process and tools, a pilot data collection effort was carried out in order to test the process and tools. Further adjustments to the process and tools were made to improve the effectiveness of data collection. 39 3.4.1 Data Collection Process A data collection process plan, shown in Figure 3.2, was developed to enhance the effectiveness of collecting data. Three cycles were included in this plan. The first consists of the process flows of conducting a district meeting to select projects for data collection, the second involves conducting a project meeting to kickoff the data collection in a project and third, the regular collection of project data at the construction site. Data Collection Process for TxDOT Project 04416 PROCESS Repeat at 8 to 10 weeks Cycle Setup District Meeting & Prepare Meeting Documents List the Priority of Work Items District Meeting Introduce research project Safety protocol Screen projects OUTPUT List of Projects Criteria: Percentage of completion < 80% Contract Working Days > 120 days Construction Projects (draw from monthly construction estimate report) Select Projects for Study List of Projects selected Project Meetings Introduce research project Safety protocol Collect Project Information Site Trip Contact Information Site Trip Site location Check the scope of work items Benchmarking Organize project information Input project information into database Collect project information General Information On-going work items & possible time table Information of Project Characteristics Repeat Once for Every Project Setup Project Meetings Project -Level Data Arrange Transporation Prepare Materials for Visiting Repeat Site Visiting Repeat for Every Trip Contact & Arrange/Confirm Site Visiting Monitor Projects via email, phone Organize Data Collected Input Data into Database Work Zone & Work Item Level Data Figure 3.2 Data Collection Process 40 3.4.1.1 Selecting Job Sites for Data Collection The first cycle in Figure 3.2 displays tasks associated with the district kickoff meeting. weeks. District meetings were conducted at an interval of every eight to ten Two important meeting tasks included the introduction of the research to the district construction engineer and engineers from area offices, and obtaining input for selecting appropriate projects for data collection. In addition, the research team was required to obtained permission and further guidance on collecting data and confirmed safety procedures. The information of on-going projects were reviewed from the construction report highways and construction monthly estimate report on the TxDOT web site (http://www. dot. state. tx. us/business/projectreports. htm). Projects which were less than 80% complete and that had a contract project duration greater than 120 working days were listed for the further screening at the district meeting. Projects with serious delays due to legal problems or change orders were eliminated as any Production Rates collected would likely be outliers. 3.4.1.2 Project Meetings Project meetings were conducted following a district meeting to achieve three objectives. First, the purpose of the research was introduced to site personnel to facilitate data collection. An introduction of the research project was presented at the beginning of the project meetings to obtain the assistance from the TxDOT 41 site personnel and to prevent redundant or incorrect information from being collected. Secondly, detailed information regarding selected Work Items and their respective work schedules were obtained. Thirdly, general information on the targeted projects was collected and work progress status was benchmarked. Sources of project information included project contracts, project drawings, and some project manager opinions. 3.4.1.3 Regular Visits The research team visited the selected job sites on a regular basis to benchmark and collect data after the project meeting. Instead of stop-watch type The first step observations, discrete observations were employed to collect data. of discrete observation was to "benchmark" the initial status of targeted activities. The location and progress of a targeted activity were documented and characteristics of the Work Zone were evaluated. Work quantity completed-to- date and crew and machinery information pertaining to the elapsed period between the first benchmarking/last observation and current observation were documented along with any disruptions and the number of working days. The production quantities were collected from the TxDOT diary and/or, site management or contractor reported quantities that had been approved by TxDOT personnel. Working day and disruption information were collected from The time interval between two observations interviews with TxDOT personnel. 42 ranged from one to two weeks. were computed. At the end of each regular visit, new data points Subsequent observations were carried out as necessary. 3.5 DATA COLLECTION A total of seven TxDOT districts, as shown in Table 3.1, were selected to collect Production Rate data for this study. The state was divided into four areas according to weather and terrain, and up to three districts were selected from each of these four areas to avoid bias caused by weather and region. Two to six projects in each district were observed simultaneously for a period of two to three months. Table 3.1 Selected Districts vs. Area Districts D1 D2 D3 D4 D5 D6 D7 Area of Texas Central and South Texas Coastal Central and South Texas North Texas Coastal Panhandle and West Texas Central and South Texas To ensure consistency in the tracking of working days, a rational for Production Rate computation was determined by the research team and TxDOT's Project Monitoring Committee (PMC) members and is displayed in Table 3.2. If there was a delay effect of not greater than two hours in a 10 working-hour day 43 and the delay was caused by weather, unworkable soil conditions, traffic accident, construction accident, equipment down time, unavailability of material, trade problem or absenteeism, the day was counted as a working day. If the delay effect was greater than two hours but less than half a day, the day was counted as a half working day. If the delay effect was more than half a day, it was For Lime-treated sub-grade, if a 1st curing considered as a non-working day. occurred on Holidays, Non-working days, Non-working weekends, and Off-day, the days were added to total working days when the duration was not greater than 2 days. Thus, the maximum total duration of 1st curing was limited to 2 days if Holidays, Non-working days, Non-working weekends, and Off-day were counted. For delays caused by Right of way, Unforeseen conditions, or Instructions from TxDOT's engineers, delays were not counted as working days. Also, total working days were adjusted if overtime was more than two hours per day. 44 Table 3.2 Rational for Production Rate Computation Factors Weather (Rain, Too Wet, Snow, Wind etc) Unworkable Soil Conditions Traffic Accident Construction Accident Equipment Down Time Material Unavailable Trade Problem Absenteeism If Delay Effect < Day If Delay Effect >= Day No Adjustment Effect Embedded in the Production Rate Corrected Effect isolated or adjusted If Number of Days of 1st Curing Holidays, Non-Working Day, <= 2 Days Non-Working Weekend, Off-Day #260 Lime Treated Subgrade 1 st Curing only Regional shortage (ROW, Unforeseen Condition, TxDOT Direction) Overtime 3.6 DATA ANALYSIS 3.6.1 Descriptive Statistics and Box Plots Descriptive statistics were often employed to summarize data such as mean, sum, counts, and frequency of variables. In this research, data are shown on scatter plots to demonstrate relationships or associations between two variables. Relationships may be observed with non-random scatter in such plots. A box plot is a statistical summary that presents mean, median, quartile, outliers and extreme values in a graphical format. Figure 3.3 is an annotated The horizontal line sketch of a box plot (SPSS Base 10.0 Applications Guide). in the shaded box represents the median or 50th percentile of the plotted sample. 45 The dark circle highlights the mean of the targeted sample. The top and bottom The end of the box represents the 3rd and 1st quartile of the sample respectively. length of the box, from 1st quartile to 3rd quartile, denotes the inter-quartile range (IQR). The horizontal line between 3rd quartile and 3rd quartile + 1.5 * IQR and between 1st quartile and 1st quartile 1.5 * IQR are the highest and lowest observed values respectively, excluding outliers in the sample. Points beyond the (3rd quartile + 1.5 * IQR) and under the (3rd quartile + 3 * IQR) as well as points under the (1st quartile 1.5 * IQR) and beyond (1st quartile 3 * IQR) are outliers. Points beyond these outer limits are considered extreme values. Extremes 1.5 * IQR 1.5 * IQR Inter-Quartile Range (IQR) 1.5 * IQR Smallest observed value that is not an outlier 1.5 * IQR Outlier Extremes Outlier Largest observed value that is not an outlier 3rd quartile Horizontal line: Median; Circle: Mean 1st quartile Figure 3.3 Annotated Sketch of the Box Plot 46 Side-by side box plots were used to compare the Production Rates of the observed data, CTDS, and historical records. Table 3.3 shows an overview of historical records utilized by this study. districts in Texas, namely D3, D4 and D8. These were collected from three Both D3 and D4 districts were also involved in the Production Rates observations. The historical records collected from these two districts were gathered from on-going projects. Production Rate information was retrieved from daily logs and payment management systems of the general contractors. The historical records from D8 were collected from quantity and working-day records in their schedule network of eleven completed projects. Table 3.3 General Information of the Sources of Historical Records District D3 D4 D8 Project ID As-Built 1 As-Built 2 As-Built 3 As-Built 4 As-Built 5 As-Built 6 Other As-Builts Number of Project 1 1 1 1 1 1 9 Progress Status of Records Start ~ 62% Start ~ 60% Start ~ 34% Start ~ 29% Start ~ End Start ~ End Start ~ End 47 3.6.2 Test of the Difference of Mean Observed Production Rates and Average CTDS Production Rates Because little information of original data is available to determine the distribution of the Production Rate data in the CTDS study, the Average CTDS Production Rates were compared with the mean observed Production Rate for the seven targeted Work Items. The one-sample t test was used for this comparison. 3.6.3 Driver Analysis Procedures used for driver analysis are shown in Figure 3.4. Factors that are suspected to have significant effects on Production Rates and are known at the design stage were considered as Candidate Drivers. Once Candidate Drivers were identified, associated data were collected during regular job visits. Scatter plots were used to examine any relationships between observed Production Rates and each Candidate Driver. Drivers having no obvious relationship with observed Production Rates were excluded from further analysis. Based on the data attributes of the Candidate Drivers, two types of analysis approaches were used for further driver analysis. For those Candidate Drivers with continuous numerical data, regression analysis was conducted to identify drivers of Production Rates and to quantify their effects. For those Candidate Drivers with discrete numerical or categorical data, the ANOVA or t-test was used to test the difference in mean Production Rates for the subsets in each Candidate Driver. 48 According to the results from statistical analyses, the drivers were thus identified. The quantitative effects of drivers on Production Rates were also investigated. In addition, the correlations between identified drivers of each targeted item were computed to be used for reference on estimating effects of multiple drivers. If data were sufficient, multiple regression analysis was used to further investigate the interaction effects of multiple drivers. Create detailed influence diagrams Establish hypotheses: Identify candidate drivers of production rates Scatter Plots: Examine correlation of observed production rates vs. candidate drivers Yes Drivers with categorical data or discrete numerical data No Stop Drivers with continuous numerical data Use ANOVA/t-test to test difference in mean production rates Use regression analysis to explore the relationship of drivers and production rates Test research hypotheses Identify drivers of production rates and quantify their effects Test correlations of identified drivers Yes Sufficient data points for multiple regression analysis No Use multiple regression analysis to identify the interaction effects of multiple drivers Stop Figure 3.4 Flow Chart of Driver Analysis 49 3.6.3.1 Test of the Difference of Mean Observed Production Rates between Sub-groups of Candidate Drivers The independent-samples t-test is one of the most popular methods of testing the differences between two population means. Three basic assumptions should be examined before applying the t-test. The three assumptions are as follows: 1. The two samples are independent 2. Populations are normally distributed 3. There are equal standard deviations between the two populations If the two samples are not independent, the independent-sample t-test will not be efficient to test the differences in mean between the two groups and other test methods such as the paired-sample t-test may be used. The second assumption that the populations are normally distributed can be examined from Q-Q plots. If all data falls on a line with a 45 degree of slope on the Q-Q plot, a typical normal distribution can be identified. If this assumption is violated, the results The of the t-test can only be used when the size of samples is reasonably large. last assumption is that the standard deviations of two tested populations should be equal. This assumption can be examined from the results of Levene's test in the The result of the t-test may be incorrect if SPSS version 11.0 for Windows. this assumption is violated, but the t-test can have an accurate result if the sample sizes are equal under this circumstance. These methods were applied in this 50 study to verify some research hypotheses and to identify some Production Rate drivers. 3.6.3.2 Regression Analysis Once a linear or non-linear relationship between two variables is observed from the scatter plot, a linear or non-linear regression analysis should be performed to verify if a relationship exists statistically. regression model is Yi = b0 + b1 * X1i + b2 * X2i. that a study is trying to predict. The form of estimating a Yi is the dependent variable In X1i and X2i are the independent variables. advance of conducting a regression analysis, the sample size should be checked if data are sufficient to perform it. According to a study conducted by Green (1991), the required sample size for a regression analysis can be determined by four values which are (the probability of making a type I error), 1- (one minus the probability of making a type II error), R2, and number of predictors. Table 3.4 displays the required sample sizes to test the hypothesis that the population multiple correlation equals zero with a power (1-) of 0.8 and of 0.05 based on power analysis (Green 1991). A regression model needs 24 data points for one predictor and 30 data points for two predictors when the , 1-, and R2 values used to determine the statistical significance of a regression model are 0.05, 0.8 and 0.26, respectively. If the required R2 used to determine the significance of a regression model increases, the number of data points can be reduced. In this study, the required R2 51 is set as 0.34. Therefore, for this study a total of 20 data points are required to perform a simple regression analysis, and 26 data points are needed to perform a multiple regression analysis with two predictors. However, less than 20 data points may be also employed for a regression analysis if a higher R square is achieved. Table 3.4 Sample Sizes Required to Test the Hypothesis that the Population Multiple Correlation Equals Zero with a Power of 0.80 and of 0.05 (adopted from Green 1991) Number of Predictors 1 2 3 4 5 6 7 8 9 10 15 20 30 40 Sample Sizes based on Power Analysis R =0.02 390 481 547 599 645 686 726 757 788 844 952 1066 1247 1407 2 R =0.13 53 66 76 84 91 97 102 108 113 117 138 156 187 213 2 R =0.26 24 30 35 39 42 46 48 51 54 56 67 77 94 110 2 In addition, the logarithmic model (Equation 3.1) and the power model (Equation 3.2) were employed to identify non-linear relationships between 52 selected cases with two variables. SPSS version 11.0 for Windows was used to perform the linear and non-linear regression analyses. Yi = b 0 + b 1 * Log X i Log Yi = Log b 0 + b 1 * Log X i (Equation 3.1) (Equation 3.2) Six steps are usually taken to perform a regression analysis. First, the dependent and independent variables should be checked to see if they are approximately normally distributed. The normal distributions of independent variables and dependent variable are basic assumptions of a regression analysis. Violation of this assumption would lead to a biased estimation due to lack of information. Secondly, a scatter plot is developed to check for a plausible linear Outliers should be removed model and a box plot is used to detect outliers. before performing a regression analysis because they impact the trend of the regression model. The third step is to fit the linear regression model and produce results of the regression analysis. In this step, the R2, the adjusted R2, and the P-values are computed. The next step is to inspect the R2 of the fitted model. The coefficient of determination, or R2, is also called the measurement of the goodness of fit of the regression line. The value of R2 is always between 0 and 1, and indicates the proportion of variation of dependent variables that can be 53 explained by the prediction model. The formula (Albright et al. 1999, p583) for calculating R2 in a simple linear model is shown in Equation 3.3. R2 = 1 - (Yi ei 2 - Y) 2 (Equation 3.3) ^ ^ Where, ei = Yi - Yi and Yi = b 0 + b1 X i ^ Yi : Observed Value; Yi : Fitted value of Yi The fifth step is to inspect the results of testing coefficients for the fitted model. The t-test is applied to test coefficients. The P-values of the t-tests should be used to check if the coefficients of the fitted model are statistically different from 0. A P-value, less than , indicates that the null hypothesis of a coefficient being equivalent to zero can be rejected at the (1- ) confidence interval. In contrast, a P-value, not less than , represents that the tested coefficient is not statistically different from zero and thus, there is no relationship between the dependent variable and the independent variable. In this study, 0.05 and 0.1 were used as the value of . The difference between applying 0.05 and 0.1 to hypothesis test is the level of confidence to conclude if the tested coefficient is significantly different from zero. The last step is to check for violation of model assumptions. Other than the approximate normal distribution of dependent and independent variable, three assumptions: (1) constant variance of errors; (2) normal distribution of errors; and (3) no high correlations between explanatory variables; should be checked. 54 The constant variance of errors can be examined by plotting the scatter plot of the predicted value of the fitted model versus the residuals. Non-constant variance of errors found in the regression model usually indicates the need for transformation of variables or adding another important variable. The normal distributions of variables and errors can be inspected by observing their Q-Q plots. If the data are perfectly normally distributed, the points in the Q-Q plot will "cluster around the 45 line. Any large deviations from a 45 line signal some type of non-normality" (Albright et al. 1999, p486). 3.6.3.3 Correlations Analysis The Pearson product-moment correlation tests were used to check the correlations between the explanatory variables. The Pearson product-moment correlation, or , is a value between -1 and 1. A correlation equal to or near zero On the other indicates no linear relationship existed between the two variables. hand, a correlation with a magnitude close to 1 indicates a strong linear relationship. 55 CHAPTER IV: DATA COLLECTION PLAN AND EXECUTION Reliable Production Rates estimation should include consideration of the impacts of drivers on Production Rates to reflect reality. Production Rates for Earthwork and Pavement Work Items and related factors were collected from thirty-five TxDOT on-going highway construction projects and analyzed statistically to investigate both the Production Rates and the drivers that have significant impacts on Production Rates. possible. Such effects were quantified whenever 4.1 RESEARCH HYPOTHESES Several PMC members believed that the Production Rates in the CTDS for most Work Items in Earthwork and Pavement are too optimistic when compared to realistic rates. the following. Hypothesis 1: The Production Rates of the CTDS are not realistic. Production Rates for Earthwork and Pavement vary significantly due to the effects of productivity factors. influence on Production Rates. Some productivity factors may have great The second hypothesis was established based on Pertaining to this issue, the first hypothesis was established as this assumption and was specified as follows. 56 Hypothesis 2: The Production Rates of targeted Work Items are driven by some productivity factors that are known at the design stage. 4.2 CANDIDATE DRIVERS OF TARGETED WORK ITEMS The purpose of this study was to investigate the Production Rates and identify the drivers that significantly influence Production Rates for Excavation, Embankment, Lime-treated sub-grade, Aggregate base course, Hot mix asphalt pavement, Slip-form concrete pavement and Conventional form concrete pavement. Influence diagrams were utilized to probe the possible drivers of Factors directly or indirectly The results Production Rates for each targeted Work Item. influencing Production Rates were listed in the influence diagrams. are intended to be used for Contract Time estimation at the design stage. Therefore, the factors identified in the influence diagram should be limited to those known at the design stage. 4.2.1 Candidate Drivers for Excavation The influence diagram of the Production Rate (CY/Crew Day) for Excavation is shown in Figure 4.1. In the influence diagram, the factors were divided into three categories: project-level, work-zone level and work-item level. Only those factors that are known at the design stage appear in bolded circles in the influence diagram. These factors which were considered as the Candidate Drivers of the Production Rates of Excavation are listed as follows. 57 Project-Level: 1. Project Type: The type of project may influence the Production Rates of Excavation Operations due to different site layout, size of work, and strategies of traffic control. 2. Project Location: Project location may influence the Production Rates of Excavation Operations because of the availability of resources, site condition and traffic condition. 3. Traffic Flow: Traffic congestion may decreases Production Rates of haul trucks and therefore may reduce Production Rates of Excavation Operations. 4. Project Complexity: Projects with higher technical complexity may result in more interactions between different crews and, thus, may have more limitations on work space and accessibility and may affect Production Rates of Excavation Operations. 5. Accelerated Construction Provision: As a result of accelerated construction provision, contractors may contribute more resources and efforts to work on Excavation Operations. Therefore, projects with accelerated construction provision may have a higher average Production Rate. 6. Contractor Management Skill: Contractors with better management skill may have higher Production Rates of 58 Excavation due to appropriate supervising and resource allocation. Work Zone-Level: 1. Work Zone Accessibility: Short distance and good haul road conditions result in more efficient transporting of excavated materials and, thus, may have higher Production Rates. 2. Work Zone Congestion: More space in a Work Zone to locate excavators, loaders and loading trucks, and for the waiting truck queue may increase the efficiency of loaders and, thus, may have higher Production Rates of Excavation. 3. Work Zone Drainage Effectiveness: The Production Rates of Excavation in Work Zones with less efficient drainage may be adversely affected as rain may worsen the condition of haul road. Work Item-Level: 1. Work Area Quantity: Based on the fact that the greater the amount of repetitive work in a Work Area leads to more efficiency of work Operations and resource allocation. This may be true for Excavation because Excavation Operations are highly repetitive. In addition, productive hours in a working day may be higher for the Work Area with greater quantity. 59 Figure 4.1 Influence Diagram of the Production Rate (CY/Crew Day) for Excavation 60 2. Soil Condition: Loose materials may be more easily excavated and therefore may have a better average Production Rate. 4.2.2 Candidate Drivers for Embankment The influence diagram of the Production Rate (CY/Crew Day) for Embankment is shown in Figure 4.2. follows: Project-Level: 1. Project Type: The type of project may influence the Production Rates of Embankment Operations due to size of work, working sequence, and strategies of traffic control. 2. Project Location: Project location may influence the Production Rates of Embankment Operations due to traffic conditions. 3. Traffic Flow: Traffic congestion usually decreases the Production Rates of haul trucks and therefore may reduce the Production Rates of Embankment Operations. 4. Project Complexity: Projects with higher technical complexity may result in more interactions between different crews and, thus, may have more limitations on work space and accessibility. This too, may affect Production Rates of The Candidate Drivers are listed as Embankment Operations. 61 5. Accelerated Construction Provision: As a result of accelerated construction provision, contractors may contribute more resources and efforts to work on Embankment Operations. Therefore, projects with accelerated construction provision may have a higher average Production Rate. 6. Contractor Management Skill: Contractors with better management skills may have higher production due to appropriate supervision and resource allocation. Work Zone-Level: 1. Work Zone Accessibility: Short distance and good haul road condition are more efficient for transporting excavated materials and so may have a better average Production Rate. 2. Work Zone Congestion: Large free space in the Work Zone allows for unloading, furthermore spreading and compacting can be operated simultaneously. This may increase the Production Rates of Embankment Operations. 3. Work Zone Drainage Effectiveness: Work Zones with less efficient drainage may have interruptions on transporting of materials after rain due to the wet condition of haul road. It may also influence the efficiency of compaction because of excessive water content. Work Zone drainage effectiveness 62 may be a driver of the Production Rates on Embankment Operations. Work Item-Level: 1. Work Area Quantity: Repetition leads to more efficient work Operations and resource allocation. This may be applicable for Embankment because these Operations are high repetitive. In addition, productive hours in a working day may be higher for the Work Area with greater quantity. 2. Soil Conditions: Soil conditions will influence the required number of compaction passes to achieve the designed density, therefore affecting Production Rate. 63 Figure 4.2 Influence Diagram of the Production Rate (CY/Crew Day) for Embankment 64 4.2.3 Candidate Drivers for Lime-Treated Sub-grade The influence diagram of the Production Rate (SY/Crew Day) for Limetreated sub-grade is shown in Figure 4.3. The Candidate Drivers are listed as follows: Project-Level: 1. Project Type: The type of Project may influence the Production Rates of Lime-treated sub-grade Operations due to different site layout, size of work, and dispersion of work. 2. Project location: The layout of drive ways and intersections in rural, urban and metro areas are very different. This may have different impacts on dispersion of work and thus, may influence the Production Rates of Lime-treated sub-grade Operations. 3. Project Complexity: Projects with higher technical complexity may lower the average Production Rate of Lime-treated subgrade for various reasons, for example higher interactions between different crews and highly dispersed works. 4. Accelerated Construction Provision: As a result of accelerated construction provision, contractors may put more resources and effort to work on Lime-treated sub-grade Operations. Therefore, projects with accelerated construction provision may have a higher average Production Rate. 65 5. Contractor Management Skill: Contractors with better management skill may have higher Production Rates due to better supervision and resource allocation. Work Zone-Level: 1. Work Zone Congestion: A large Work Zone may allow mixing, compacting and finishing simultaneously and may have higher Production Rates of Lime-treated sub-grade Operations. 2. Work Zone Clay Content: The Work Zone with higher clay content needs more lime for the treatment of the soil. Therefore, the mixing speed may be slower so the Production Rates of Lime-treated sub-grade may be lower. 3. Work Zone Land Slope: The slope of a Work Zone will affect the speed of operating equipment and influence the efficiency of elevation and grade control. Work Item-Level: 1. Work Area Quantity: Lime-treated sub-grade Operations may experience increasing productivity as it is highly repetitive. In addition, productive hours in a working day may be higher for the Work Area with greater quantity. 2. Length of Work Area: When the number of repetitions increases, Production Rates will be higher due to learning 66 effects. A longer Work Area means this work function can expect increasing productivity. 3. Type of Lime Used: The required duration of curing may vary according to types of lime used, therefore it may influence the average Production Rate of Lime-treated sub-grade. 4. Lift-Thickness: The construction time of mixing and compacting may be longer in a thicker lift and thus, the Production Rates in a thicker lift may be lower. 5. Soil Condition: The condition of soil may influence the speed of mixing and compacting. 67 Figure 4.3 Influence Diagram of the Production Rate (SY/Crew Day) for Lime-Treated Sub-grade 68 4.2.4 Candidate Drivers for Aggregate Base Course The influence diagram of the Production Rate (Lift-SY/Crew Day) for Aggregate base course is shown as Figure 4.4. The Candidate Drivers are listed as follows: Project-Level: 1. Project Type: The type of project may influence the Production Rates of Aggregate base Operations due to different site layout size of work and dispersion of work. 2. Project Location: The layout of drive ways and intersections in rural, urban and metro areas are very different. This may influence the Production Rates of Aggregate base construction. 3. Project Complexity: Projects with higher technical complexity may have lower Production Rates of Aggregate base Operations. 4. Accelerated Construction Provision: Projects with accelerated construction provision may have higher Production Rates. 5. Contractor Management Skill: Contractors with better management skill may have higher Production Rates. Work Zone-Level: 69 1. Work Zone Congestion: Large Work Zones may allow spreading, compacting, and finishing simultaneously and thus, may have better average Production Rates. 2. Work Zone Land Slope: The slope of a Work Zone will affect the speed of operating equipment and the efficiency of elevation and grade control. Work Item-Level: 1. Work Area Quantity: Aggregate base may over time experience increasing productivity because its Operations are highly repetitive. In addition, productive hours in a working day may be higher for the Work Area with greater quantity. 2. Lift-Length of Work Area: The longer the Lift-length of a Work Area for Aggregate base the more the number of repetitions there will be. When the number of repetitions increases, Production Rates will be higher due to learning effects. 3. Width of Work Area: A wider Work Area may allow more equipment to work at the same time and thus may have higher average Production Rates. 4. Lift-Thickness: The construction time of compacting may be longer in a thicker lift. Therefore, the Production Rates of Aggregate base in a thicker lift may be lower. 70 Figure 4.4 Influence Diagram of the Production Rate (Lift-SY/Crew Day) for Aggregate Base Course 71 4.2.5 Candidate Drivers for Hot Mix Asphalt Pavement The influence diagram for the Production Rate (Ton/Crew Day) of Hot mix asphalt pavement is shown in Figure 4.5. The Candidate Drivers are listed as follows: Project-Level: 1. Project Type: The type of project may influence the Production Rates of Hot mix asphalt pavement Operations due to different site layout, size of Hot mix asphalt, strategies of traffic control and dispersion of work. 2. Project Location: The frequency of drive ways and intersections in rural, urban and metro area are very different. This may have different impacts on dispersion of work and, thus, may influence the Production Rates of Hot mix asphalt pavement. In addition, the traffic condition in different type of location may have an impact on the Production Rates. 3. Traffic Flow: Traffic flow may influence the Production Rates of Hot mix asphalt pavement Operations because it influences the efficiency of logistics. 4. Project Complexity: Projects with higher technical complexity may result in more interactions between work and traffic. This 72 may lead to high dispersion of work and thus, may lead to lower productivity. 5. Accelerated Construction Provision: As a result of accelerated construction provision, contractors may put more resources and effort into work on Hot mix asphalt pavement Operations, leading to a higher average Production Rate. 6. Contractor Management Skill: Contractors with better management skills may have higher Production Rates of Hot mix asphalt pavement due to better supervision, engineering and resource allocation. Work Zone-Level: 1. Work Zone Accessibility: Working in easily accessible Work Zones, contractors can better manage the transportation of Hot mix asphalt and may reduce the frequency of interruptions due to material shortage. Therefore, a better average Production Rate may be expected when working in easily accessible Work Zones. 2. Work Zone Congestion: Working in congested Work Zones, contractors may need more time to unload Hot mix asphalt and thus, increase the waiting time of the lay-down machine. may have lower Production Rates. They 73 3. Work Zone Land Slope: The slope of a Work Zone may influence the speed of operating equipments and affect the efficiency of elevation and grade control. Therefore, slope may influence the Production Rate of Hot mix asphalt Operations. Work Item-Level: 1. Work Area Quantity: Increased repetition in a Work Area leads to more efficiency. This may be true for Hot mix asphalt In pavement because its Operations are highly repetitive. addition, productive hours in a working day may be higher for the Work Area with greater quantity. 2. Course Type: The Surface course is usually built with a higher standard of quality as compared to the Base course. Therefore, a lower average Production Rate of the Surface course may be expected. 3. Main Lane vs. Non-main Lane Application: The location of work such as on the main lane, the frontage road, or on a ramp may influence the Production Rates of Hot mix asphalt Operations due to dispersion of work. The main lane usually has lesser dispersion of Hot mix asphalt work than the frontage 74 road or on a ramp. The Production Rates of Hot mix asphalt in the main lane may be higher than other areas. 75 Figure 4.5 Influence Diagram of the Production Rate (Ton/Crew Day) for Hot Mix Asphalt Pavement 76 4.2.6 Candidate Drivers for Slip-form Concrete Pavement The influence diagram of the Production Rate (SY/Crew Day) for Slip-form concrete pavement is shown as Figure 4.6. follows: Project-Level: 1. Project Type: The type of project may influence the Production Rates of Slip-form concrete pavement Operations due to different site layout, size of work and dispersion of work. 2. Project Location: Rural, urban and metro area may have very different impacts on dispersion of work and thus, may influence the Production Rates of Slip-form concrete pavement. 3. Project Complexity: Projects with higher technical complexity will result in more interactions between different crews and may lead to lower average Production Rates of Slip-form concrete pavement. 4. Traffic Flow: Traffic flow can affect the efficiency of logistics which may influence the Production Rates of Slip-form concrete pavement. 5. Accelerated Construction Provision: As a result of accelerated construction provision, contractors may put more resources and effort into work on Slip-form concrete pavement Operations. 77 The Candidate Drivers are listed as Projects with accelerated construction provision, thus they may have a higher average Production Rate. 6. Contractor Management Skill: Contractors with good management may have higher Production Rates of Slip-form concrete pavement due to better supervision, engineering and resource allocation. Work Zone-Level: 1. Work Zone Accessibility: Working in easily accessible Work Zones, contractors can better manage the transportation of concrete so that the frequency of interruptions due to materials shortage may be reduced. Therefore, a higher average Production Rate may be expected in easily accessible Work Zones 2. Work Zone Congestion: Large Work Zones will allow transitmix trucks to wait in the Work Zone, which may reduce the waiting time due to materials shortage. Therefore, Work Zones with less congestion may have higher Production Rates. 3. Work Zone Land Slope: The slope of a Work Zone may influence the speed of a paver, the transition time of locating transit-mix trucks, and the duration of grading. Therefore, the 78 land slope of a Work Zone may be a driver of the Production Rate of Slip-form concrete pavement Operations. Work Item-Level: 1. Type of Concrete Pavement: Three types of Slip-form concrete pavement are used in Texas. Each has a different scope of work, for example, reinforced concrete pavement may require more working days on rebar installation than the other two types. Therefore, the reinforced concrete pavement may have In a lower average Production Rate than other two types. addition, productive hours in a working day may be higher for the Work Area with greater quantity. 2. Work Area Quantity: Based on the fact that the greater the amount of repetitive work in a Work Area the more efficient the work Operations and resource allocation will be, Slip-form concrete pavement may experience better Production Rate because its Operations are highly repetitive. 3. Length of Work Area: The longer the length of a Work Area for Slip-form concrete pavement the more the number of repetitions there will be. When the number of repetitions increases, Production Rates will be higher due to learning effects. 79 4. Width of Work Area: A wider Work Area usually has minor problems of congestion and it is convenient for transit-mix trucks to wait and unload concrete. Therefore, a wider Work Area may have a better Production Rate. 5. Thickness of Concrete Pavement: Thicker concrete pavement needs more concrete for a certain area. Therefore, it requires more time for unloading and consolidating of poured concrete and thus, may have lower Production Rates. 80 Figure 4.6 Influence Diagram of the Production Rate (SY/Crew Day) for Slip-form Concrete Pavement 81 4.2.7 Candidate Drivers for Conventional Form Concrete Pavement The influence diagram of the Production Rate (SY/Crew Day) for Conventional form concrete pavement is shown as Figure 4.7. Drivers were listed as follows. Project-Level: 1. Project Type: The type of project may influence the Production Rates of Conventional form concrete pavement Operations. 2. Project Location: Project location may influence the Production Rates of Conventional form concrete pavement Operations due to its impact on traffic conditions. 3. Project Complexity: Projects with higher technical complexity may have lower Production Rates. 4. Traffic Flow: Traffic flow affects the efficiency of logistics and thus, may influence the Production Rates of Conventional form concrete pavement. 5. Accelerated Construction Provision: Projects with accelerated construction provision may have a higher average Production Rate. 6. Contractor Management Skill: Contractors with better The Candidate management skills may have higher Production Rates. 82 Work Zone-Level: 1. Work Zone Accessibility: Working in easily accessible Work Zones may lead to higher Production Rates. 2. Work Zone Congestion: Work Zone with large space may allow transit-mix truck to wait in the Work Zone and thus, reduce the waiting time due to material shortage and the transition duration of unloading concrete. Therefore, Work Zones with less congestion may have better Production Rates. 3. Work Zone Land Slope: The slope of a Work Zone may influence the transition duration of unloading of transit-mix trucks, and have an impact on the duration of grading and finishing. Therefore, Work Zone land slope may influence the Production Rates. Work Item-Level: 1. Work Area Quantity: The repetitive nature of conventional from concrete pavement may allow for increasing productivity. In addition, productive hours in a working day may be higher for the Work Area with greater quantity. 2. Configuration of Concrete Pavement: When the Configuration of Concrete Pavement has any curve or sharp angle, the duration of performing the formwork and rebar installation for this 83 Concrete Pavement may be longer than the Concrete pavement that has a Configuration without any curve or sharp angle. Therefore, a lower average Production Rate may be expected for the Concrete pavement for the former Configuration. 3. Thickness of Concrete Pavement: A thicker depth concrete pavement usually needs more concrete to complete a fixed area. It may also require more time for unloading and consolidating of poured concrete. Rate may be expected. Therefore, a lower average Production 84 Figure 4.7 Influence Diagram of the Production Rate (SY/Crew Day) for Conventional form Concrete Pavement 85 Table 4.1 summarizes the selected Candidate Drivers for the seven targeted Work Items. Data attributes of the Candidate Drivers at the project-level are Candidate Drivers at the displayed in the project-level data collection tool. Work Zone-level were discussed in detail in Section 4.4.2. Table 4.2 displays data attributes of Candidate Drivers at the Work Item-level. These Candidate Drivers were further investigated in the driver analysis of Chapter 6 and 7. Table 4.1 Candidate Drivers vs. Seven Targeted Work Items Candidate Drivers Project Type Project Location Project Level Traffic Flow Project Complexity Accelerated Construction Provision Contractor Management Skill Work Zone Accessibility Work Zone Level Work Zone Congestion Work Zone Drainage Effectiveness Work Zone Clay Content Work Zone Land Slope Work Area Quantity Soil Condition Length of Work Area Type of Lime Used Work Item Level Thickness Width of Work Area Course Type Main Lane vs. Non-main Lane Type of Concrete Pavement Configuration X X X X Excavation X X X X X X X X X LimeAggregate Base Embankmant Treatment Sub- (Flexible Base grade and CTB) X X X X X X X X X X X X X X X X X (Lift) X X X X X X X X X (Lift-Length) X X X X X X X X X X X X X X X X X X X X X Hot Mix Asphalt Pavement X X X X X X X X Slip-from Concrete Pavement X X X X X X X X Conventional Form Concrete Pavement X X X X X X X X 86 Table 4.2 Work Item Level Candidate Drivers and Data Attributes Excavation (Unit: CY/Crew Day) Work Area Quantity Soil Condition (Numerical: CY) Loose Stiff Rocky Embankment (Unit: CY/Crew Day) Work Area Quantity Soil Condition (Numerical: CY) Loose Stiff Rocky Lime-treated Sub-grade (Unit: SY/Crew Day) Work Area Quantity Length of Work Area Type of Lime Used Lift-Thickness Soil Condition (Numerical: SY) (Numerical: LF) Type C Lime (Numerical: INCH) Loose Stiff Rocky Others Aggregate Base Course (Unit: Lift-SY/Crew Day) Work Area Quantity Lift-Length of Work Area Width of Work Area Lift-Thickness (Numerical: Lift-SY) (Numerical: LF) (Numerical: LF) (Numerical: INCH) Hot Mix Asphaltic Concrete Pavement (Unit: Ton/Crew Day) Work Area Quantity Course Type Main Lane vs. Non-main Lane (Numerical: TON) Base Main Lane Surface Non-main Lane Slipform Concrete Pavement (Unit: SY/Crew Day) Type of Concrete Pavement Work Area Quantity Length of Work Area Width of Work Area Thickness CRCP (Numerical: SY) (Numerical: LF) (Numerical: LF) (Numerical: INCH) JCP NRCP Conventional Form Concrete Pavement (Unit: SY/Crew Day) Work Area Quantity Configuration (Curve or Sharp Angle) Thickness (Numerical: SY) None (Numerical: INCH) Yes 87 4.3 DATA COLLECTION TOOLS A package of data collection tools consisting of project-level data collection, Work Zone-level and Work Item-level data collection and Work Items sheets was developed. 4.3.1 Project-Level Data Collection Tool The project-level data collection tool, shown in Appendix C, consists of three major parts. The first section documents general project information including the name of the road, station range, prime general contractor, project duration, percentage of completion, as well as city and county. The second section helps identify Work Items from the TxDOT site personnel according to their planned schedule. Such information in this section allows data collectors to pin-point their Work Items and to reflect on projects in which they would be interested. The last section was used by data collectors to evaluate the characteristics of the selected projects. The information collected for this section includes project type, location, traffic flow, traffic account, annual precipitation, winter season length, percentage of completion, contract amount, technical complexity, contract day, accelerated construction provision, liquidated damages, soil types, clay content, land slope, water table depth below grade, scheduling technique used, days per week, hours per day, contract administration system and contractor's management skill. 88 4.3.2 Work Zone-level and Work Item-level The Work Zone- and Work Item-level data collection tool is shown in Appendix D. 4.3.2.1 Work Zone-Level Six Work Zone (WZ)-level factors were identified as the possible factors influencing the Production Rates of highway construction: accessibility, congestion, drainage effectiveness, clay content, land slope and water table depth. The measurements of these six factors are discussed in the following sections. Due to the complexity of construction task, the Work Zone defined for each type of construction tasks may vary in terms of its physical outline, and thus the six factors were measured in different ways. 4.3.2.1.1 WZ Accessibility Work Zone accessibility in this study was characterized in one of three ways: difficult, moderate and easy. According to the Candidate Drivers selected in Section 4.1, Work Zone accessibility influences the Production Rates of Excavation, Embankment, Hot mix asphalt pavement, and Concrete pavement. For Excavation and Embankment, rolling resistance, grade resistance, and haul road distance can influence the travel time of hauling materials (Simon 1999; Peurifoy et al. 2002) Therefore, the different levels of Work Zone accessibility are defined as follow: 89 Difficult Haul distance is greater than ten miles, or Haul distance is less than ten miles but greater than five miles and the access road has high total resistance Moderate Haul distance is less than ten miles but greater than five miles and the access road has low total resistance, or Haul distance is less than five mile but greater than one mile, and the access road has high total resistance Easy Haul distance is greater than one mile but less than five miles and the access road has low total resistance, or Haul distance is less than one mile For Hot mix asphalt pavement and Concrete pavement, the Work Zone accessibility measurement was based on the distance between the Hot mix asphalt plant or concrete batch plant and the Work Area and the ease of accessing the Work Area by the transporting trucks. Difficult Haul distance is greater than ten miles, or Access road is not well constructed Moderate Haul distance is less than ten miles but greater than five miles, or Access road is not well maintained Easy Haul distance is less than five miles, and Access road is well maintained 90 4.3.2.1.2 WZ Congestion Ovararin and Popescu (2001) defined Work Zone congestion as the frequency of additional crews working in the same Work Area. In this study, working procedures are very different among the selected seven Work Items, so it was necessary to separate the definition of the Work Zone congestion for each Work Item. According to the Candidate Drivers selected in Section 4.1, Work Zone congestion influenced the Production Rates of all seven targeted Work Items. For Excavation, Work Zone congestion refers to the space allowed for the truck queue when loading and the space allowed for the excavators to perform Excavation and loading. Severe There is no other space in the Work Zone for the truck queue waiting to load, and There is limited space for loaders to load trucks Moderate There is free space for loaders but there is limited space in the Work Zone for truck queue waiting for loading Minor There is free space for loaders and truck queue For Embankment, Work Zone congestion refers to the space allowed in the Work Zone for unloading, spreading, and compacting. Severe Work Zone allows only one of tree different tasks (Dumping, Spreading, or Compacting) at a time Moderate Work Zone area allows only two different tasks 91 simultaneously Minor Work Zone allows three tasks simultaneously For Lime-treated sub-grade, Work Zone congestion refers to the space allowed in the Work Zone for mixer, motor grader and compactor to work simultaneously. Severe Moderate Only one piece of equipment can be operated each time Two out of three pieces of equipment can be operated simultaneously Minor Mixer, motor grader, and compactor can work simultaneously For Aggregate base course, Work Zone congestion refers to the space allowed in the Work Zone for the motor grader and compactors to work simultaneously. Severe Moderate Only one piece of equipment can be operated at a time Two out of three pieces of equipment can be operated simultaneously Minor More than three pieces of equipment can work simultaneously For Hot mix asphalt (HMA) pavement, Work Zone congestion refers to the space allowed in the Work Zone for unloading the truck and the waiting truck queue. Severe Work Zone area is adjacent to heavy traffic and has limited 92 space for unloading Hot mix asphalt Moderate Work Zone area is not adjacent to heavy traffic but has limited space for unloading Hot mix asphalt or waiting trucks Minor Work Zone area has enough space for unloading HMA and truck queue For Slip-form concrete pavement, Work Zone congestion refers to the space allowed in the Work Zone for installing rebar and unloading concrete. Severe Work Zone area is adjacent to heavy traffic and has limited space for installing rebar and unloading concrete Moderate Work Zone area is not adjacent to heavy traffic but has limited space for unloading concrete or for a truck queue Minor Work Zone area has enough space for installing rebar, unloading concrete and for the truck queue For Conventional form concrete pavement, Work Zone congestion refers to the space allowed in the Work Zone for installing rebar and unloading concrete. Severe Work Zone area is adjacent to heavy traffic and has limited space for unloading concrete and eight rebar workers Moderate Work Zone area is not adjacent to heavy traffic but has limited space for unloading concrete or eight rebar workers Minor Work Zone area has enough space for unloading concrete and eight rebar workers 93 4.3.2.1.3 WZ Drainage Effectiveness Work Zone drainage effectiveness is a measurement of the frequency of flooding after rain. According to the Candidate Drivers selected in Section 4.1, this Candidate Driver only influenced the Production Rates of Excavation and Embankment Operations. This Candidate Driver was based on TxDOT site personnel's judgment since they were in charge of assessing the site condition when it rained. Easily Flooded Moderate Quickly Drains Frequently floods after rain Sometimes floods after a heavy rain Rarely floods after rain 4.3.2.1.4 WZ Clay Content According to the Candidate Drivers selected in Section 4.1, only the Production Rates of Excavation, Embankment and Lime-treated sub-grade were influenced by clay content. judgment of site personnel. High Moderate Low Soil becomes very sticky after rainfall Soil becomes somewhat sticky after rainfall Soil does not become sticky after rainfall The clay content was evaluated based on the 94 4.3.2.1.5 WZ Land Slope Land slope affects the Production Rates of all seven targeted Work Items. The land slope of the Work Zone was determined during site visits. Steep Moderate The slope of the Work Zone is greater than 15 The slope of the Work Zone is greater than 5 but less than 15 Flat The slope of Work Zone is less than 5 4.3.2.1.6 WZ Water Table Depth Excavation and Embankment Operations are affected by the Water table depth. >10' The Water table depth was measured by TxDOT site personnel. The typical water table is more than 10' below the original ground level 4'~10' The typical water table is between 4' and 10' below the original ground level <4' The typical water table is less than 4' below the original ground level 4.3.2.2 Work Item Level The Work Item-level data collection tool was used to document the completed quantity of work, crew information, equipment information, total working days, 95 as well as disruptions. A tracking calendar that was made a part of the Work Item-level data collection tool was created to track the information. 4.3.3 Work Item Sheets Lack of standardization for measuring productivity is an obstacle for the comparison of construction productivity between projects (Borcherding and Alarcon 1991). A consistent data collection technology is required to study productivity in the construction industry (Sanders and Thomas 1991). A Work Item sheet (Appendix E) was developed to guide data collectors to consistently document Production Rates. Each Work Item sheet includes item number, Work Item description, measured unit, scope of measurement, specific factors and crew definition for each Work Item. The scope of each Work Item The start node, end was determined by the research team and PMC members. node, in-scope activities, and out-scope activities were listed on each Work Item sheet. Notes on Work Item-specific Candidate Drivers were listed on the Work Item sheet to remind data collectors during the data collection process as well. 4.4 PILOT DATA COLLECTION As the data collection tools were developed and the process was established, a pilot data collection effort was conducted to test if the data collection tools and planned process were effective. A TxDOT district was selected in which to 96 perform the pilot data collection. levels of complexity were selected. Three construction projects with different At the beginning of the pilot data collection, the research team concentrated their efforts on a simple project to collect Production Rate data. Two months after working on the first project, two construction projects that were more complicated were studied. A total of twelve data points were collected in the The efficiency of the data first district from late February to late July of 2002. collection was low because only one or two projects were studied concurrently and the targeted Work Items in the on-going projects were not performed according to the planned schedule. The data collection process then was adjusted to collect data on more than three construction projects simultaneously. Despite these difficulties encountered during the pilot data collection, the data collection tools were refined in such a way as to be more complete and efficient. 4.5 DATA COLLECTION Project-level information was collected during the project meetings. Other information was collected at regular job visits. The scope of each targeted Work Item is presented in this section. 4.5.1 Excavation Table 4.3 presents the scope of Excavation employed for data collection. The scope of Excavation starts with removing the soil or excavating for a 97 construction phase and ends when the elevation of the sub-grade or the working phase is reached. The activities included in the scope are removing the top soil, excavating from the original elevation to the elevation of the sub-grade, loading excavated materials, and disposal of materials. The rock Excavation Operation was excluded. Table 4.3 Scope of Excavation for Data Collection SCOPE - Included - Not Included Survey & Layout Access road construction and maintenance Unsuitable material replacement Reshaped by blade and then sprinkled and rolled for sub-grade surface (about 6" depth) Temporary drainage maintenance Shaping slop Rock Removing top soil Excavation from original elevation to the elevation which is at least 6" below the required sub-grade elevation Disposal of material Starting NODE - - Remove top soil. Starting the excavation of any working phase. Sub-grade surface is completed. Reach the anticipated elevation of the working phase Ending - In the construction industry, tracking the completed quantity in an Excavation Operation is a cumbersome task for clients and contractors. Contractors usually prefer to claim the completed quantity as large as possible to obtain higher payment on a unit price contract. In contrast, clients tend to pay for a completed Therefore, an quantity as low as possible to reduce the risk of overpayment. 98 agreement between clients and contractors on the methods of tracking the excavated quantity should be made at the start of a construction project. There are three methods of tracking the completed quantity for an Excavation Operation employed by TxDOT. First, the general contractor proposes a quantity report according to the quantity calculated from the numbers of loaded trucks, trailers, or scrapers. Based on the proposed quantity, TxDOT will This method approve or adjust the quantity depending on the planned quantity. is mostly applied to projects with a large amount of Excavation at a certain Work Area. Secondly, the general contractor proposes a quantity report based on the quantity calculated from the planned quantity. TxDOT will review the quantity report and then adjust or approve the quantity report. Thirdly, the general contractor and TxDOT evaluate the percentage of completion together and then calculate the completed quantity by multiplying the planned quantity and the percentage of completion. The completed quantity in the study was collected from TxDOT's approved quantity. The standard resource used in the Excavation Operation is one excavator with a 2 cubic yardage bucket and trucks. 4.5.2 Embankment The scope of the Embankment Operation, shown in Table 4.4, starts from placing the first load of material in a working phase and ends by reaching the planned elevation. This is somewhat different than the scope described in TxDOT's construction specification, which starts from removing the top soil and 99 ends by reaching the elevation of the sub-grade or the elevation instructed by the engineers. It is difficult to measure the Production Rate starting from the Since the removal of the top soil until it reaches the elevation of the sub-grade. sources of materials used in Embankment for a certain Work Area could be obtained from varied Work Areas or projects, the Embankment Operation is usually divided into multiple phases. The time intervals between two successive phases are varied and not predicable. The standard resource used in the Embankment Operation is one or two dozers and a compactor. Table 4.4 Scope of Embankment for Data Collection SCOPE Included - Not Included Survey & Layout Constructing access road Temporary drainage maintenance (Construction of roadway embankments, levees and dykes or any designated section of the roadway) - Placing materials - Spread material - Sprinkling - Compaction Starting NODE - Place the first load of embankment material. Sub-grade surface is completed. . Reach the elevation of the working phase if there are more than one phases of embankment Ending - 4.5.3 Lime-Treated Sub-grade Lime-treated sub-grade is a common method of stabilizing the sub-grade in Texas due to expansive soil. Adding lime to the soil can make it more flexible 100 and thus reduces the possibility of cracking in the surface of sub-grade. Another advantage is to prevent the intrusion of water into the sub-grade. When the elevation of a sub-grade has been reached and compaction and grading have been completed, slurry or dry lime is placed on the sub-grade. A cutting and pulverizing machine is used to cut the sub-grade uniformly to a proper depth, usually six inches, and then the cutting material is mixed with the lime. After the mixing process, a motor grader, a sheep-foot roller and a steel roller are used to compact and seal the sub-grade. mixing". After the first mixing, the sub-grade is left to cure for one to four days depending on the decision of the TxDOT engineers. If a "Type C" lime is used, The above processes are called the "first the sub-grade needs two to seven days for curing. The typical duration of the first curing in the observations is two days. After curing, the sub-grade is mixed again. This is the second mixing. This time the sub-grade is shaped to the Following the second mixing, the sub- required grade after mixing is completed. grade is cured again and then the Aggregate base course or pavement structure is placed on top of the sub-grade (TxDOT 1993). The scope of Lime-treated sub-grade, shown in Table 4.5, starts with spreading lime for the first mixing and ends with finishing sub-grade surface. The second curing is excluded in the scope because of the high variation in its duration. The second curing varies from one day to fourteen days depending on 101 the engineers' instruction or the contractor's working plan. The standard resource used in the Lime-treated sub-grade Operation is a mixer, a motor grader, a sheep-foot roller, a steel roller, one or two spreaders and a water truck. Table 4.5 Scope of Lime-Treated Sub-grade for Data Collection SCOPE - Included - Not Included Survey & layout Equipment move in Transport material Curing (after finishing) Density tests Setup blue top Cutting & pulverizing Spread Lime Mixing Sprinkling or aerating Compaction Finishing 1ST curing and 2nd mixing Starting NODE Ending - Spread lime or cut & pulverize sub-grade. Finishing sub-grade surface is completed. 4.5.4 Aggregate Base Course Two types of Base course were observed in this study. The first type is the Flexible base course. Such a Base course usually contains multiple lifts. observed thickness of a lift varies from three to six inches. starts from hauling the flexible base material to the job site. shaping is usually performed on the same day of hauling. The This Work Item Spreading and Following that, Sometimes, processing, which includes compacting and finishing, is performed. the timing between processing and hauling is long because contractors wait for 102 the sub-grade of another section to be complete so they can process the two sections together. After a Flexible base course is completed, it will be cured for about two days, as suggested by the TxDOT inspectors, before the surfacing will be placed on the completed Base. The second type of Aggregate base course is the Cement-treated base (CTB) course. mixing. Two mixing methods are applied to the CTB: plant mixing and road No data for road mixing CTB was collected for this study, therefore only the CTB for plant mixing will be studied. In this type of Operation, the CTB material is delivered from the plant to the job site and then is placed and spread on the top of the sub-grade. Following that, the compaction is completed within two hours for each lift as there is limited duration on processing cement mixed material because of the interactions between the cement and water. If the CTB has multiple lifts, the compaction of all lifts must be completed within five hours (TxDOT 1993). The CTB Operation observed in this study was constructed in a single lift which had a uniform thickness of six inches. After the CTB was completed, it usually requires at least seventy two hours for curing. The scope of the Aggregate base Operation, shown in Table 4.6, starts from delivering Base materials to the job site and ends at the completion of the Base course. Curing, material tests, and density tests were excluded in the scope. The completed quantity of this Work Item was the area in which the contractors process an Aggregate base Operation. The unit measured in the 103 study was Lift-SY/Crew Day. It refers to the number of square yards of a Base The standard resource used lift that is completed using a standard size of crew. in the Aggregate base Operation is a motor grader, one or two rollers and a water truck. Table 4.6 Scope of Aggregate Base for Data Collection SCOPE - Included - Not Included Survey & layout Shaping the sub-grade or existing roadbed Stockpiled All material tests excluded Curing (Flexible Base: Directed by Engineers, usually 2 days; CTB: 72 hours) Density tests Rework caused by failing to achieve required density Placing materials Spread uniformly & shaping Blade & shaping Sprinkling Compaction Dry-out (if required) Starting NODE Ending - Place the first load of base material. Finishing a lift of base course is completed. 4.5.5 Hot Mix Asphalt Pavement Hot mix asphalt pavement (HMA) includes the construction of the HMA base and surface. Two types of HMA base were included in the scope. The first type is the HMA base constructed for use under Concrete pavement and the other type is constructed as the base course of the HMA surface. 104 The scope of an HMA Operation, shown in Table 4.7, includes transporting HMA materials, setting up the lay-down machine, placing HMA and compaction. The completed quantity is measured as tonnages of HMA placed on a targeted Work Area by a standard resource. The standard resource used in the Hot mix asphalt pavement is a lay-down machine, a pneumatic roller, a steel roller, and a finishing crew consisting of six to eight workers. Table 4.7 Scope of Hot Mix Asphalt Pavement for Data Collection SCOPE - Included - Not Included Transport materials Cleaning surface before applying for tack coat Shoot tack coat (if tack coat required) Survey and layout Mixing materials in the plant Equipment setup Lay Hot Mix Asphalt Compaction (Roller or lightly oiled tamps) Starting NODE Ending - Place the first load of Hot Mix Asphalt material. Complete compaction. 4.5.6 Slip-form Concrete Pavement Two types of Slip-form concrete pavement Operations were observed in the study period. They include continuously reinforced concrete pavement (CRCP) and jointed concrete pavement (JCP). The difference between these two types is 105 the usage of reinforced steel in the Concrete pavement. In this study, JCP was excluded because it is rarely used by TxDOT. The scope of Slip-form concrete pavement, shown in Table 4.8, starts at the setting of the string line and ends at the finishing of the surface of Concrete pavement. For CRCP, it was found that there was often a long time after rebar installation and before concrete placement. This is because contractors can achieve better production by reducing the number of times for setting up the slipform paver. The standard resource used in the Slip-form concrete pavement Operation is a slip-form paver, material-transfer equipment and a rebar crew consisting of eight to ten workers and a concrete crew consisting of six to eight workers. Table 4.8 Scope of Slip-form Concrete Pavement for Data Collection SCOPE - Included - Not Included Survey & Layout Surface preparation Equipments move in Ride quality test Core test Unloading reinforcing steel Curing Saw cutting Setting string line Placing dowels Installing reinforcing steel Placing joint assemblies Initial equipment setup Placing concrete Finishing Starting NODE Ending - Set string line. Complete concrete placement. 106 4.5.7 Conventional Form Concrete Pavement Table 4.9 presents the scope of Conventional form concrete pavement employed for the data collection of this study. Conventional form concrete pavement starts with setting up formwork and ends at the completion of concrete placement. The standard resource used in Conventional form concrete pavement is a formwork crew with four to six workers, a rebar crew with six to ten workers, and a concrete crew with six to ten workers. Table 4.9 Scope of Conventional Form Concrete Pavement for Data Collection SCOPE - Included - Not Included Survey & Layout Surface preparation Cutting & bending Reinforcing steel Core test Curing Removing formwork Formwork Installing reinforcing steel Placing concrete Spread and finishing Starting NODE Ending - Start to setup formwork Complete concrete placement. 4.6 SUMMARY OF STUDY DISTRICTS AND PROJECTS A total of thirty-five on-going projects from seven districts were investigated for this study. The visited districts and their respective number of visited Among the seven districts, three were located projects are listed in Table 4.10. in Central and South Texas, two were in the coastal region, and one each in North 107 Texas and the Panhandle and West Texas. General project information for the study projects is presented in Appendix G. Production Rate data in this study were collected from early March, 2002 to late March, 2004. A total of 196 data points were collected. Table 4.10 Dates of Collecting Data, Number of Investigated Projects and Number of Observed Data by Visited Districts Total number of Projects 4 2 4 7 9 2 7 35 No. of Observed Data Total Number of Total Data Points Work Items Visited Districts D1 D2 D3 D4 D5 D6 D7 Area Dates of Collecting Data Central and South Texas Coastal Texas Central and South Texas North Texas Coastal Texas Panhandle and West Texas Central and South Texas Total 3/1/02 ~ 7/31/02 7/1/02 ~ 9/1/02 9/1/02 ~ 2/10/03 11/7/02 ~ 2/25/03 3/20/03 ~ 11/1/03 9/16/03 ~ 11/1/03 11/15/03~3/31/04 5 2 5 7 7 3 6 13 13 34 33 68 10 25 196 Twenty-two different contractors built these projects, as shown in Table 4.11. The average ratio of observed projects versus prime contractors is 1.5. Table 4.12 displays the number of projects according to contract amount. Most of the projects had a contract amount of less than thirty million dollars. projects had a contract amount of more than one hundred million. Only two 108 Table 4.11 Number of Projects by Prime Contractor I.D. Contractors GC1 GC2 GC3 GC4 GC5 GC6 GC7 GC8 GC9 GC10 GC11 Number of Projects 1 1 1 2 2 2 1 1 1 1 1 Contractors GC12 GC13 GC14 GC15 GC16 GC17 GC18 GC19 GC20 GC21 GC22 Number of Projects 1 1 2 1 1 1 1 2 5 1 5 Total 35 Table 4.12 Number of Projects by Contract Amount Contract Amount (Million) 0 ~ 9.99 10 ~ 19.99 20 ~ 29.99 30 ~ 39.99 40 ~ 49.99 50 ~ 59.99 60 ~ 69.99 70 ~ 79.99 80 ~ 89.99 90 ~ 99.99 >100 Number of Projects 14 8 5 1 1 1 0 1 2 0 2 Total 35 4.7 SUMMARY OF PRODUCTION RATE DATA A tabular summary of data observations for each Work Item is listed in Table 4.13. Production Rates for each Work Item were collected at a wide variety of 109 projects in several districts. Several different prime contractors were observed, so biases due to region and contractor would be limited. Table 4.13 Sources for Data and Observed Quantity for Seven Work Items Work Item Excavation Embankment Lime-treated Sub-grade Aggregate Base Course Hot Mix Asphalt Pavement Slip-form Concrete Pavement Conventional Form Concrete Pavement Number of Data Points 26 34 32 29 32 23 20 Number of Districts 5 5 6 6 6 3 4 Number of Projects 12 16 18 15 19 10 8 Number of Prime Total of Observed Contractors Quantity 10 12 12 13 14 6 5 154,570 CY 237,415 CY 317,235 SY 414,826 SY-LIFT 61,152 TON 169,357 SY 21,889 SY 110 CHAPTER V: DESCRIPTIVE STATISTICS OF OBSERVED PRODUCTION RATES A survey conducted in January 2003 indicated that most TxDOT districts considered the Production Rates of CTDS to be unrealistic and in need of revision. In addition, most members of the PMC believed that the Production Box plots Rates for Earthwork and Pavement construction were too optimistic. are used herein to compare the Production Rates of the observed data, CTDS, and historical records. Table 5.1 displays the range as well as the mean of the observed Production Rates for each Work Item. Table 5.1 Range Data for Seven Work Items Work Item Excavation Embankment Lime-treated Sub-grade Aggregate Base Course Hot Mix Asphalt Pavement Slip-form Concrete Pavement Conventioinal Form Concrete Pavement Unit CY/Crew Day CY/Crew Day SY/Crew Day SY-Lift/Crew Day TON/Crew Day SY/Crew Day SY/Crew Day Minimum 199 249 82 526 158 462 30 Mean 1163 1097 1563 3398 817 1253 306 Maximum 3558 3000 3722 6500 1460 2154 582 Other existing Production Rate sources such as the Caterpillar Performance Handbook which documents the Production Rates of each piece of caterpillar equipment are not applicable for comparison. This study was more concerned 111 about the Production Rates of Operations, which usually combine the work of several pieces of equipment and/or labors, rather than the Production Rates of a single machine. 5.1 Excavation Figure 5.1 displays data summaries for comparison of Production Rates from different sources of Excavation. Production Rate data were observed directly from twelve projects in five districts in Texas. The average observed Production Rate was 1,163 CY/Crew Day, which is much slower than the average CTDS rate, 3,400 CY/Crew Day, but faster than the Production Rates found in most of the historical records. Almost no information is available to further explore the causes of lower Excavation Production Rates for historical data. Data Source No. Districts 5 No. Projects 12 No. Data Points 26 No. CY Observations 154,570 CTDS N/A N/A N/A N/A As-Built 1 1 (D3) 1 129 175,556 As-Built 2 1 (D3) 1 456 327,394 As-Built 3 1 (D4) 1 23 4,910 As-Built 4 1 (D4) 1 250 87,832 As-Built Projects 1 (D8) 11 126 89,819 Unit: CY/Crew Day 0 2000 4000 6000 8000 Figure 5.1 Comparison of Excavation Production Rates from Different Sources 112 5.2 Embankment Figure 5.2 presents the distribution of Production Rates among different sources for Embankment. Production Rate data were collected from sixteen projects in five districts in Texas by direct observation. The average observed Production Rate was 1,097 CY/Crew Day, which is much slower than the average CTDS rate of 3,500 CY/Crew Day. The observed Production Rates were close to the Production Rates in As-built 2 and As-built 4 and faster than other as-built rates. Data Source No. Districts 5 No. Projects 16 No. Data Points 34 No. CY Observations 237,415 CTDS N/A N/A N/A N/A As-Built 1 1 (D3) 1 22 13,683 As-Built 2 1 (D3) 1 543 688,520 As-Built 3 1 (D4) 1 16 4,272 As-Built 4 1 (D4) 1 197 190,436 As-Built Projects 1 (D8) 10 125 52,285 Unit: CY/Crew Day 0 2000 4000 6000 8000 Figure 5.2 Comparison of Embankment Production Rates from Different Sources 113 5.3 Lime-Treated Sub-grade Figure 5.3 shows the distribution of Production Rates obtained from different sources for Lime-treated sub-grade. Thirty-two data points were observed from eighteen projects in six districts. The work scope included in the Production Rates for CTDS and RS Means is different than that for this study and As-built rates. For CTDS and RS Means, the first curing is excluded; however first Therefore, for curing is included in the rates for this study and As-built rates. comparison purposes only, observed Production Rates were computed excluding the first curing and plotted in Fig 5.3. Data Source Observations (1st Curing included) No. Districts 6 No. Projects 18 No. Data Points 32 No. SY 317,235 Observations (1st Curing excluded) 6 18 32 317,235 CTDS (1st Curing excluded) N/A N/A N/A N/A As-Built 1 1 (D3) 1 5 19,125 As-Built 2 1 (D3) 1 0 0 As-Built 3 1 (D4) 1 4 6,552 As-Built 4 1 (D4) 1 20 53,250 Unit: SY/Crew Day 0 2000 4000 6000 8000 10000 Figure 5.3 Comparison of Lime-Treated Sub-grade Production Rates from Different Sources 114 The average CTDS Production Rate is 4,000 SY/Crew Day. The average observed Production Rate, including the first curing, was 1,563 SY/Crew Day, and 2,348 SY/Crew Day excluding the first curing. 5.4 Aggregate Base Course Figure 5.4 presents a comparison of Production Rates collected from this study and those retrieved from the CTDS for Aggregate base course. No historical records were used for this comparison because it was difficult to identify the working days for processing Aggregate base from such records. Due to the different working process between Cement-Treated Base (CTB) and Flexible base, the observed Production Rates were separated into two groups for better comparison. Fourteen data points from six projects in one district were collected for CTB, with a total quantity of 157,308 LIFT-SY. For Flexible base course, fifteen data points from nine projects in five districts were collected with a total of 257,518 LIFT-SY. The average observed Production Rate was 4,050 LIFT-SY/Crew The Day for CTB and 2,788 LIFT-SY/Crew Day for Flexible base course. average CTDS Production Rate for both CTB and Flexible base is 3,000 LIFTSY/Crew Day. Therefore, the CTDS Production Rate of CTB is much lower than the average observed rate. In contrast, the average CTDS Production Rate for Flexible base is higher than that from field observations. 115 Data Source Observations (Cem. T. Base) No. Districts 1 No. Projects 6 No. Data Points 14 No. SY-LIFT 157,308 Observations (Flex. Base) 5 9 15 257,518 CTDS (Cem. T. Base) N/A N/A N/A N/A CTDS (Flex. Base) N/A N/A N/A N/A Unit: SY-LIFT/Crew Day 0 1000 2000 3000 4000 5000 6000 7000 Figure 5.4 Comparison of Aggregate Base Course Production Rates from Different Sources 5.5 Hot Mix Asphalt Pavement Figure 5.5 presents side-by-side box plots of Production Rates of Hot mix asphalt (HMA) pavement collected from different sources. Thirty-two data points were observed for Hot mix asphalt pavement. Nineteen projects in six districts were investigated with a total of 61,152 tons of HMA placed. The average CTDS Production Rate for HMA pavement is 1,200 Tons/Crew Day. The average observed Production Rate fell between the average CTDS rate and the average rates from historical records. 116 Data Source No. Districts 6 No. Projects 19 No. Data Points 32 No. Tons Observations 61,152 CTDS N/A N/A N/A N/A As-Built 1 1 (D3) 1 48 23,457 As-Built 2 1 (D3) 1 135 37,271 Unit: Tons/Crew Day 0 400 800 1200 1600 2000 2400 Figure 5.5 Comparison of Hot Mix Asphalt Pavement Production Rates from Different Sources 5.6 Slip-form Concrete Pavement Figure 5.6 displays side-by-side box plots of Production Rates from observations, as-built projects, and CTDS. The Production Rates of twentythree Work Zones from ten projects in three districts were investigated. Of those data points, three were Jointed Concrete Pavement (JCP). The others were for Continuously Reinforced Concrete Pavement (CRCP). A total of 161,133 square yards of CRCP Operations and 8,224 square yards of JCP were observed. The Production Rate of CRCP is much lower than JCP because JCP does not involve time-intensive rebar work. 117 The average Production Rate of the observed CRCP Operations was 1,357 SY/Crew Day, which was faster than the average Production Rate of the As-built project 4 but much slower than the average CTDS Production Rate of 3,000 SY/Crew Day. The observed Production Rate of JCP Operations was 1,729 SY/Crew Day, which is relatively close to the average rate of the As-built project 3. Data Source No. Districts No. Projects No. Data Points No. SY Observations (JCP) 1 1 3 8,224 Observations (CRCP) 3 9 20 161,133 CTDS N/A N/A N/A N/A As-Built 3 (JCP) 1 (D4) 1 19 22,617 As-Built 4 (CRCP) 1 (D4) 1 1 342 Unit: SY/Crew Day 0 1000 2000 3000 4000 5000 6000 Figure 5.6 Comparison of Slip-form Concrete Pavement Production Rates from Different Sources 5.7 Conventional Form Concrete Pavement Figure 5.7 displays observed Production Rates and three as-built projects for Conventional from concrete pavement. 118 A total of 20,944 square yards of Concrete pavement construction were observed. The average observed Production Rate was 338 SY/Crew Day. The range of observed Production Rates is similar to the range obtained from the As-built projects. No. Districts 3 No. Projects 8 No. Data Points 20 No. SY Data Source Observations 21,889 As-Built 5 1 (D8) 1 7 3,877 As-Built 6 1 (D8) 1 14 4,869 As-Built Projects 1 (D8) 7 9 1,745 Unit: SY/Crew Day 0 200 400 600 800 1000 1200 Figure 5.7 Comparison of Conventional Form Concrete Pavement Production Rates from Different Sources 119 CHAPTER VI: DATA ANALYSIS AND HYPOTHESIS TESTING FOR EARTHWORK-RELATED WORK ITEMS The ANOVA/t test was employed to test if there were different mean Production Rates within groups for drivers involving categorical or discrete numerical data. A non-linear or linear regression analysis was employed to This was explore the relationships between Production Rates and drivers. applied for drivers involving continuously numerical data. The logarithmic model and the power model were used to identify non-linear relationships of Production Rates and drivers. For each factor analysis, the R-squares and the adjusted R-square of the linear, logarithmic, and power models were examined. The assumptions of each model were examined in order to determine the best model for studying the effects of the driver on the Production Rates of the targeted Work Item. When the regression analysis was employed to analyze the effects of drivers on Production Rates, outliers of each variable were identified and removed from the data set (shown in Appendix H). A linear or non-linear model was employed to study the effects of drivers and then the violations of the assumptions of the regression analysis were checked. 120 6.1 TEST DIFFERENCE IN MEAN PRODUCTION RATES Mean observed Production Rates were compared with average CTDS rates to test the first hypothesis which is presented as follows: Hypothesis 1: The Production Rates of the CTDS are not realistic. The established null hypothesis is that the mean observed Production Rates are equivalent to the average CTDS rates for major Work Items of Earthwork construction. In other words, the established alternative hypothesis is that average CTDS Production Rates are different from mean observed Production Rates. The results of the hypothesis testing are displayed in Table 6.1. Except for Flexible base, other Work Items show that there is a significant difference between the average rate of the CTDS and the mean observed Production Rates. As clearly indicated, the Production Rates of the CTDS are too optimistic for Excavation, Embankment and Lime-treated sub-grade. The average Production Rate of the CTDS for Cement-treated base is slower than the mean observed Production Rate, however. 121 Table 6.1 Average Production Rates of the CTDS and Mean Observed Production Rates Work Item Excavation Embankment Lime-Treated Sub-grade Flexible Base Course Cement Treated Base Number of Data Points 26 34 32 15 14 Unit CY/Crew Day CY/Crew Day SY/Crew Day SY-Lift/Crew Day SY-Lift/Crew Day Mean Observed Production Rate 1163 1097 2348 2788 4050 Average CTDS Mean Rate Difference 3400 3500 4000 3000 3000 -2237 -2403 -1652 -212 1050 P-Value *0.000 *0.000 *0.000 0.583 *0.013 * indicates that P-value is less than 0.05 and thus, the null hypothesis (Mean Observed Production Rate = Average CTDS Rate) is rejected at 95% confidence interval. 6.2 ANALYSIS OF DRIVERS OF PRODUCTION RATES Several Candidate Drivers were selected in Section 4.1 and further investigated for their effects on Production Rates. The following is the formulation of the second hypothesis. Hypothesis 2: The Production Rates of the targeted Work Items are driven by some productivity factors that are known at the design stage. 6.2.1 Excavation Scatter plots, shown in Appendix I-1, were used to examine the relationship between twelve Candidate Drivers and Excavation Production Rates. All but Work Area Quantity were excluded from further driver analysis as the observed Production Rates did not exhibit any specific relationship to the variability of Candidate Drivers. Figure 6.1 displays the scatter plot of observed Production Rates versus Work Area Quantity. A possible non-linear or linear relationship 122 of Work Area Quantity and Production Rates can be observed from the scatter plot. The sub-hypotheses of the second hypothesis, listed as follows, were further tested for Excavation. Sub-hypothesis 1: Operations and resource allocation will be more efficient when the amount of repetitive work in a Work Area is great. More resources may be used to amplify daily This may be true for Excavation Production Rates. because its Operations are highly repetitive. In addition, productive hours in a working day may be higher for the Work Area with greater quantity. A Production Rate (CY/Crew Day) 3000 A A 2000 A A A A AA A 1000 A A A A A A A A A A A A A A 0 5000 10000 15000 Work Area Quantity (CY) Figure 6.1 Scatter Plot for Excavation: Observed Production Rates (CY/Crew Day) and Work Area Quantity (CY) 123 Excavation: Observed Production Rates and Work Area Quantity The logarithmic model was found to be the most efficient model for the relationship between observed Production Rates and Work Area Quantity for Excavation among three selected models (linear model, logarithmic model and power model). Prior to using the logarithmic model, the box plots of the dependent variable (i.e. the observed Production Rate) and the independent (i.e. variable the logarithmic transformation of Work Area Quantity), shown in Figure 6.2 and Figure 6.3 respectively, were employed for outlier analysis. The seventh data point was found to be an outlier. This outlier was removed before conducting further regression analysis. The fitted logarithmic model, which was used to explore the relationships between the observed Production Rates and the Work Area Quantity for Excavation construction, is shown in Figure 6.4. 4000 7 Production Rate (CY/Crew Day) 3000 2000 1000 0 N=26 Figure 6.2 Excavation: Box Plot of Observed Production Rates (CY/Crew Day) 124 10 Log (Work Area Quantity (CY)) 9 8 7 6 5 N=26 Figure 6.3 Excavation: Box Plot of Logarithmic Transformation of Work Area Quantity (CY) 3000 Production Rate (CY/Crew Day) 2000 1000 0 0 10000 20000 W ork Area Quantity (CY) Figure 6.4 Scatter Plot and Fitted Logarithmic Model for Excavation: Observed Production Rates (CY/Crew Day) vs. Work Area Quantity (CY) 125 This model, shown as Equation 6.1, was statistically significant at the 95% confidence interval. Table 6.2 displays the results of a regression analysis using The R2 and adjusted R2 are 0.692 and 0.678 respectively. the logarithmic model. The coefficients of this model were statistically different from zero at the 95% confidence interval since the P-values were less than 0.05. Production Rate = -3995 + 649 Log (Work Area Quantity) (Equation 6.1) Table 6.2 Logarithmic Model for Excavation: Production Rates (CY/Crew Day) by Work Area Quantity (CY) R2 Adjusted R2 Standard Error Regression Model Variable Work Area Quantity (Constant) F 51.64 B 649 -3995 0.692 0.678 411 P value 0.0000 P value 0.0000 0.0000 Further tests were performed on the fitted logarithmic model to find violations of the assumptions of the regression analysis. The plots used to check for this are displayed in Appendix J. No violation of the assumptions was found as the plots show. Therefore, this model is statistically significant. 126 This fitted model is only applicable to Work Area quantities within the range of 472 CY to 16,798 CY since this model was developed based on observed data within that range. The predicted Production Rate will be negative if the Work Therefore, the estimated Area Quantity is less than 472 CY, using this model. Production Rates of this model can range from 199 CY/Crew Day to 2,319 CY/Crew Day. When the estimated Production Rate is less than 199 CY/Crew Day, the minimum observed Production Rate, 199, may be more reasonable than the predicted value. The effects of Work Area Quantity on the Production Rates of Excavation can be computed by differentiation, as shown in the Equation 6.2, of the fitted logarithmic model. Therefore, if two Work Areas have a quantity of Excavation close to 10,000 CY but different by 1,000CY, they may have a difference of about 65 CY/Crew Day in their average Production Rate. d( Production Rate) = d( Work Area Quantity) Work 649 (Equation 6.2) Area Quantity Contractors tend to use a larger size of loading resources when the excavation quantity for a Work Area increases. The research identifies that the three types of loading machines that were commonly used for the excavation Operations: excavators, loaders and scrapers. Table 6.3 lists the types of loading machines found during the observations and the daily Production Rates of each machine published in the Heavy Construction Cost Data of RS Means 2002. 127 Table 6.3 Daily Production Rates (CY/Day) of Different Loading Machine (Adopted from Heavy Construction Cost Data of RS Means 2002) Loading Machine Scraper (21CY, 1,500' Haul Distance) Scraper (14CY, 1,500' Haul Distance) Scraper (14CY, 3,000' Haul Distance) Scraper (21CY, 5,000' Haul Distance) Scraper (14CY, 5,000' Haul Distance) Excavator (2CY Cap) Excavator (1-1/2CY Cap) Loader (Track, 3CY Cap) Loader (Wheel, 2-1/4CY Cap) Loader (Track, 2-1/2CY Cap) Daily Production Rate (RS Means) 1,030CY/DAY 800CY/DAY 700CY/DAY 650CY/DAY 560CY/DAY 1,040CY/DAY 800CY/DAY 1,040CY/DAY 800CY/DAY 760CY/DAY However, as mentioned before, the resources used in different Operations vary. The daily Production Rate of 1,040 CY/Day using an excavator with 2CY bucket as listed in RS Means 2002 is treated as the standard resource. In order to study the relationship between size of loading resources and Work Area Quantity, the size of loading resources of each observed excavation data point was computed based on the suggested daily Production Rates in Table 6.3. For example, when two 14 CY scrapers with a 3000' haul distance and one 2-1/4 CY wheel loader are used, the size of loading resources is equivalent to 800+2*(700))/1040 = 2.12. Figure 6.5 shows the relationship between observed Work Area Quantity and Size of an employed loading crew. A linear relationship with R2 of 0.56 was found for the Work Area Quantity and the Size of resources. 128 A 15000 A Work Area Quantity (CY) A 10000 A A 5000 A A A A A A A A A A A A A A A A A A A A 0 1.00 1.50 2.00 2.50 Size of an Employed Loader Fleet Figure 6.5 Scatter Plot and Fitted Linear Model for Excavation: Observed Work Area Quantity (CY) vs. Size of an Employed Loader Fleet In addition, the observed Production Rates were standardized by the size of the loading resources for driver analyses. The scatter plots, shown in Appendix I-2, were used to examine the relationship between the twelve Candidate Drivers and the standardized Production Rates of Excavation. relationship was found in those plots. No statistically significant 6.2.2 Embankment The scatter plots, shown in Appendix K, were used to examine the relationships of twelve Candidate Drivers on the Production Rates for Embankment. A non-linear relationship for Work Area Quantity and a 129 difference in mean Production Rates for Work Zone accessibility and Work Zone congestion were observed. The scatter plots of those three Candidate Drivers are shown in Figure 6.6, Figure 6.7 and Figure 6.8. Other Candidate Drivers were excluded from further driver analysis. Three sub-hypotheses, listed as follows, were further tested for Embankment. Sub-hypothesis 1: Based on the fact that repetition usually leads to greater efficiency and better resource allocation, Embankment Operations may be more productive in larger Operations. In addition, productive hours in a working day may be higher for the Work Area with larger quantity. Sub-hypothesis 2: Work Zone congestion is negatively related to Production Rate. This assumption is made as it was observed that less congested Work Zones may allow more pieces of machinery to work simultaneously leading to higher Production Rates. Sub-hypothesis 3: Work Zone accessibility has a significant impact on Production Rate. This is based on the assumption that higher Production Rate is possible if haul distance is short and the haul road condition is good. 130 3000 A Production Rate (CY/Crew Day) A 2000 A A A A A A A A A A A A A A A 1000 A AA A AA A AA AA AA A A A 0 10000 20000 30000 Work Area Quantity (CY) Figure 6.6 Scatter Plot for Embankment: Observed Production Rate (CY/Crew Day) vs. Work Area Quantity (CY) 3000 A Production Rate (CY/Crew Day) A 2000 A A A A A A A A A A A A A A A A A A A A A 1000 A A A A A A A A Easy Moderate Difficult Work Zone Accessibility Figure 6.7 Scatter Plot for Embankment: Observed Production Rate (CY/Crew Day) vs. Work Zone Accessibility 131 3000 A Production Rate (CY/Crew Day) A 2000 A A A A A A A A A A A A A A A A A A A A A A A A A A A A 1000 Minor Moderate Severe Work Zone Congestion Figure 6.8 Scatter Plot for Embankment: Observed Production Rate (CY/Crew Day) vs. Work Zone Congestion 6.2.2.1 Embankment: Observed Production Rates and Work Area Quantity The logarithmic model was found to be the most efficient model for the relationship between observed Production Rates and Work Area Quantity for Embankment. Prior to using the logarithmic model to identify the relationships, the box plots of the dependent variable (i.e. the observed Production Rate) and the independent variable (i.e. the logarithmic transformation of Work Area Quantity), shown in Figure 6.9 and Figure 6.10, were employed for outlier analysis. One data point was found to be an outlier and was removed before conducting further regression analysis. The fitted logarithmic model, which was used to explore the 132 relationships between the observed Production Rates and Work Area Quantity for Embankment, is shown in Figure 6.11. 4000 Production Rate (CY/Crew Day) 3000 25 2000 1000 0 N=34 Figure 6.9 Embankment: Box Plot of Observed Production Rates (CY/Crew Day) 11 Log (Work Area Quantity) 10 9 8 7 6 N=34 Figure 6.10 Embankment: Box Plots of Log (Work Area Quantity (CY)) 133 4000 Production Rate (CY/Crew Day) 3000 2000 1000 0 0 10000 20000 30000 40000 Work Area Quantity (CY) Figure 6.11 Scatter Plot and Fitted Logarithmic Model for Embankment: Observed Production Rates (CY/Crew Day) vs. Work Area Quantity (CY) This model, shown as Equation 6.3, was statistically significant at the 95% confidence interval. Table 6.4 displays the results of the regression analysis The R2 and adjusted R2 are 0.343 and 0.322 using the logarithmic model. respectively. The coefficients of this model were statistically different from zero at the 95% confidence interval since the P-values of testing coefficients for Work Area Quantity and constant were less than 0.05. Production Rate = -1531 + 309 Log (Work Area Quantity) (Equation 6.3) 134 Table 6.4 Logarithmic Model for Embankment: Production Rates (CY/Crew Day) by Work Area Quantity (CY) R2 Adjusted R2 Standard Error Regression Model Variable Work Area Quantity (Constant) F 16.2 B 308.8 -1530.9 0.343 0.322 444.99 P value 0.0003 P value 0.0003 0.0237 Next, tests to catch violations of the assumptions of the regression analysis were performed. The plots are displayed in Appendix L-1. No violation was found, as these plots exhibited. Thus, this model is statistically significant. This model is only applicable to Work Area quantities within the range of 1,064 CY to 33,938 CY since this model was developed on this range. Therefore, the estimated Production Rates of this model can range from 621 CY/Crew Day to 1,691 CY/Crew Day. The effects of the Work Area Quantity can be computed by the differentiation of the fitted logarithmic model, as shown in Equation 6.4. Therefore, when two Work Areas have the quantity of Embankment Operations close to 10,000 CY but differ by 1,000 CY they may exhibit the difference of about 30 CY/Crew Day in their average Production Rate. d( Production Rate) = d( Work Area Quantity) Work 135 309 (Equation 6.4) Area Quantity 6.2.2.2 Embankment: Observed Production Rates and Work Zone accessibility The number of data points collected for easy, moderate and difficult Work Zone accessibility was respectively 8, 18, and 8. The equality of variances of the three levels was tested by the homogeneity of variance test. Table 6.5 shows the results of the group variances test and the ANOVA test. The P-value of 0.503 indicated that the variances of the observed Production Rates in the three levels are statistically equal. ANOVA was then employed to test the difference The P- in mean Production Rate for the three levels of Work Zone accessibility. value of the ANOVA test was 0.215, which is greater than 0.05. According to the results, it can be concluded that the three levels did not have different mean Production Rates at the 95% confidence interval. With this limited data sample, this suggested that Work Zone accessibility is not a driver of Production Rates for Embankment. In addition, Table 6.6 presents the number of data points and mean Embankment Production Rate for the three levels of Work Zone accessibility. The average Production Rate in the Work Zone with moderate Therefore, it appears that accessibility was higher than with easy accessibility. Work Zone accessibility does not have a huge influence on Embankment. 136 Table 6.5 Results of Group Variances Test and ANOVA Test for Embankment: Work Zone Accessibility Homogeneity of Group Variances Test P value Test Equality of Group Variances ANOVA Test P value Test Equality of Means among Groups 0.215 0.503 Table 6.6 Numbers of Observed Data Points and Mean Production Rate for Embankment: Work Zone Accessibility Number of Data Points Easy Accessibility Moderate Accessibility Difficult Accessibility 8 18 8 Mean Production Rate (CY/Crew Day) 1124 1233 762 6.2.2.3 Embankment: Observed Production Rates and Work Zone congestion The number of data points collected for the three levels of Work Zone congestion; minor, moderate and severe, were 15, 18, and 1 respectively. As only one data point was available for severe congestion, it was excluded from the driver analysis. The t-test was employed to test the difference in mean Production Rate between minor and moderate Work Zone congestion, since the two levels are independent and normally distributed, as shown in Appendix L-2. 137 Table 6.7 presents the results of group variances test and t-test for the two levels. A P-value of 0.34 in the group variances test indicated that the two levels had equal variances at the 95% confidence interval. Based on the equal variances, the P-value of the t-test was found to be 0.009, which is less than 0.05. Therefore, the average Production Rate in Work Zones with minor congestion is significantly different than that with moderate congestion. Table 6.8 indicates that the average Production Rate in the Work Zone is 1,424 CY/Crew Day with minor congestion and 872 CY/Crew Day with moderate congestion. The difference between the two levels is 552 CY/Crew Day. Table 6.7 Results of Group Variances Test and ANOVA Test for Embankment: Work Zone Congestion Homogeneity of Group Variances Test P value Test Equality of Group Variances ANOVA Test P value Test Equality of Means among Groups 0.009 0.34 138 Table 6.8 Numbers of Observed Data Points and Mean Production Rate for Embankment: Work Zone Congestion Number of Data Points Minor Congestion Moderate Congestion 15 18 Mean Production Rate (CY/Crew Day) 1424 872 6.2.3 Lime-Treated Sub-grade The scatter plots, shown in Appendix M, were used to examine the relationships of fourteen Candidate Drivers and observed Production Rates of Lime-treated sub-grade. Relationships for Work Area Quantity, Length of Work These scatter plots are shown in Figures Area, and Location were observed. 6.12, 6.13 and 6.14 respectively. tested for Lime-treated sub-grade. Three sub-hypotheses, listed as follows, were Sub-hypothesis 1: Based on the fact that repetitive work leads to more efficient work Operations and resource allocation, Limetreated sub-grade may be more efficient at higher quantities. In addition, productive hours in a working day may be higher for the Work Area with larger quantity. Sub-hypothesis 2: For Lime-treated sub-grade, the longer the Length of a Work Area the more the number of repetitions there will 139 be. When repetitions increase, the work will be more efficient due to learning effects. Sub-hypothesis 3: Denser population and traffic will cause the work of Limetreated sub-grade to be more dispersed. Dispersion of work has a negative relationship with productivity. A Production Rate (SY/Crew Day) 3000 A A A A A A 2000 A A A A A A AA A AA A A A A A 1000 A AA A A A A A 0 0 10000 20000 30000 40000 50000 Work Area Quantity (SY) Figure 6.12 Scatter Plot for Lime-Treated Sub-grade: Production Rate (SY/Crew Day) vs. Work Area Quantity (SY) 140 A Production Rate (SY/Crew Day) 3000 A A A A A A 2000 A A A A A A A A A A A AA A A A A A A A A A A A A 1000 0 0 10000 20000 30000 Length of Work Area (LF) Figure 6.13 Scatter Plot for Lime-Treated Sub-grade: Production Rate (SY/Crew Day) vs. Length of Work Area (LF) A Production Rate (SY/Crew Day) 3000 A A A A A A 2000 A A A A A A A A A A A A A A 1000 A A A A A 0 Rural Urban A Metro Location Figure 6.14 Scatter Plot for Lime-Treated Sub-grade: Production Rate (SY/Crew Day) vs. Location 141 6.2.3.1 Lime-Treated Sub-grade: Observed Production Rates and Work Area Quantity The logarithmic model was found to be the most efficient model for the relationship between observed Production Rates and Work Area Quantity for Lime-treated sub-grade among the three selected models. Prior to modeling, the box plots of the dependent variable and the independent variable, shown in Figure 6.15 and Figure 6.16, were employed for outlier analysis. The 3rd and 12th data points were found to be outliers and were removed before conducting further regression analysis. The fitted logarithmic model is shown in Figure 6.10. 4000 12 Production Rate (SY/Crew Day) 3000 2000 1000 0 N=32 Figure 6.15 Lime-Treated Sub-grade: Box Plot of Observed Production Rates (SY/Crew Day) 142 12 11 Log (Work Area Quantity (SY)) 10 9 8 7 6 5 6 N=32 Figure 6.16 Lime-Treated Sub-grade: Box Plot of Log Transformation of Work Area Quantity (SY) 4000 Production Rate (SY/Crew Day) 3000 2000 1000 0 0 10000 20000 30000 40000 50000 60000 Work Area Quantity (SY) Figure 6.17 Scatter Plot and Logarithmic Model for Lime-Treated Sub-grade: Observed Production Rates (SY/Crew Day) vs. Work Area Quantity (SY) 143 This model, shown as Equation 6.5, was statistically significant at the 95% confidence interval. Table 6.9 displays the results of a regression analysis using The R2 and adjusted R2 are 0.714 and 0.704 respectively. the logarithmic model. The coefficients of this model were statistically different from zero at the 95% confidence. Production Rate = -6457 + 878 Log (Work Area Quantity) (Equation 6.5) The violations of assumptions of the regression analysis were further tested for the fitted logarithmic model. displayed in Appendix N-1. The plots used to check for violations are No violation of the assumptions was found. Therefore, this model is statistically significant, meaning that Work Area Quantity affects the Production Rate of Lime-treated sub-grade construction. Table 6.9 Logarithmic Model for Lime-Treated Sub-grade: Observed Production Rates (SY/Crew Day) by Work Area Quantity (SY) R2 Adjusted R2 Standard Error Regression Model Variable Work Area Quantity (Constant) F 70.0 B -6457 878 0.714 0.704 432 P value 0.0000 P value 0.0000 0.0010 144 This model is only applicable to Work Area quantities within the range of 1,632 SY to 26,645 SY, because this model was developed on this range. Therefore, the predicted Production Rates of this logarithmic model can range from 38 SY/Crew Day to 2,490 SY/Crew Day. The effect of Work Area Quantity on the Production Rate of Lime-treated sub-grade Operations can be computed by differentiation, shown in the Equation 6.6, of the fitted model. As an example, when two Work Areas have a Work Area Quantity close to 10,000 SY but have a difference of 1,000 CY in quantity, they may exhibit a difference of 88 SY/Crew Day in average Production Rate. d( Production Rate) 878 (Equation 6.6) = d( Work Area Quantity) Work Area Quantity 6.2.3.2 Lime-Treated Sub-grade: Observed Production Rates and Length of Work Area The logarithmic model was found to be the most efficient model for the Length of Work Area for Lime-treated sub-grade. Prior to using the logarithmic model, box plots of the observed Production Rate and the logarithmic transformation of the Length of Work Area, as shown in Figure 6.15 and Figure 6.18, were employed for outlier analysis. The 11th and 12th data point were found to be outliers. These outliers were removed before conducting further 145 regression analysis. The fitted logarithmic model, which was used to explore the relationships between the observed Production Rates and the Length of Work Area for Lime-treated sub-grade construction, is shown in Figure 6.19. This model, shown as Equation 6.7, was statistically significant at the 95% confidence interval. Table 6.10 displays the results of a regression analysis The R2 and adjusted R2 are respectively 0.631 and using the logarithmic model. 0.617. The coefficients of this model were statistically different from zero at the 95% confidence interval since the P-values of the testing coefficients for Work Area Quantity and constant were less than 0.05. 11 11 10 Log (Length of Work Area (LF)) 9 8 7 6 5 4 N=32 Figure 6.18 Lime-Treated Sub-grade: Box Plot of Log Transformation of Length of Work Area (LF) 146 4000 Production Rate (SY/Crew Day) 3000 2000 1000 0 0 2000 4000 6000 8000 10000 Length of Work Area (LF) Figure 6.19 Scatter Plot and Logarithmic Model for Lime-Treated Sub-grade: Observed Production Rates (SY/Crew Day) vs. Length of Work Area Quantity (LF) Production Rate = -3446 + 661 Log (Length of Work Area) (Equation 6.7) Table 6.10 Logarithmic Model for Lime-Treated Sub-grade: Production Rates (SY/Crew Day) by Length of Work Area (LF) R2 Adjusted R2 Standard Error Regression Model Variable Length of Work Area (Constant) F 47.79 B 661 -3446 0.631 0.617 483 P value 0.0000 P value 0.0000 0.0000 147 The plots used to check for violations of the assumptions are displayed in Appendix N-2. No violation of the assumptions was found, as the plots indicate. Therefore, this model is statistically significant and the effects of Length of Work Area on the Production Rate of Lime-treated sub-grade are identified. This model is only applicable to Length of Work Area within the range of 1,632 LF to 50,490 LF, since this model was developed based on observed data in this range. Therefore, the estimated Production Rates of this model can range from 1,444 SY/Crew Day to 3,712 SY/Crew Day. The effects of the Length of Work Area on the Production Rates of Lime-treated sub-grade Operations can be computed by differentiation of the fitted model, as shown in the Equation 6.8. Therefore, when two Work Areas have a length close to 10,000 LF but have a difference of about 1,000 LF in length, they experience a difference of about 66 CY/Crew Day in average Production Rate. d(Production Rate) 661 = d(Length of Work Area) Length of Work Area (Equation 6.8) 6.2.3.3 Lime-Treated Sub-grade: Observed Production Rates and Project location The number of data points collected for Rural, Urban and Metro areas were respectively 5, 24 and 3. The sample size for Metro was too small for 148 comparison. Therefore, ANOVA was used to test the difference in mean Table 6.11 presents the Production Rates between Rural and Urban areas. results of the group variances test and t-test for the two groups. A P-value of 0.018 in the group variances test indicated that the two groups did not have equal variances at a 95% confidence interval. Based on the unequal variances between the two groups, the P-value of the t-test was 0.599 which was not less than 0.05. Therefore, the Rural average Production Rate was not significantly different from Urban at the 95% confidence interval. Table 6.11 Results of Group Variances Test and t-test for Lime-Treated Subgrade: Project Location (Rural and Urban) Homogeneity of Group Variances Test P value Test Equality of Group Variances Independent-Samples T Test P value Test Equality of Means between Groups 0.599 0.018 6.2.4 Aggregate Base Course Two types of aggregate Operations were observed for this study: Flexible base and Cement-treated base (CTB) Operations. Figure 6.20 shows their observed 149 Production Rates. Due to the different requirements on processing operations of these two types of Base course construction, the driver analyses were performed separately. The scatter plots, shown in Appendix O and Appendix P for Cement-treated base and Flexible base respectively, were used to examine the relationship between eleven Candidate Drivers and observed Production Rates. Relationships between observed Production Rates and Work Area Quantity, as well as Length of Work Area were analyzed for both Cement-treated base and Flexible base Operations. 6.23 and Figure 6.24. driver analyses. These are shown in Figure 6.21, Figure 6.22, Figure Other Candidate Drivers were excluded from further Two sub-hypotheses, listed as follows and applicable to both types of Aggregate base Operations, were further tested. Sub-hypothesis 1: Due to their repetitive nature, CTB and Flexible base may become more efficient with greater quantities. In addition, productive hours in a working day may be higher for the Work Area with greater quantity. Sub-hypothesis 2: For Cement-treated base and Flexible base, the longer the length of a Work Area the more the number of repetitions there will be. When repetitions increase, the work will be more efficient due to learning effects. 150 Production Rates (SY-Lift/Crew Day) A 6000 A A A 5000 A A A A 4000 A A A A A A A A A A A A 3000 A 2000 A 1000 A A A A Cement Treated Base Flexible Base Type of Aggregate Base Construction Figure 6.20 Aggregate Base: Scatter Plot of Observed Production Rates (LiftSY/Crew Day) vs. Types of Aggregate Base Operations Production Rate (SY-LIFT/Crew Day) A 6000 A A 5000 A A A A A A A 4000 A 3000 A A 2000 A 10000 20000 30000 Work Area Quantity (SY-LIFT) Figure 6.21 Cement-Treated Base: Scatter Plot of Observed Production Rates (Lift-SY/Crew Day) vs. Work Area Quantity (Lift-SY) 151 A Produc tion Rate (SY-LIF T/Crew Day) 6 00 0 A A 5 00 0 A A A A AA A 4 00 0 3 00 0 A A A 2 00 0 A 0 2 00 0 4 00 0 6 00 0 Lift-Length (LF) Figure 6.22 Cement-Treated Base: Scatter Plot of Observed Production Rates (Lift-SY/Crew Day) vs. Lift-Length of Work Area (LF) A Production Rate (SY-Lift/Crew Day) 5000 A 4000 A A A A 3000 A A A A 2000 A A 1000 A A A 0 10000 20000 30000 40000 Work Area Quantity (SY-Lift) Figure 6.23 Flexible Base: Scatter Plot of Observed Production Rates (LiftSY/Crew Day) vs. Work Area Quantity (Lift-SY) 152 A Production Rate (SY-Lift/Crew Day) 5000 A 4000 A A A A A A 3000 A A 2000 A A 1000 A A A 0 2500 5000 7500 10000 Lift-Length of Work Area (LF) Figure 6.24 Flexible Base: Scatter Plot of Observed Production Rates (LiftSY/Crew Day) vs. Lift-Length of Work Area (LF) 6.2.4.1 Cement-Treated Base: Observed Production Rates and Work Area Quantity The logarithmic model was found to be the most efficient model for the relationship between observed Production Rates and Work Area Quantity for Cement-treated base, among the three selected models. The box plots of the dependent variable and the independent variable, shown in Figure 6.25 and Figure 6.26, were employed for outlier analysis. No outlier was observed from these two plots. The fitted logarithmic model is shown in Figure 6.27. 153 7000 Production Rate (SY-Lift/Crew Day) 6000 5000 4000 3000 2000 1000 N=14 Figure 6.25 Cement-Treated Base: Box Plot of Observed Production Rates (LiftSY/Crew Day) 11 Log (Work Area Quantity (SY-Lift)) 10 9 8 7 N=14 Figure 6.26 Cement-Treated Base: Box Plot of Log Transformation of Work Area Quantity (Lift-SY) 154 7000 Production Rate (SY-Lift/Crew Day) 6000 5000 4000 3000 2000 1000 0 10000 20000 30000 40000 Work Area Quantity (SY-LIFT) Figure 6.27 Scatter Plot and Linear Model for Cement-Treated Base: Production Rates (Lift-SY/Crew Day) vs. Work Area Quantity (Lift-SY) This model, shown as Equation 6.9, was statistically significant at the 95% confidence interval. Table 6.12 displays the results of a regression analysis The R2 and adjusted R2 are 0.627 and 0.596 using the logarithmic model. respectively. The coefficients of this model were found to be statistically different from zero at the 95% confidence interval. Production Rate = -6991 + 1231 Log (Work Area Quantity) (Equation 6.9) The violations of the assumptions of the regression analysis were further tested for the fitted logarithmic model and the plots used to check for them are 155 displayed in Appendix Q-1. None were found. Therefore, this model is statistically significant, meaning that Work Area Quantity positively affects Production Rates in Cement-treated base construction, according to the fitted model. Table 6.12 Linear Model for Cement-Treated Base: Production Rates (LiftSY/Crew Day) by Work Area Quantity (Lift-SY) R2 Adjusted R2 Standard Error Regression Model Variable Work Area Quantity (Constant) F 20.15 B 1231 -6991 0.627 0.596 873.4 P value 0.0007 P value 0.0007 0.0152 This model is only applicable for Work Area quantities between 1,416 LiftSY and 35,956 Lift-SY. Therefore, the estimated Production Rates of this fitted logarithmic model can range from 1,941 Lift-SY/Crew Day to 5,922 SY/Crew Day. The effects of the Work Area Quantity on the Production Rates of Cement-treated base Operations can be computed from differentiation of the fitted logarithmic model, shown in the Equation 6.10. Therefore, when two Work Areas have a quantity to place close to 10,000 Lift-SY, yet are different by 1,000 Lift-SY in Length of Work Area, they will experience a difference of about 123 Lift-SY/Crew Day in average Production Rate. 156 d( Production Rate) 1231 (Equation 6.10) = d( Work Area Quantity) Work Area Quantity 6.2.4.2 Cement-Treated Base: Observed Production Rates and Lift-length of Work Area The linear model was found to be the most efficient model for the relationship between observed Production Rates and Lift-length of Work Area for Cementtreated base. Prior to using the linear model, the box plots of the dependent variable (i.e. the observed Production Rate) and the independent variable (i.e. the Lift-length of Work Area), as shown in Figure 6.25 and Figure 6.28, were employed for outlier analysis. outliers. analysis. The 12th and 14th data points were found to be These two outliers were removed before conducting further regression The fitted linear model is shown in Figure 6.29. 8000 12 Lift-Length of Work Area (LF) 6000 14 4000 2000 0 N=14 Figure 6.28 Cement-Treated Base: Box Plot of Lift-Length of Work Area (LF) 157 7000 Production Rate (SY-Lift/Crew Day) 6000 5000 4000 3000 2000 1000 0 1000 2000 3000 4000 Lift-Length (LF) Figure 6.29 Scatter Plot and Linear Model for Cement-Treated Base: Observed Production Rates (Lift-SY/Crew Day) vs. Lift-Length of Work Area (LF) Table 6.13 displays the results of a regression analysis to model the relationship between observed Production Rates and Lift-length of Work Area for Cement-treated base construction. This model, shown in Equation 6.11, was statistically significant at the 95% confidence interval. The R2 and adjusted R2 are 0.393 and 0.332 respectively. The coefficients of this fitted model were statistically different from zero at the 95% confidence interval since the P-values of testing coefficients for Work Area Quantity and constant were less than 0.05. Production Rate = 2366 +1.02 (Lift-Length of Work Area) (Equation 6.11) 158 Table 6.13 Linear Model for Cement-Treated Base: Production Rates (LiftSY/Crew Day) by Lift-Length of Work Area (LF) R2 Adjusted R2 Standard Error Regression Model Variable Lift-Length of Work Area (Constant) F 6.47 B 1.02 2366 0.393 0.332 1106 P value 0.0291 P value 0.0291 0.0053 A check for violations of the assumptions of the regression analysis was performed but none were found, as presented in Appendix Q-2. model is statistically significant. This model is only applicable to Work Area quantities within the range of 250 LF to 3,250 LF. Therefore, the estimated Production Rates of the fitted linear Therefore, this model can range from 2,621 Lift-SY/Crew Day to 5,681 Lift-SY/Crew Day. The effects of Lift-length of Work Area on the Production Rate of Cement-treated base Operations can be computed from differentiation of the fitted model, as shown in Equation 6.12. Therefore, when two Work Areas are 100 LF different in Lift-length, they may experience a Production Rate difference of about 102 Lift-SY/Crew Day. 159 d( Production Rate) = 1.02 d( Lift - Length of Work Area ) (Equation 6.12) 6.2.4.3 Flexible Base: Observed Production Rates and Work Area Quantity The logarithmic model was found to be the best model for Flexible base. Prior to using the logarithmic model, an outlier analysis was performed but none were found. Figure 6.30 and Figure 6.31 were employed for outlier analysis. The fitted logarithmic model, which was used to identify the relationship between observed Production Rates and Work Area Quantity for Flexible base construction, is shown in Figure 6.32. 6000 Production Rate (SY-Lift/Crew Day) 5000 4000 3000 2000 1000 0 N=15 Figure 6.30 Flexible Base: Box Plot of Observed Production Rates (Lift-SY/Crew Day) 160 11 Log (Work Area Quantity (SY-Lift)) 10 9 8 7 N=15 Figure 6.31 Flexible Base: Box Plot of Logarithmic Transformation of Work Area Quantity (Lift-SY) 6000 Production Rate (SY-Lift/Crew Day) 5000 4000 3000 2000 1000 0 0 10000 20000 30000 40000 50000 Work Area Quantity (SY-Lift) Figure 6.32 Scatter Plot and Linear Model for Flexible Base: Observed Production Rates (Lift-SY/Crew Day) vs. Work Area Quantity (LiftSY) 161 Table 6.14 displays the results of a regression analysis used for the relationship between observed Production Rate and Work Area Quantity for Flexible base construction. This model, shown as Equation 6.13, is statistically The R2 and adjusted R2 are 0.594 and significant at the 95% confidence interval. 0.562 respectively. Production Rate = -7761 + 1126 Log (Work Area Quantity) (Equation 6.13) Violations of the assumptions of regression analysis were further tested for the fitted logarithmic model and associated plots are displayed in Appendix R-1. violation of the assumptions was found, as the plots indicate. No Therefore, this model is statistically significant and the relationship between Work Area Quantity and Production Rate is identified, according to the fitted model. Table 6.14 Logarithmic Model for Flexible Base: Production Rates (LiftSY/Crew Day) by Work Area Quantity (Lift-SY) R2 Adjusted R2 Standard Error Regression Model Variable Work Area Quantity (Constant) F 19 B 1126 -7761 0.594 0.562 966 P value 0.0008 P value 0.0008 0.0071 162 This model is only applicable within the range from 1,579 Lift-SY to 41,607 Lift-SY. Therefore, the estimated Production Rates of the fitted logarithmic The model can range from 531 Lift-SY/Crew Day to 4,215 SY/Crew Day. effects of Work Area Quantity on Production Rate for Flexible base Operations can be computed from differentiation of the fitted model, as shown in the Equation 6.14. Therefore, when two Work Areas have a quantity of Flexible base close to 10,000 Lift-SY but differ by 1,000 Lift-SY, they may have a difference of about 112 Lift-SY/Crew Day in average Production Rate. d( Production Rate) 1126 = d( Work Area Quantity) Work Area Quantity (Equation 6.14) 6.2.4.4 Flexible Base: Observed Production Rates and Lift-length of Work Area The logarithmic model was found to be the most efficient model for the relationship between observed Production Rates and Lift-length of Work Area for Flexible base. Prior to using the logarithmic model, box plots of the dependent and independent variables, shown in Figure 6.30 and Figure 6.33, were employed for outlier analysis. No outlier was observed from these two plots. The fitted logarithmic model is shown in Figure 6.34. 163 10 Log (Lift-Length of Work Area (LF)) 9 8 7 6 5 N=15 Figure 6.33 Flexible Base: Box Plot of Logarithmic Transformation of Work Area Quantity (Lift-SY) This model, shown as Equation 6.15, was statistically significant at the 95% confidence interval. Table 6.15 displays the results of the regression analysis using the logarithmic model. Production Rate = -4990 + 997 Log (Lift-Length of Work Area) (Equation 6.15) 164 6000 Production Rate (SY-Lift/Crew Day) 5000 4000 3000 2000 1000 0 0 2000 4000 6000 8000 10000 12000 Lift-Length of Work Area (LF) Figure 6.34 Scatter Plot and Logarithmic Model for Flexible Base: Observed Production Rates (Lift-SY/Crew Day) vs. Lift-Length of Work Area (LF) Table 6.15 Logarithmic Model for Flexible Base: Production Rates (LiftSY/Crew Day) by Lift-Length of Work Area (LF) R2 Adjusted R2 Standard Error Regression Model Variable Length of Work Area (Constant) F 16.33 B 997 -4990 0.557 0.523 1009 P value 0.0014 P value 0.0014 0.0233 165 Violations of the assumptions of regression analysis were further tested for the fitted logarithmic model. violation was found. The plots are displayed in Appendix R-2. No Therefore, the fitted model is statistically significant. This model is only applicable to Lift-length of Work Area within the range of 263 LF to 11,371 LF because this model was developed based on the observed data in this range. Therefore, the estimated Production Rates of the fitted logarithmic model can range from 565 Lift-SY/Crew Day to 4,321 Lift-SY/Crew Day. The effects of Lift-length of Work Area on Production Rate of Flexible Therefore, when two base can be computed, as shown in the Equation 6.16. Work Areas have a Lift-length of Work Area close to 10,000 Lift-SY yet differ by 1,000 Lift-SY, they may differ in Production Rate by about 100 Lift-SY/Crew Day. d(Production Rate) 997 = d(Lift - Length of Work Area Lift - Length of Work Area ) (Equation 6.16) 6.3 CORRELATIONS TESTING OF DRIVERS The next step in this study involved testing of the identified drivers for correlations. A Pearson correlation of 0 indicates that two variables are totally independent, and the interaction effects of two variables on Production Rates can be computed as the sum of the effects of each variable. In contrast, a Pearson 166 correlation value further away from 0 indicates increasing correlation between the two variables. The combined effects of multiple variables should be obtained from further analysis, such as multiple regression analysis. If two drivers have high correlations, multicollinearity will limit the investigation of effects from a multiple regression model (Wonnacott and Wonnacott 1987). Table 6.16 shows the correlation table of two drivers for Embankment construction. A Pearson correlation of -0.247 implies that the two drivers are Therefore, the effects of the two drivers should be not highly correlated. considered together to increase accuracy of the Production Rate model. Table 6.16 Correlations Test for Work Area Quantity and Work Zone Congestion of Embankment Construction Work Zone Congestion Logarithmic Transformation of Work Area Quantity -0.247 0.173 Pearson Correlations Sig. (2-tailed) For Lime-treated sub-grade construction, the correlations between drivers are shown in Table 6.17. Not surprising, high correlations between Work Area Therefore, the effects of Quantity and Length of Work Area were observed. Work Area Quantity and Length of Work Area should not be considered together in one model. 167 Table 6.17 Correlations Test for Work Area Quantity and Length of Work Area of Lime-Treated Sub-grade Construction Logarithmic Transformation of Length of Work Area Logarithmic Transformation of Work Area Quantity 0.875 ** Pearson Correlations Sig. (2-tailed) 0.000 **. Correlation is significant at the 0.01 level (2 tailed) Correlations tests on Work Area Quantity and Lift-length of Work Area for Cement-treated base and Flexible base are presented in Table 6.18 and Table 6.19. Not surprising, both indicated high correlations between Work Area Therefore, they should not be Quantity and Lift-length of Work Area. considered together in one model. Table 6.18 Correlations Test for Work Area Quantity and Lift-Length of Work Area of Cement-Treated Base Construction Lift-Length of Work Area Logarithmic Transformation of Work Area Quantity 0.854 ** Pearson Correlations Sig. (2-tailed) 0.000 **. Correlation is significant at the 0.01 level (2 tailed) 168 Table 6.19 Correlations Test for Work Area Quantity and Lift-Length of Work Area of Flexible Base Construction Logarithmic Transformation of Lift-Length of Work Area Logarithmic Transformation of Work Area Quantity 0.954 ** Pearson Correlations Sig. (2-tailed) 0.000 **. Correlation is significant at the 0.01 level (2 tailed) 6.4 EFFECTS OF MULTIPLE DRIVERS ON PRODUCTION RATES Multiple regression analysis was used to explore the effects of multiple drivers on Production Rates for targeted Work Items. According to the required sample size for regression analysis and the assumption of independent variables, only Embankment was eligible for further multiple regression analysis. 6.4.1 Embankment: Production Rates by Logarithmic Transformation of Work Area Quantity and Work Zone Congestion In the multiple regression analysis for Embankment, the dependent variables are estimated Production Rates and the independent variables are logarithmic transformation of Work Area Quantity and Work Zone congestion. One outlier and the data pertaining to a severely congested Work Zone were removed before conducting a multiple regression analysis. The data for Work Zone congestion were recoded as binary data (minor congested Work Zones were recoded as 0, and moderately congested Work Zones were recoded as 1). 169 Table 6.20 displays the results of the multiple regression analysis. The fitted model, shown as Equation 6.17, is statistically significant at the 95% confidence interval. The R2 and adjusted R2 are 0.4 and 0.358 respectively. The coefficients of the fitted model are statistically different from zero at the 95% confidence interval since the P-values of testing coefficients for Work Area Quantity and the constant term are less than 0.05. Production Rate = -575 + 254 Log (Work Area Quantity) 313 (Work Zone Congestion) (Equation 6.17) Table 6.20 Multiple Regression Model for Embankment R2 Adjusted R2 Standard Error Regression Model Variable Log (Work Area Quantity) Work Zone Congestion (Constant) F 9.65 B 254 -313 -575 0.4 0.358 424 P value 0.001 P value 0.003 0.055 0.449 The test for violation of the assumption of constant variances was employed on the fitted multiple regression model. The plot used to check for the violation No violation of the assumption of the assumption is displayed in Appendix S. 170 was found. Thus, this model is statistically significant and the effects of Work Area Quantity and Work Zone congestion on Production Rates were further quantified. This model is only applicable for Work Area quantities within the range of 1,064 CY to 33,938 CY, and not for the Work Zones with severe congestion. Therefore, the estimated Production Rates of this model can range from 882 CY/Crew Day to 2,075 CY/Crew Day. The effects of Work Area Quantity on the Production Rates of Embankment Operations can be computed by differentiation of the fitted multiple regression model, as shown in Equation 6.18. Therefore, when two Work Areas have the quantity of Embankment close to 10,000 CY but differ by 1,000 CY in quantity, they may have a difference of about 25 CY/Crew Day in average Production Rate. The effect of minor Work Zone congestion results in a Production Rate of 313 CY/Crew Day better than moderate Work Zone congestion. d( Production Rate) = d( Work Area Quantity) Work 254 (Equation 6.18) Area Quantity 6.5 SUMMARY OF FINDINGS ON DRIVER ANALYSES Table 6.21 summarizes the results of driver analysis. Project type was not analyzed due to insufficient data. None of the investigated Candidate Drivers at 171 the project level was found to significantly affect Production Rates. For Candidate Drivers at the Work Zone-level, only Work Zone congestion was found to be a Production Rate driver for Embankment. For Candidate Drivers at the Work Item-level, Work Area Quantity was identified as a driver for the four targeted Earthwork-related Work Items. Length of Work Area was identified as a Production Rate driver of Lime-treated sub-grade and Aggregate base. Table 6.21 Summary of Results of Driver Analyses Lime-Treated Sub-grade Cement Treated Base Candidate Drivers Excavation Embankmant Flexible Base Project Type Project Level Project Location Traffic Flow Project Complexity Accelerated Construction Provision Contractor Management Skill Work Zone Level Work Zone Accessibility Work Zone Congestion Work Zone Drainage Effectiveness Work Zone Clay Content Work Zone Land Slope Work Area Quantity Work Item Level Soil Condition Length of Work Area Type of Lime Used Thickness Width of Work Area : Driver found to be statistically significant : Investigated but not statistically significant : Insufficient data for analysis (Lift) (Lift-Length) 172 Table 6.22 lists the identified drivers that have a statistically significant relationship with Production Rates for major Earthwork construction Work Items. A multiple regression model was developed for Embankment to illustrate interaction effects of the identified drivers. For Lime-treated sub-grade and Aggregate base, multiple regression analysis is not applicable because of a high correlation between drivers. Table 6.22 Summary of Identified Production Rate Drivers Work Item Excavation Embankment Drivers Work Area Quantity Work Area Quantity Work Zone Congestion Lime Treated Sub-grade Work Area Quantity Length of Work Area Cement Treated Base Aggregate Base Flexible Base Work Area Quantity Length of Work Area Work Area Quantity Length of Work Area Type of Regression Model used for Analysis Log model Log model ******** Log model Log model Log model Linear Model Log model Log model Regression Analysis/T Test R = 0.692 R = 0.343 P=0.009 R = 0.714 R = 0.631 R = 0.627 R = 0.393 R = 0.594 R = 0.557 2 2 2 2 2 2 2 2 Multiple Regression Analysis None R = 0.4 Adjusted R = 0.358 *None 2 2 *None *None *None: Because the two drivers are highly correlated, multiple regression analysis is not applicable. **None: The sample size was not sufficient for multiple regression analysis. 173 CHAPTER VII: DATA ANALYSIS AND HYPOTHESIS TESTS FOR PAVEMENT-RELATED WORK ITEMS 7.1 TEST DIFFERENCE IN MEAN PRODUCTION RATES The mean observed Production Rates of Pavement-related Work Items were compared with average CTDS rates to test the first hypothesis which is presented as follows: Hypothesis 1: The Production Rates of the CTDS are not realistic. The established null hypothesis is that the mean observed Production Rates are equivalent to the average CTDS rates. In other words, the established alternative hypothesis is that the average CTDS Production Rates are different from mean observed Production Rates. The results of the hypothesis testing are displayed in Table 7.1. Conventional form concrete pavement was excluded in this hypothesis test since its Production Rates are not available from the CTDS. It appears that the Production Rates of the CTDS are too optimistic for Hot mix asphalt pavement (HMAP) and Slip-form concrete pavement. 174 Table 7.1 Average CTDS Production Rates and Mean Observed Production Rates for HMAP and Slip-form Concrete Pavement Work Item Hot Mix Asphalt Pavement Slip-form Concrete Pavement Number of Data Points 32 20 Unit TON/Crew Day SY-Lift/Crew Day Mean Observed Production Rate 817 1253 Average CTDS Mean Rate Difference 1200 3000 -383 -1747 P-Value *0.000 *0.000 * indicates that P-value is less than 0.05 and thus, the null hypothesis (Mean Observed Production Rate = Average CTDS Rate) is rejected at 95% confidence interval. 7.2 ANALYSIS OF DRIVERS OF PRODUCTION RATES Several Candidate Drivers were selected in Section 4.1 and further investigated for their effects on Production Rates. The following is the formulation of the second hypothesis. Hypothesis 2: The Production Rates of the targeted Work Items are driven by some productivity factors that are known at the design stage. 7.2.1 Hot Mix Asphalt Pavement The scatter plots, shown in Appendix T, were used to examine the relationships between eleven Candidate Drivers and observed Production Rates for Hot mix asphalt pavement Operations. Relationships for Work Area Quantity and Difference in mean Production Rate for Course types were observed. The scatter plots of these two Candidate Drivers are shown in Figures 7.1 and 7.2. Other Candidate Drivers were excluded from further driver analyses. The two sub-hypotheses, listed as follows, were tested further. 175 Sub-hypothesis 1: Hot mix asphalt pavement Operations may experience increased productivity in larger Operations. In addition, Work Area with higher quantity may yield more effective daily working hours. Sub-hypothesis 2: Surface courses are usually built to a higher standard of quality than Base courses, therefore Hot mix asphalt pavement Base course Operations may have higher Production Rates. 1500 A A A A Production Rate (Ton/Crew Day) A A A A 1000 A A A A A A A A A A A A A A A A 500 A A A A A A A A 1000 2000 3000 4000 5000 6000 Work Area Quantity (Ton) Figure 7.1 Hot Mix Asphalt Pavement: Scatter Plot of Observed Production Rates (Ton/Crew Day) vs. Work Area Quantity (Ton) 176 1500 A A A A Production Rate (Ton/Crew Day) A A A 1000 A A A A A A A A A A A A A A A 500 A A A A A A A A Base Course Surface Surface and Base Course Type Figure 7.2 Hot Mix Asphalt Pavement: Scatter Plot of Observed Production Rates (Ton/Crew Day) vs. Course Type 7.2.1.1 Hot Mix Asphalt Pavement: Observed Production Rates and Work Area Quantity The logarithmic model was found to be the most efficient model for the relationship between observed Production Rates and Work Area Quantity for Hot mix asphalt pavement among three selected models (linear model, logarithmic model and power model). Box plots, shown in Figure 7.3 and Figure 7.4, were No outlier was observed. The fitted logarithmic employed for outlier analysis. model is shown in Figure 7.5. 177 1600 1400 Production Rate (Ton/Crew Day) 1200 1000 800 600 400 200 0 N=32 Figure 7.3 Hot Mix Asphalt Pavement: Box Plot of Observed Production Rates (Ton/Crew Day) 9 Log (Work Area Quantity (Ton)) 8 7 6 5 N=32 Figure 7.4 Hot Mix Asphalt Pavement: Box Plot of Logarithmic Transformation of Work Area Quantity (Ton) 178 1600 1400 Production Rate (Ton/Crew Day) 1200 1000 800 600 400 200 0 0 1000 2000 3000 4000 5000 6000 Work Area Quantity (Ton) Figure 7.5 Scatter Plot and Logarithmic Model for Hot Mix Asphalt Pavement: Observed Production Rates (Ton/Crew Day) vs. Work Area Quantity (Ton) The fitted model, shown as Equation 7.1, was found to be statistically significant at the 95% confidence interval. Table 7.2 displays the results of a regression analysis using the logarithmic model. The R2 and adjusted R2 are 0.432 and 0.414 respectively. Production Rate = -1198 + 278 Log (Work Area Quantity) (Equation 7.1) No violation of assumptions was found, as shown in Appendix U-1. Therefore, this model is statistically significant. This model is only applicable to Work Area quantities within the range of 227 Tons to 5,840 Tons. 179 Therefore, the estimated Production Rates of the fitted logarithmic model can range from 310 Tons/Crew Day to 1,213 Tons/Crew Day. The effects of Work Area Quantity on the Production Rates of Hot mix asphalt pavement can be computed from differentiation of the fitted model, as shown in the Equation 7.2. As an example, when two Work Areas have a quantity of Hot mix asphalt pavement close to 1,000 Tons but differ by 100 Tons in total quantity, they may experience a difference of about 28 Tons/Crew Day in average Production Rate. Table 7.2 Logarithmic Model for Hot Mix Asphalt Pavement: Production Rates (Ton/Crew Day) by Work Area Quantity (Ton) R2 Adjusted R2 Standard Error Regression Model Variable Work Area Quantity (Constant) F 22.86 B 278 -1198 0.432 0.414 268 P value 0.0000 P value 0.0000 0.0083 d( Production Rate) 278 (Equation 7.2) = d( Work Area Quantity) ( Work Area Quantity) 180 7.2.1.2 Hot Mix Asphalt Pavement: Observed Production Rates and Course type A total of thirty-two data points were observed in this portion of the study. Twenty-two pertained to Base course construction and nine pertained to Surface course. One observation included both Surface and Base course construction. In order to investigate Production Rate difference between Base and Surface course, the data point observed with both Surface and Base course construction was removed. The t-test was employed to test the difference in mean Production Rate between Surface and Base course construction, since the two groups are independent and both groups are normally distributed (Appendix U-2). Table 7.3 presents the results of the t-test for the two groups. The homogeneity testing of variance yield a P-value of 0.103, thus, indicated that the two groups had equal variance at 90% confidence interval. Based on the assumption of equal variance between two groups, the P-value of t-test was 0.093, which was less than 0.1. Therefore, it can be concluded that the average Production Rate between Surface and Base course construction is different at the 90% confidence interval. 181 Table 7.3 Results of Group Variances Test and ANOVA Test for Hot Mix Asphalt Pavement: Course Type Homogeneity of Group Variances Test P value Test Equality of Group Variances Independent-Samples T Test P value Test Equality of Means among Groups 0.093 0.103 Table 7.4 shows that the average Production Rate of Surface course construction was 646 Tons/Crew Day and the average Production Rate of the Base course construction was 882 Tons/Crew Day. The difference of average Production Rate between the two types of course construction was 236 Tons/Crew Day. Table 7.4 Hot Mix Asphalt Pavement: Numbers of Data Points and Mean Production Rate Number of Data Points Base Course Surface Course 22 9 Mean Production Rate (Tons/Crew Day) 882 646 182 7.2.2 Slip-form Concrete Pavement The scatter plots, shown in Appendix V, were used to examine the relationship between fourteen Candidate Drivers and observed Production Rates for Slip-form concrete pavement. Relationships for Work Area Quantity and The scatter plots for these two Candidate Length of Work Area were observed. Drivers are shown in Figure 7.6 and Figure 7.7. Sub-hypothesis 1: Slip-form concrete pavement Operations may experience increased Production Rate for larger Operations. In addition, Work Areas with higher quantity may yield more effective daily working hours. Sub-hypothesis 2: A longer Length of Work Area may contribute to increased Production Rate. 183 A 2000 A Production Rate (SY/Crew Day) A A A 1500 A A A A A A A A 1000 A AA A A A 500 0 A 10000 20000 30000 40000 Work Area Quantity (SY) Figure 7.6 Slip-form Concrete Pavement: Scatter Plot of Observed Production Rates (SY/Crew Day) vs. Work Area Quantity (SY) A 2000 A Production Rate (SY/Crew Day) A A A 1500 A A A A A A A A 1000 A A A A A A 500 0 A 5000 10000 15000 Length of Work Area (LF) Figure 7.7 Slip-form Concrete Pavement: Scatter Plot of Observed Production Rates (SY/Crew Day) vs. Length of Work Area (LF) 184 7.2.2.1 Slip-form Concrete Pavement: Observed Production Rates and Work Area Quantity The logarithmic model was found to be the most efficient model for the relationship between observed Production Rates and Work Area Quantity for Slip-form concrete pavement. Prior to using the fitted logarithmic model to model the relationships, the box plots, shown in Figures 7.8 and 7.9, were employed for outlier analysis. The 4th data points were found to be an outlier. The fitted It was removed before conducting further regression analysis. logarithmic model for Slip-form concrete pavement construction is shown in Figure 7.10. 3000 Production Rate (SY/Crew Day) 2000 1000 0 N=20 Figure 7.8 Slip-form Concrete Pavement: Box Plot of Observed Production Rates (SY/Crew Day) 185 11 10 Log (Work Area Quantity (SY)) 9 8 7 4 6 N=20 Figure 7.9 Slip-form Concrete Pavement: Box Plot of Logarithmic Transformation of Work Area Quantity (SY) 3000 Production Rate (SY/Crew Day) 2000 1000 0 0 10000 20000 30000 40000 Work Area Quantity (SY) Figure 7.10 Scatter Plots and Logarithmic Model for Slip-form Concrete Pavement: Observed Production Rates (SY/Crew Day) vs. Work Area Quantity (SY) 186 This model, shown as Equation 7.3, is statistically significant at the 95% confidence interval. Table 7.5 displays the results of a regression analysis using The R2 and adjusted R2 are 0.653 and 0.632 respectively. the logarithmic model. The coefficients of this model were statistically different from zero at the 95% confidence interval since the P-values of testing coefficients for Work Area Quantity and constant term were less than 0.05. Production Rate = -2274 + 408 Log (Work Area Quantity) (Equation 7.3) Table 7.5 Logarithmic Model for Slip-form Concrete Pavement: Production Rates (SY/Crew Day) by Work Area Quantity (SY) R2 Adjusted R2 Standard Error Regression Model Variable Work Area Quantity (Constant) F 32 B 408 -2274 0.653 0.632 256 P value 0.0000 P value 0.0000 0.0022 The plots used to check for violations of the assumptions are displayed in Appendix W-1, and none were found. Therefore, the fitted model is statistically significant. This model is only applicable to Work Area quantities within the range of 1,156 SY to 18,592 SY. Therefore, the estimated Production Rates of 187 the fitted logarithmic model can range from 591 SY/Crew Day to 1,752 SY/Crew Day. The effects of the Work Area Quantity on the Production Rates of Slip- form concrete pavement construction can be computed from differentiation of the fitted model, as shown in the Equation 7.4. Therefore, when two Work Areas have the quantity of Slip-form concrete pavement close to 10,000 SY and yet differ by 1,000 SY in Work Area Quantity, productivity may differ by 42 SY/Crew Day in the average daily Production Rate. d( Production Rate) 418 (Equation 7.4) = d( Work Area Quantity ) ( Work Area Quantity ) 7.2.2.2 Slip-form Concrete Pavement: Observed Production Rates and Length of Work Area The logarithmic model was found to be the most efficient model for the relationship between observed Production Rates and Length of Work Area for Slip-form concrete pavement construction. Prior to using the logarithmic model to model the relationship, the box plots of the dependent variable (i.e. the observed Production Rate) and the independent variable (i.e. the logarithmic transformation of Length of Work Area), shown in Figure 7.8 and Figure 7.11, were employed for outlier analysis. The 4th and 7th data points were found to be outliers. Two outliers were removed before conducting further regression 188 analysis. The fitted logarithmic model for Slip-form concrete pavement construction is shown in Figure 7.12. 10 7 9 Log (Length of Work Area) 8 7 6 4 5 N=20 Figure 7.11 Slip-form Concrete Pavement: Box Plot of Logarithmic Transformation of Length of Work (LF) The fitted model, shown as Equation 7.5, was statistically significant at the 95% confidence interval. Table 7.6 displays the results of a regression analysis that models the relationship of observed Production Rates and Length of Work Area for Slip-form concrete pavement construction. were respectively 0.356 and 0.316. The R2 and adjusted R2 The coefficient for the Length of Work Area of the fitted model was statistically different from zero at the 95% confidence interval. Although the constant term was not statistically different from zero in the fitted model at the 95% confidence interval, the fitted model can still be used 189 to quantify the relationship between Work Area Quantity and observed Production Rates. Production Rate = -1193 + 306 Log (Length of Work Area) (Equation 7.5) Violations of the assumptions of regression analysis were further tested for the fitted logarithmic model and are displayed in Appendix W-2. No violation of the assumptions was found, as the plots indicate. Therefore, the fitted model is statistically significant, meaning that Length of Work Area significantly affects Production Rates in Slip-form concrete pavement according to the fitted model. 3000 Production Rate (SY/Crew Day) 2000 1000 0 0 2000 4000 6000 8000 Length of Work Area (LF) Figure 7.12 Scatter Plot and Logarithmic Model for Slip-form Concrete Pavement: Observed Production Rates (SY/Crew Day) vs. Work Area Quantity (SY) 190 Table 7.6 Logarithmic Model for Slip-form Concrete Pavement: Production Rates (SY/Crew Day) by Work Area Quantity (SY) R2 Adjusted R2 Standard Error Regression Model Variable Length of Work Area (Constant) F 8.85 B 306 -1193 0.356 0.316 328 P value 0.0089 P value 0.0089 0.1659 This model is only applicable to Length of Work Area with the range from 473 LF to 7,783 LF. Therefore, the estimated Production Rates of the fitted logarithmic model can range from 692 SY/Crew Day to 1,549 SY/Crew Day. The effects of the Length of Work Area on the Production Rates of Slip-form concrete pavement construction can be computed from differentiation of the fitted model, as shown in the Equation 7.6. Therefore, two Work Areas with Length of Work Area close to 1,000 LF and that differ by 100 LF may show a difference of about 31 SY/Crew Day in the average Production Rate. d(Production Rate) 306 (Equation 7.6) = d(Length of Work Area) ( Length of Work Area) 191 7.2.3 Analysis of Drivers of Production Rates for Conventional Form Concrete Pavement The scatter plots shown in Appendix X were used to examine the relationship between fourteen Candidate Drivers and observed Production Rates for Conventional form concrete pavement. Relationships for Work Area Quantity The scatter plots of Other were found as well as for different types of Configuration. these two Candidate Drivers are shown in Figure 7.13 and Figure 7.14. Candidate Drivers were excluded from further analysis. Sub-hypothesis 1: Repetition and higher quantities should contribute to an increased Production Rate for Conventional form concrete pavement. In addition, Work Areas with higher quantity may yield more effective daily working hours. Sub-hypothesis 2: Formwork and rebar installation for Convention form concrete pavement with curves or sharp angles take longer than for the Concrete pavement with regular shapes. This may explain why Concrete pavement Operations with curve(s) or sharp angle(s) have lower Production Rates. 192 600 A A A Production Rate (SY/Crew Day) 500 A A 400 A A A A 300 A A A A A 200 A 100 A A 0 1000 2000 3000 4000 Work Area Quantity (SY) Figure 7.13 Conventional Form Concrete Pavement: Scatter Plot of Observed Production Rates (SY/Crew Day) vs. Work Area Quantity (SY) 600 A A Production Rate (SY/Crew Day) 500 A A 400 A A A A A A A A A 300 A 200 A A 100 A A A None Yes Configuration (Curve or Sharp Angle) Figure 7.14 Conventional Form Concrete Pavement: Scatter Plot of Observed Production Rates (SY/Crew Day) vs. Configuration 193 7.2.3.1 Conventional Form Concrete Pavement: Observed Production Rates and Work Area Quantity The logarithmic model was found to be the best model for the relationship between observed Production Rates and Work Area Quantity for Conventional form concrete pavement. Prior to using the logarithmic model, box plots shown The 13th and in Figure 7.15 and Figure 7.16 were employed for outlier analysis. 14th data points were found to be outliers. conducting the regression analysis. These outliers were removed before The fitted logarithmic model for Conventional form concrete pavement construction is shown in Figure 7.17. 700 600 Production Rates (SY/Crew Day) 500 400 300 200 100 0 N=20 Figure 7.15 Conventional Form Concrete Pavement: Box Plot of Observed Production Rates (SY/Crew Day) 194 9 8 Log (Work Area Quantity (SY)) 7 6 5 4 3 14 13 2 N=20 Figure 7.16 Conventional Form Concrete Pavement: Box Plot of Logarithmic Transformation of Work Area Quantity (SY) 600 500 Production Rates (SY/Crew Day) 400 300 200 100 0 0 1000 2000 3000 4000 5000 Work Area Quantity (SY) Figure 7.17 Pavement: Observed Production Rates (SY/Crew Day) vs. Logarithmic Transformation of Work Area Quantity (SY) 195 Table 7.7 displays the results of a regression analysis using the logarithmic model. The fitted model, shown in Equation 6.23, is statistically significant at The R2 and adjusted R2 are 0.511 and 0.481 the 95% confidence interval. respectively. The coefficients of the fitted model are statistically different from zero at the 95% confidence interval since the P-values of testing coefficients for Work Area Quantity and the constant term were less than 0.05. Production Rate = -583 + 135 Log (Work Area Quantity) (Equation 7.7) Violations of the assumptions of regression analysis were further tested for the fitted logarithmic model, and the plots used to check for such violations are displayed in Appendix Y-1. No violation was found. Therefore, the fitted model is statistically significant and effects of Work Area Quantity on Production Rates were further investigated according to the fitted model. Table 7.7 Linear Model for Conventional Form Concrete Pavement: Production Rates (SY/Crew Day) by Work Area Quantity (SY) R2 Adjusted R2 Standard Error Regression Model Variable Work Area Quantity (Constant) F 16.7 B 135 -583 0.511 0.481 107 P value 0.0009 P value 0.0009 0.0200 196 This model is only applicable for Work Area quantities within the range of 211 SY to 4,320 SY. Therefore, the estimated Production Rates of the fitted The logarithmic model can range from 140 SY/Crew Day to 547 SY/Crew Day. effects of Work Area Quantity on Production Rates for Conventional form concrete pavement construction can be computed from differentiation of the fitted model, as shown in the Equation 7.8. Therefore, when two Work Areas have a quantity of Conventional form concrete pavement close to 1,000 SY but differ by 100 SY in quantity, they may experience about 14 SY/Crew Day difference in average Production Rate. d( Production Rate) = d( Work Area Quantity ) Work 135 Area Quantity (Equation 7.8) 7.2.3.2 Conventional Form Concrete Pavement: Observed Production Rates and Configuration Conventional form concrete pavement observations were divided into two categories according to Configuration. The first category was for the Concrete The second A Pavement Operation that included sharp angle(s) or curve(s). category was for the Concrete Pavement without any curve or sharp angle. total of twenty data points were observed. points. Each category has ten observed data The t-test was employed to test the difference in mean Production Rate between the two categories, since the two groups are independent and both groups 197 are normally distributed (Appendix Y-2). Table 7.8 presents the results of the group variances test and the t-test. A P-value of 0.6 of homogeneity of variances test indicated that two groups had equal variances at the 95% confidence interval. Based on the assumption of equal variances between two groups, the P-value of the t-test was 0.000 which is less than 0.05. Therefore, it can be concluded that the average Production Rates of the Conventional form concrete pavement construction are different between the two categories at the 95% confidence interval. Table 7.9 shows that the average Production Rate for Conventional form concrete pavement construction is 420 SY/Crew Day for the Configuration without any curve or sharp angle and 192 SY/Crew Day for the Configuration with curve(s) or sharp angle(s). The difference of average Production Rate between the two categories is 228 SY/Crew Day. Table 7.8 Results of Group Variances Test and T Test for Conventional Form Concrete Pavement: Configuration Homogeneity of Group Variances Test P value Test Equality of Group Variances Independent-Samples T Test P value Test Equality of Means among Groups 0.000 0.6 198 Table 7.9 Numbers of Data Points and Mean Production Rate for Conventional Form Concrete Pavement: Configuration Number of Data Points Configuration without any curve or sharp angle Configuration with any curve or sharp angle 10 10 Mean Production Rate (SY/Crew Day) 420 192 7. 3 CORRELATIONS TESTING OF DRIVERS The Work Area Quantity and Course type in Hot mix asphalt pavement construction were not highly correlated according to results of the correlations test shown in Table 7.10. Therefore, the effects of the two drivers on Production Rates of Hot mix asphalt pavement should be considered simultaneously during the estimation of Production Rates. Table 7.10 Correlations Test for Work Area Quantity and Course Type of Hot Mix Asphalt Pavement Construction Course Type Logarithmic Transformation of Work Area Quantity 0.165 0.367 Pearson Correlations Sig. (2-tailed) Table 7.11 indicated a high correlation between Work Area Quantity and Length of Work Area in Slip-form concrete pavement construction. Therefore, 199 the effects of Work Area Quantity and Length of Work Area should not be considered at the same time during estimation of Production Rates. Table 7.11 Correlations Test for Work Area Quantity and Length of Work Area of Slip-form Concrete Pavement Construction Logarithmic Transformation of Length of Work Area Logarithmic Transformation of Work Area Quantity 0.951 ** Pearson Correlations Sig. (2-tailed) 0.000 **. Correlation is significant at the 0.01 level (2 tailed) Table 7.12 shows the results of the correlations test between Work Area Quantity and Configuration for Conventional form concrete pavement. The correlation of -0.391 indicated that the Configuration is not highly correlated with Work Area Quantity. Therefore, the effects of both drivers should be considered together during estimation of the Production Rate for Conventional form concrete pavement. Table 7.12 Correlations Test for Work Area Quantity and Configuration of Conventional Form Concrete Pavement Construction Configuration Logarithmic Transformation of Work Area Quantity -0.391 0.108 Pearson Correlations Sig. (2-tailed) 200 7.4 EFFECTS ON MULTIPLE DRIVERS ON PRODUCTION RATES According to the required sample size for regression analysis and the assumption of independent variables, only Hot mix asphalt pavement was eligible for further multiple regression analysis. 7.4.1 Hot Mix Asphalt Pavement: Production Rates by Logarithmic Transformation of Work Area Quantity and Course Type In the multiple regression analysis for Hot mix asphalt pavement, the dependent variables are estimated Production Rates and the independent variables are logarithmic transformation of Work Area Quantity and Course type. The data for Course type were recoded as binary data. Data for Base course construction were recoded as 1, and Surface course construction were recoded as 0. Table 7.13 displays the results of the multiple regression analysis. The fitted model, shown as Equation 7.9, is statistically significant at the 95% confidence interval. The R2 and adjusted R2 are 0.488 and 0.452 respectively. Production Rate = -1263 + 269 Log (Work Area Quantity) 181 (Course Type) (Equation 7.9) 201 Table 7.13 Multiple Regression Model for Hot Mix Asphalt Pavement R2 Adjusted R2 Standard Error Regression Model Variable Log (Work Area Quantity) Course Type (Constant) F 13.35 B 269 181 -1263 0.488 0.452 263 P value 0.000 P value 0.000 0.095 0.006 The plots used to check for violation of assumptions are displayed in Appendix Z. No violation was found. Thus, this model was statistically significant and the effects of Work Area Quantity on Production Rate are established herein. This model is only applicable to Work Area quantities within the range of 227 tons to 5,840 tons, since this model was developed based on the observed data that had this range. Therefore, the estimated Production Rates for Surface course construction of this model can range from 196 Tons/Crew Day to 1,070 Tons/Crew Day, and for Base course construction can range from 377 Tons/Crew Day to 1,251 Tons/Crew Day. The effects of Work Area Quantity on the Production Rate of Hot Mix Asphalt Operations can be computed by differentiation of the fitted multiple regression model, as shown in Equation 6.28. Therefore, when two Work Areas have a quantity of Hot mix asphalt pavement 202 close to 1,000 tons but differ by 100 tons in Work Area Quantity, they may experience a difference about 27 Ton/Crew Day in average Production Rate. The effects of Course type on Production Rate of Hot mix asphalt pavement are 181 Ton/Crew Day when the two Work Areas have the same Work Area Quantity. d( Production Rate) 269 = (Equation 7.10) d( Work Area Quantity) Work Area Quantity 7.5 SUMMARY OF FINDINGS ON DRIVER ANALYSES Table 7.14 summarizes the results of driver analysis for Pavement-related Work Items. Project type was not analyzed due to insufficient data. None of the investigated Candidate Drivers at the project- and Work Zone- level was found to significantly affect Production Rate. For Candidate Drivers at the Work Item-level, Work Area Quantity was identified as a driver for the three targeted Pavement-related Work Items. Length of Work Area was identified as a Production Rate driver of Slip-form concrete pavement. In addition, Course type is a Production Rate driver of Hot mix asphalt pavement, and Configuration is a Production Rate driver of Conventional form concrete pavement construction. Table 7.15 lists the drivers identified as having a significant Production Rate impact for major Pavement-related Work Items. A multiple regression model 203 was developed for Hot mix asphalt pavement to determine the interaction effects of the identified drivers. For Slip-form concrete pavement, multiple regression analysis was not applicable because a high correlation was found between its drivers. For Conventional form concrete pavement, insufficient data limited the development of a multiple regression model. Table 7.14 Summary of Results of Driver Analyses Candidate Drivers Project Type Project Level Project Location Traffic Flow Project Complexity Accelerated Construction Provision Contractor Management Skill Work Zone Level Work Zone Accessibility Work Zone Congestion Work Zone Land Slope Work Area Quantity Length of Work Area Work Item Level Thickness Width of Work Area Course Type Main Lane vs. Non-main Lane Type of Concrete Pavement Configuration : Driver found to be statistically significant : Investigated but not statistically significant : Insufficient data for analysis Hot Mix Asphalt Slip-from Concrete Conventional Form Pavement Pavement (CRCP) Concrete Pavement 204 Table 7.15 Summary of Identified Production Rate Drivers Work Item Drivers Work Area Quantity Course Type Slip-form Concrete Pavement Work Area Quantity Length of Work Area Conventional Form Concrete Pavement Work Area Quantity Configuration Type of Regression Model used for Analysis Log model ******** Log model Log model Log model ******** Regression Analysis/T Test R2= 0.432 P=0.093 R = 0.653 R = 0.356 R = 0.511 P<0.001 2 2 2 Multiple Regression Analysis R = 0.488 Adjusted R = 0.452 *None 2 2 Hot Mix Asphalt Pavement **None *None: Because the two drivers are highly correlated, multiple regression analysis is not applicable. **None: The sample size was not sufficient for multiple regression analysis. 205 CHAPTER VIII: CONCLUSIONS OF THIS RESEARCH The purpose of this study was to investigate realistic Production Rates, and to identify drivers known at the design stage which influence Production Rates as well as to quantify their effects for seven Work Items in Earthwork and Pavement construction. Based on the findings in this study, the highway construction industry can improve the reliability of their Production Rates database, which should lead to more reasonable Contract Time estimation. Since the drivers discussed in this study were those that should be available at the design stage, more accurate estimation should be possible. 8.1 CONCLUSIONS Except for Flexible base, the observed Production Rates collected from ongoing projects were significantly different from the Production Rates in the CTDS. The CTDS Production Rates for most major Work Items in Earthwork and Pavement construction were found to be too optimistic. Sizable variation of observed Production Rates leads one to believe that Production Rate estimation can be far from realistic rates without consideration of statistically Significant Drivers. The Production Rate data found from historical records did not contain sufficient information to explain the variation in Production Rates. The drivers significantly influencing Production Rates of the seven targeted Work Items in Earthwork and Pavement construction were identified and their 206 effects were quantified. Table 8.1 displays the identified significant Production Rate drivers for seven targeted Work Items as well as the sensitivity factors considered in the CTDS. Work Area Quantity was found to have a positive relationship with the Production Rates of all seven targeted Work Items in Earthwork and Pavement construction. One reason could be that all seven targeted Work Items are highly repetitive in their nature. When a Work Area involves a large quantity of work, Production Rates are higher due to learning effects. Another reason could be that contractors are more willing to contribute more effort on engineering and more resources to a larger quantity of work in order to reduce total cost and construction time. A final reason could be that productive hours per working day are higher when a Work Area involves a large quantity. Although the CTDS study indicates that Soil Condition, Location and Traffic Condition are sensitive to the respective Work Items (refer to Table 8.1), this research did not find that they were statistically significant. For Lime-treated sub-grade, Aggregate base, and Slip-form concrete pavement, the results of correlation tests of identified drivers showed that Length of Work Area was highly correlated with Work Area Quantity. longer Work Area implies larger Work Area Quantity. It is obvious that Therefore, the effects of Work Area Quantity and Length of Work Area should not be considered concurrently during Production Rate estimation. 207 Table 8.1 Work Items vs. Significant Drivers of this Research and the CTDS Research Targeted Work Items Earth Excavation Significant Drivers Work Area Quantity Work Area Quantity Work Zone Congestion Work Items Earth Excavation CTDS Sensitivity Factors Quantity of Work Soil Condition Quantity of Work Soil Condition Quantity of Work Soil Condition Quantity of Work Location Quantity of Work Soil Condition Quantity of Work Location Quantity of Work Traffic Condition Quantity of Work Concrete Paving Location Embankment Embankment Work Area Quantity Lime Ttreated Sub-grad Length of Work Area Flexible Base Work Area Quantity Length of Work Area Work Area Quantity Length of Work Area Work Area Quantity Hot Mix Asphalt Pavement Course Type Slip-from Concrete Pavement Conventional Form Concrete Pavement Work Area Quantity Length of Work Area Work Area Quantity Configuration Lime Stablization Flexible Base Material Cement Treated Base Material Hot Mix Asphalt Base Hot Mix Asphalt Surface Cement Treated Base Course type is a Production Rate driver of Hot mix asphalt pavement construction. The difference in quality requirements of Base courses and Surface courses could be the main reason for the difference in average Production Rates between the two Course types. Surface courses are usually constructed with a higher standard of quality than Base courses. To reach a higher standard 208 of quality control, slower paving speed in order to achieve more precise grading and compacting is applied to Surface course construction. Configuration of concrete pavement was a significant Production Rate driver on Conventional form concrete pavement construction. Curve(s) or sharp angle(s) increases the technical complexity for formwork and rebar installation. This is the most likely reason for lower Production Rates for the Conventional form concrete pavement with any curve or sharp angle. Once the drivers of Production Rates for the seven targeted items were identified and their effects were found, their correlations were also explored. Based on the findings in this study the Production Rates of the seven targeted Work Items could be estimated. application. First, the effects of drivers on Production Rates should only be used as a reference to estimate Production Rates. The effects on Production Rates from However, there are some limitations to the multiple drivers are not equivalent to the sum of the effects of each driver. Therefore, designers should carefully evaluate the combined effects of all drivers during Production Rate estimation. Second, because of the presence of a maximum Production Rate, a non-linear model was more appropriate than a linear model to model Production Rates for construction activities. Furthermore, the limited number of data points may not 209 be representative for all applications. must be recognized. Therefore, the limitations of the model Third, this study is limited to Production Rate estimation for Contract Time estimation. In other words, the scope for measuring Production Rates would be different for cost management purposes. 8.2 RECOMMENDATIONS The Production Rate data of seven major earthwork-and pavement-related Work Items have been collected for this study. Future research should collect data on additional TxDOT Work Items frequently on the critical path, such as rock excavation, and concrete curb and gutter. In addition, a reliable Production Rate database should be created to facilitate Contract Time estimation. Significant Drivers known at the design stage were identified for each targeted Work Item. Future studies should seek to better understand remaining sources of Production Rate variability such as weather impact and operator skill. Moreover, lead and lag time information should be investigated to enhance information for time-estimating. 210 APPENDICES 211 Appendix A. Questionnaire for Selecting Work Items for the Study Name District Site/Office Address Phone Number : _________________ : _________________ : _________________ Position : _____________________ :__________________________________________________________ E-mail Address : _____________________ Please check as you think it is most appropriate Pay Items Definitely Track? Yea/No Initial traffic control Detour ROW Preparations Clear & Grub Remove old structure(small) Remove old pavement Remove old curb & gutter Remove old sidewalks Major structure demolition Yes/No Yes/No Yes/No Yes/No Yes/No Low Moderate High Low Moderate High Low Moderate High Low Moderate High Low Moderate High Low Moderate High Low Moderate High Rarely Rarely Rarely Rarely Rarely Rarely Rarely Sometimes Sometimes Sometimes Sometimes Sometimes Sometimes Sometimes Often Often Often Often Often Often Often Usually Usually Usually Usually Usually Usually Usually Yes/No Yes/No Degree of Variability in Crew Productivity Low Moderate High Low Moderate High Low Moderate High How often On or Near Critical Path Rarely Sometimes Often Usually Rarely Rarely Sometimes Sometimes Often Often Usually Usually Remove old drainage/utility structures Yes/No Yes/No 212 Appendix A. Questionnaire for Selecting Work Items for the Study (Cont'd) Pay Items Excavation/embankment Earth excavation Rock excavation Embankment Drainage structures/storm sewers Pipe Box culverts Inlets & Manholes Bridge Structures Erect temporary bridge Bridge demolition Cofferdams Piling Footings Columns, caps & bents Wingwalls Yes/No Yes/No Yes/No Yes/No Yes/No Yes/No Yes/No Low Moderate High Low Moderate High Low Moderate High Low Moderate High Low Moderate High Low Moderate High Low Moderate High Rarely Rarely Rarely Rarely Rarely Rarely Rarely Sometimes Sometimes Sometimes Sometimes Sometimes Sometimes Sometimes Often Often Often Often Often Often Often Usually Usually Usually Usually Usually Usually Usually Yes/No Yes/No Yes/No Low Moderate High Low Moderate High Low Moderate High Rarely Rarely Rarely Sometimes Sometimes Sometimes Often Often Often Usually Usually Usually Yes/No Yes/No Yes/No Low Moderate High Low Moderate High Low Moderate High Rarely Rarely Rarely Sometimes Sometimes Sometimes Often Often Often Usually Usually Usually Definitely Track? Degree of Variability in Crew Productivity How often On or Near Critical Path 213 Appendix A. Questionnaire for Selecting Work Items for the Study (Cont'd) Pay Items Definitely Track? Yea/No Beams (erection only) Bridge deck (total depth) Bridge curb/walk Bridge handrail Remove temporary bridge Retaining walls Base Preparations Lime stabilization Flexible base material Cement treated base material New curb & gutter Hot mix asphalt base Concrete paving Hot mix asphalt surface Yes/No Yes/No Yes/No Yes/No Yes/No Yes/No Yes/No Low Moderate High Low Moderate High Low Moderate High Low Low Low Low Moderate Moderate Moderate Moderate High High High High Rarely Rarely Rarely Rarely Rarely Rarely Rarely Sometimes Sometimes Sometimes Sometimes Sometimes Sometimes Sometimes Often Often Often Often Often Often Often Usually Usually Usually Usually Usually Usually Usually Yes/No Yes/No Yes/No Yes/No Yes/No Yes/No Degree of Variability in Crew Productivity Low Moderate High Low Moderate High Low Moderate High Low Moderate High Low Moderate High Low Moderate High Low Moderate High How often On or Near Critical Path Rarely Sometimes Often Usually Rarely Rarely Rarely Rarely Rarely Rarely Sometimes Sometimes Sometimes Sometimes Sometimes Sometimes Often Often Often Often Often Often Usually Usually Usually Usually Usually Usually 214 Pay Ite ms Appendix A. Questionnaire for Selecting Work Items for the Study (Cont'd) Definitely Degree of Variability How often On or Near Track? in Cre w Productivity Critical Path Yea/No Permanent signing & traffic signals Small signs Overhead signs Major traffic signals Seeding & Landscape Permanent pavement markings Final clean up Others __________________________ __________________________ Yes/No Yes/No Yes/No Yes/No Yes/No Yes/No Yes/No Yes/No Yes/No Low Moderate High Rarely Sometimes Often Usually Low Low Low Low Low Low Moderate High Moderate High Moderate High Moderate High Moderate High Moderate High Rarely Rarely Rarely Rarely Rarely Rarely S ometimes S ometimes S ometimes S ometimes S ometimes S ometimes Often Often Often Often Often Often Usually Usually Usually Usually Usually Usually Low Low Low Moderate High Moderate High Moderate High Rarely Rarely Rarely S ometimes S ometimes S ometimes Often Often Often Usually Usually Usually ________________________ Your Comment (We appreciate your comment) Are you interested in continued participation in this study? Thank You. Yes No 215 Appendix B. Results of the Survey for Selecting Work Items to be Tracked Results of the Survey for Selectiing Work Items to be tracked Work Items Initial traffic control Detour ROW Preparations Clear & Grub Remove old structure(small) Remove old pavement Remove old curb & gutter Remove old sidewalks Remove old drainage/utility structures Major structure demolition Definitely Track? - 'Yes' Response Bob. H. Carlos C. Doug W. Dan D. Mike L. Yes Yes Yes Yes Harry P. Mario R.G. Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes David H. Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Pat W. Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Mike B. Yes Yes Duane S. Yes Yes Tom N. Yes Yes Mike C. Yes Yes Yes Total of 'Yes' 7 8 6 4 6 3 3 Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes 5 9 9 6 10 8 10 8 3 7 4 Excavation/embankment *Earth excavation Rock excavation *Embankment Drainage structures/storm sewers Yes Yes Yes Pipe Box culverts Yes Yes Yes Yes Inlets & Manholes Bridge Structures Yes Yes Yes Yes Yes Erect temporary bridge Bridge demolition Cofferdams Piling Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Footings Columns, caps & bents Yes Yes Yes Yes Yes 9 7 10 7 9 10 9 3 2 4 Wingwalls Beams (erection only) Bridge deck (total depth) Bridge rail Yes Yes Bridge curb/walk Bridge handrail Remove temporary bridge Retaining walls Base Preparations *Lime stabilization *Flexible base material *Cement treated base material Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes 40 Drill Shaft/ Surface Treatment 10 7 9 7 7 9 9 10 3 Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes New curb & gutter *Hot mix asphalt base *Concrete paving *Hot mix asphalt surface Permanent signing & traffic signals Small signs Overhead signs Major traffic signals Seeding & Landscape Permanent pavement markings Final clean up Total of 'Yes' Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes 0 0 35 Utility Installation /adjustment 0 Yes Yes Yes 36 Yes Yes Yes Yes Yes 37 Yes Yes Yes 25 Yes Yes 31 Yes Yes Yes 25 Yes Yes Yes 22 7 7 3 9 6 Yes Yes 17 21 Traffic Switches, Temporary Striping, CTB Move & Reset Others Planning Hot Mix Pav't Drill Shaft 216 Appendix C. Project-Level Data Collection Tool Production Rate Tracking : Project level CCSJ # : : Highway # : Station Range City/County : : Million --( Calandar/Working days) Project ID: Project Length District : Prime Contractor: % of Project Completion : Work Items to be tracked: % Contract Amount : $ Project(Construction) Period : Item # Work Item Unit Approx. Total Quantity Scheduled Start Date Scheduled End Date Sub- Contracted? Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes No No No No No No No No No No Comments Please, fill out next page. 217 Appendix C. Project-Level Data Collection Tool (Cont'd) Project Level Variables Evaluation Project CCSJ: Variable Project Type Unit Seal Coat Overlay Optional Values Rehabilitate Existing Road New Location Freeway Bridge Replacement/ New Bridge Metro Most hours congested > 20 K > 40" North Texas 70-100 20M ~ 50 M Complex >50M Panhandle & West Texas Convert NonFreeway to Freeway New Location Non-Freeway Upgrade Freeway to Standards Upgrade NonFreeway to Standards Widen Freeway Widen NonFreeway Bridge Widening/ Rehabilitation Urban Only rush hours congested 5 K ~ 20 K 15"~40" Central & South Texas 30-70 5M ~ 20 M Moderate Working Day Incentive Using Contract Administrative Cost Lane Rental Disincentive 300~3K Stiff Moderate Moderate 4' ~ 10' Interchanges Location Traffic Flow Traffic Count (ADT) Annual Precipitation Weather Winter Season Length % of Construction Completion at 1st Data Collection Date Size : Construction Contract Amount Technical Complexity Contract Day % $ Costal 0-30 <5M Simple Calendar Day Veh./ Day /Year Rural Rarely congested <5K < 15" None Contract Accelerated Constrcution Provision Substantial Completion I/D Liquidated damages Soil types Clay Content (Plastic Soils) Local site Land Slope Drainage Effectiveness Water Table Depth below Grade Scheduling Technique Used Days per Week (typical) Day/Week $/Day < 300 Loose Low Flat < 4' Bar Chart 4 8 C.I.S. Good Milestones with Incentives/ Disincentives A+B Provisions 3K~6K Rocky High Steep > 10' CPM (Resource-loaded) 7 2 Shifts 6K~12K > 12K CPM (Not Resource-loaded) 5 10 Site Mgmt. Average Poor 6 12 Work Schedule Hours per Day(typical) Hours/Day Contract Admin. System CMS (Contractor Management Skill) 218 Appendix D. Work Zone & Work Item -Level Data Collection Tool Production Rate Tracking: Work Zone Level Work Zone & Work Item Assessed Project ID: ___________________ Work Zone Description/Sketch: Description: _________________________________________________________ _________________________________________________________ Typical Workday Start Time: _________________________ Typical Workday Stop Time: _________________________ Is observed Work Item on critical path? Workers are from: Union Yes No No. indicates the No. of Traffic lines Double line indicates that WZ is not affected at all by its side of traffic. Sketch District: ______________________ Work Item (No.): ________________________________________________________________ Non-Union How much quantity included in the Work Area: ________________________________________________________ Work Zone Level Variables Evaluation Variable 1 WZ Accessibility WZ Construction Congestion Work Zone Site Drainage Effectiveness Clay Content in Soil Difficult Characterization Moderate Easy Not applicable Comment 2 3 3.1 Severe Easily Flooded High Moderate Moderate Moderate Minor Quickly Drained Low Not applicable Not applicable Not applicable 3.2 Land Slope Water Table Depth Below Grade Steep Moderate Flat Not applicable 3.3 <4' 4'~10' >10' Not applicable Check if data Collection completed 219 Appendix D. Work Zone & Work Item -Level Data Collection Tool (Cont'd) Observation Record Date: Production Rate Tracking: Work Item Level Recorder: _______________ Approximate % of Completion Completion Status: Fully Labeled Sketch and Description Quantity Completed Unit 220 Appendix D. Work Zone & Work Item -Level Production Observation Record Data Collection Tool (Cont'd) Rate Tracking (Work Item Level) Resource Efforts for Work Item Crew Crew Type Novice Novice Novice Novice Average Skill Level Typical Typical Typical Typical Experienced Experienced Experienced Experienced Crew Size Equipment Equipment Piece Equipment Size Number in Operation Note: #1~#5 is the observation number. 221 Appendix D. Work Zone & Work Item -Level Data Collection Tool (Cont'd) Production Rate Tracking Calendar (Work Item Level) Sunday Monday II VI Tuesday Wednesday Thursday Friday Saturday / I III / / / / / / / / / / / / / / / / / / / / / / / / / / / / / or / / / and / / , VI: Comment No. I: Observation #, II: X, , III: Indication except X, Total Working Days: _______________ Indication H: Holiday or Day Off T - #: This Observation # W: Weather day (< 2 Hrs of work) S: Work Day With Some Weather Effect N: UNworkable Soil Condition I: Incomplete Crew E: Equipment Downtime/not Available M: Material Unavailable U: Utility Conflicts F: UnForeseen Condition C: Construction Accident A: Traffic Accident O: Overtime D: Other Delay (specify in comments) : Working Day with : Normal Working Day X: Non Working Day Delay Comments: 1 2 222 Appendix D. Work Zone & Work Item -Level Data Collection Tool (Cont'd) Production Rate Tracking Calendar (Work Item Level) Comments (Continued): 3 4 5 6 7 8 9 10 11 12 General Comment 223 Appendix E. Work Item Sheets Work Item Excavation Sub-Item Earth Excavation Included - Work Item # 110 Unit of Measurement CY/Crew Day SCOPE Removing top soil Not Included Survey & Layout Access road construction and maintenance Unsuitable material replacement Reshaped by blade and then sprinkled and rolled for sub-grade surface (about 6" depth) Temporary drainage maintenance Shaping slop Rock Excavation from original elevation to the elevation which is at least 6" below the required sub-grade elevation Disposal of material PRODUCTIVITY FACTOR (Work Item) Construction Type(Haul to Waste, Cut to Fill), (Note:_________________________) Haul road distance (Specify: ______________________________________________) - - Equipment number/Equipment size/Soil Type/Clay content in soil - Remove top soil. Starting NODE Starting the excavation of any working phase. Sub-grade surface is completed. Reach the anticipated elevation of the working phase Ending - - Equipment: 1 Excavator (2CY Bucket), Trucks (Number is according to the distance from disposal field to Work Zone and traffic condition.) A Crew Definition Comments; Verified ________ Node; In a special case, a data collector can judge the Starting and the Ending Node based on his/her professional experience. 224 Appendix E. Work Item Sheets (Cont'd) Work Item Embankment Sub-Item Embankment Included - Work Item # 132 Unit of Measurement CY/Crew Day SCOPE Not Included Survey & Layout Constructing access road Temporary drainage maintenance (Construction of roadway embankments, levees and dykes or any designated section of the roadway) Placing materials Spread material Sprinkling Compaction PRODUCTIVITY FACTOR (Work Item) - Material Type (Type A, Type B, Type C, Type D), (Note: _________________) Density Requirement (Ordinary Compaction, Density Control) Construction Type(Borrow to Fill, Cut to Fill), (Note: ________________________) Slope (Steep, Moderate, Flat), (Note: _______________________________________) - - Equipment number/Equipment size/Work Zone Congestion/Clay content in soil/Work Zone drainage effectiveness Starting NODE Ending - Place the first load of embankment material. Sub-grade surface is completed. . Reach the elevation of the working phase if there are more than one phases of embankment A Crew Definition Equipment: 1~2 Dozer, 1 Compactor Comments; Verified ________ Node; In a special case, a data collector can judge the Starting and the Ending Node based on his/her professional experience. 225 Appendix E. Work Item Sheets (Cont'd) Work Item Lime-Treated for materials used as sub-grade Sub-Item Lime-Treated for materials used as subgrade Included - Work Item # 260 Unit of Measurement SY/Crew Day SCOPE Cutting & pulverizing Spread Lime Mixing Sprinkling or aerating Compaction Finishing 1ST curing and 2nd mixing Not Included Survey & layout Equipment move in Transport material Curing (after finishing) Density tests Setup blue top PRODUCTIVITY FACTOR (Work Item) Number of Mixing (Specify: _________________________________________) Lift Height (Specify: ________________________________________________) Type C Lime Used (Yes, No) (Note: ___________________________________) Total Length Ready For Work (Specify: _______________________________) Average Width of Work Area (Specify: ________________________________) Slope (Steep, Moderate, Flat), (Note: __________________________________) - - Work Zone Congestion/Soil Type/# of working days only for curing/# of nonworking days on curing Starting NODE Ending - Spread lime or cut & pulverize sub-grade. Finishing sub-grade surface is completed. Equipment: 1 Stabilizer, 1 Motor Grader, 1 or 2 Spreader, 1 Sheep-foot Roller, 1 Flat Roller A Crew Definition Comments; Verified ________ Node; In a special case, a data collector can judge the Starting and the Ending Node based on his/her professional experience. 226 Appendix E. Work Item Sheets (Cont'd) Work Item Aggregate Base Course Sub-Item Aggregate Base Course Included Work Item # 247, 262, 263, 275, 276 Unit of Measurement LIFT-SY/Crew Day SCOPE Placing materials Not Included Survey & layout Shaping the sub-grade or existing roadbed Stockpiled All material tests excluded Curing (Flexible Base: Directed by Engineers, usually 2 days; CTB: 72 hours) Density tests Rework caused by failing to achieve required density Spread uniformly & shaping Blade & shaping Sprinkle Compact Dry-out (if required) PRODUCTIVITY FACTOR (Work Item) - Lift Height (Specify: ___________________________) Total Lift Length (Specify: _____________________) Average Width (Specify: _______________________) Number of Lifts (Specify: ______________________) Type of treatment (None, Lime treatment, Portland Cement), (Note: ________) Treatment Mixing Method (Plant mixing, Roadway mixing), (Note: _________) Slope (Steep, Moderate, Flat), (Note: ___________________________________) - Location/ Soil Type/ Work Zone Congestion Starting NODE Ending - Place the first load of base material. Finishing a lift of base course is completed. A Crew Definition Equipment: 1 Motor Grader, 1~2 Steel Roller, 1 Water Truck, Trucks (Number is according to the distance from Work Zone to material resource) Comments; Verified ________ Node; In a special case, a data collector can judge the Starting and the Ending Node based on his/her professional experience. 227 Appendix E. Work Item Sheets (Cont'd) Work Item Hot Mix Asphalt Sub-Item Hot Mix Asphaltic Concrete Pavement, Asphalt Stabilized Base Included Work Item # 340, 345 Not Included Survey and layout Transport materials Cleaning surface before applying for tack coat Shoot tack coat (if tack coat required) Mixing materials in the plant Equipment setup Unit of Measurement Ton/Crew Day SCOPE Lay Hot Mix Asphalt Compaction (Roller or lightly oiled tamps) PRODUCTIVITY FACTOR (Work Item) - Thickness of Lifts (Specify: _________________________________________) (Bond Breaker, Base Course, Surface) construction, (Note: ______________) Asphalt Plant Capacity (Production Rate) (Specify: _______________tons/hr) (Machine Laid, Blade Laid), (Note: __________________________________) Slope (Steep, Moderate, Flat), (Note: _________________________________) - Traffic Condition/ Location Starting NODE Ending - Place the first load of Hot Mix Asphalt material. - Complete compaction. A Crew Definition Labors: One crew (6-8) Equipment: 1 Lay down Machine, 1 Pneumatic Roller, 5 Trucks Comments; Verified ________ Node; In a special case, a data collector can judge the Starting and the Ending Node based on his/her professional experience. 228 Appendix E. Work Item Sheets (Cont'd) Work Item Concrete Paving Sub-Item Slip-form Included Work Item # 360-1 Not Included Survey & Layout Surface preparation Equipments move in Ride quality test Core test Unloading reinforcing steel Curing Saw cutting Unit of Measurement SY/Crew Day SCOPE Setting string line Placing dowels Installing reinforcing steel Placing joint assemblies Initial equipment setup Placing concrete Finishing PRODUCTIVITY FACTOR (Work Item) Starting NODE Ending - (Continuously reinforced concrete pavement, Jointed concrete pavement, Nonreinforced concrete pavement), (Note: ________________________________) Thickness of Concrete Pavement (Specify: _____________________________) Total Length Ready for Slip (Specify: _________________________________) Width of Pass (Specify: _____________________________________________) Number of Moving Slip-Form Paver (Specify: _________________________) Quantity of Concrete Poured (Specify: ________________________________) Slope (Steep, Moderate, Flat) (Note: __________________________________) Location Set string line. Complete concrete placement. Labors: One crew for reinforcing bar (8-10), One crew for concrete feeding and placing (6-8) Equipment : 1 Slip-form Paver, 1 Material Transfer A Crew Definition Comments; Verified ________ Node; In a special case, a data collector can judge the Starting and the Ending Node based on his/her professional experience. 229 Appendix E. Work Item Sheets (Cont'd) Work Item Concrete Paving Sub-Item Conventional Handform Included Work Item # 360-2 Not Included Survey & Layout Surface preparation Cutting & bending Reinforcing steel Core test Curing Removing formwork Unit of Measurement SY/Crew Day SCOPE Formwork Installing reinforcing steel Placing concrete Spread and finishing PRODUCTIVITY FACTOR (Work Item) - Spread roller used (Yes, No), (Note: ___________________________________) Slope (Steep, Moderate, Flat) (Note: ___________________________________) Shape (Simple Configuration, With any Curve and Sharp Angle)(Note: _____) - Crew size/Work Zone congestion Start to setup formwork Starting NODE Ending - Complete concrete placement. A Crew Definition Labors: One crew for formwork (3-4), One crew for reinforcing bar (6-8), One crew for concrete pouring (6-10) Comments; Verified ________ Node; In a special case, a data collector can judge the Starting and the Ending Node based on his/her professional experience. 230 Appendix F. Safety Protocol Safety Protocol for Construction Site Visits (TXDOT Project 0-4416) READ, FAMILIZE and OBEY THIS SAFETY PROTOCOL BEFORE SITE VISIT Ensure compliance with all regulations concerning the standard safety procedures of TxDOT and site. Site protocol Arrival: On each and every visit, the GRA must report to field office and gain permission to enter the site. Departure: Report back to the field office on departure. Vacant Sites: If there are no site representatives on site, then access is prohibited. Instructions: GRA must follow any instructions given to them whilst on site, from the site representative or TxDOT personnel. Safety Procedures Responsibility Avoiding accidents: GRA can avoid accidents by concentrating and thinking before acting. Remember that acting on impulse and taking shortcuts causes many accidents. Parking & Transportation: GRA should park near the field office and go to job site with TxDOT personnel. Clothing Safety vest: Wear safety vest all the times in the job site. Hardhats: Wear safety hardhats all the times in the job site. Footwear: Wear steel-toed boots if required. 231 Appendix F. Safety Protocol (Cont'd) Hearing protection: Ear protection should be worn if required. Safety glass: Wear safety glass in required area. Loose clothing: Do not wear loose clothing. Moving around the site Barricades: Do not lean over or go beyond any protective handrails or barricades. Openings: Be careful where you walk. Pay attention to openings, barriers, protective covers and changes in levels. Access: Use correct access at all times. Restricted areas: Keep out of restricted areas. Movement: Running on any part of the site is prohibited. Never walk backwards in a construction area. Do not jump from equipment, platforms or scaffolds. Do not stand or walk under any loads being lifted. Weather: Beware of slippery surfaces (particularly after or during rain). Be careful in windy weather. Behaviors on-site: Restrict communication with workers unless it is necessary for the research. Traffic: Be aware of moving equipment and vehicles. Traffic rules should be obeyed and strict attention should be paid to all warning signs at all times. Taking pictures: GRA can freely take the pictures on the surveyed Work Items unless it is restricted. 232 Appendix G. General Information of Investigated Projects Project ID P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 P11 P12 P13 P14 P15 P16 P17 P18 P19 P20 P21 P22 P23 P24 P25 P26 P27 P28 P29 P30 P31 P32 P33 P34 P35 District D1 D1 D1 D1 D2 D2 D3 D3 D3 D3 D4 D4 D4 D4 D4 D4 D4 D5 D5 D5 D5 D5 D5 D5 D5 D5 D6 D6 D7 D7 D7 D7 D7 D7 D7 Prime GC 11 GC 22 GC 22 GC 8 GC 3 GC 12 GC 14 GC 14 GC 5 GC 2 GC 7 GC 17 GC 6 GC 22 GC 18 GC 22 GC 6 GC 20 GC 22 GC 20 GC 19 GC 20 GC 20 GC 19 GC 13 GC 20 GC 10 GC 1 GC 4 GC 4 GC 16 GC 21 GC 5 GC 15 GC 9 Contract Amount 10.90 29.60 18.90 35.90 20.00 9.00 87.80 50.49 6.80 4.50 8.55 12.00 8.60 75.13 17.00 261.00 8.30 16.10 15.20 9.60 3.77 23.29 80.90 16.60 1.52 104.00 47.00 7.43 9.62 25.36 5.56 3.60 1.35 12.26 20.16 Project Type Widen Nonfreeway Widen Freeway Widen Freeway Upgrade Nonfreeway To Standards Covert Nonfreeway to Freeway Covert Nonfreeway to Freeway Interchanges Covert Nonfreeway to Freeway Widen Nonfreeway Widen Nonfreeway Widen Freeway Widen Nonfreeway Rehailitate Existing Road Upgrade Freeway to Standards Widen Nonfreeway Interchanges Rehailitate Existing Road Upgrade Freeway to Standards Upgrade Freeway to Standards Upgrade Freeway to Standards Bridge Replacement/New Bridge New Location Freeway Upgrade Nonfreeway To Standards Widen Freeway Bridge Widening/Rehibilitation Widen Freeway Covert Nonfreeway to Freeway Rehailitate Existing Road Upgrade Freeway to Standards Upgrade Nonfreeway To Standards Bridge Replacement/New Bridge Widen Freeway Widen Nonfreeway New Location Freeway New Location Freeway Location Urban Metro Metro Urban Urban Urban Metro Metro Urban Rural Urban Urban Urban Urban Rural Metro Urban Urban Urban Urban Rural Urban Metro Rural Rural Urban Urban Rural Urban Rural Urban Urban Urban Rural Rural Contract Amount Range 5 ~ 20 Million 20~50 Million 5 ~ 20 Million 20~50 Million 20~50 Million 5 ~ 20 Million >50 Million >50 Million 5 ~ 20 Million 5 ~ 20 Million 5 ~ 20 Million 5 ~ 20 Million 5 ~ 20 Million >50 Million 5 ~ 20 Million >50 Million 5 ~ 20 Million 20~50 Million 20~50 Million 20~50 Million < 5 Million 20~50 Million 20~50 Million 5 ~ 20 Million < 5 Million >50 Million 20~50 Million 5 ~ 20 Million 5 ~ 20 Million 20~50 Million 5 ~ 20 Million < 5 Million < 5 Million 5 ~ 20 Million 20~50 Million Project Complexity Simple Complex Complex Complex Moderate Moderate Complex Moderate Moderate Simple Moderate Moderate Simple Complex Moderate Complex Simple Moderate Moderate Moderate Moderate Moderate Complex Moderate Simple Moderate Complex Moderate Moderate Complex Moderate Simple Simple Moderate Moderate Accelerated Construction Provision None None None Milestones with Incentives/Discentive None None None None None None None None Substantial Completion I/D None None Milestones with Incentives/Discentive Substantial Completion I/D None None None None None Milestones with Incentives/Discentive None None Milestones with Incentives/Discentive Incentive using contract administrative cost None None None None None None None None Contract Management Skill Good Good Average Average Good Good Average Good Good Good Good Average Good Average Good Good Good Good Good Good Good Good Good Average Average Good Average Average Good Good Average Average Good Good Good 233 Appendix H. Production Rates Data for this Study Excavation DP ID 110001 110002 110003 110004 110005 110006 110007 110008 110009 110010 110011 110012 110013 110014 110015 110016 110017 110018 110019 110020 110021 110022 110023 110024 110025 110026 Project ID P12 P13 P16 P8 P8 P7 P7 P7 P7 P8 P7 P7 P7 P8 P8 P4 P12 P11 P16 P19 P19 P23 P30 P35 P30 P33 Work Zone Accessibility Moderate Difficult Difficult Moderate Moderate Moderate Easy Moderate Moderate Easy Moderate Moderate Easy Easy Difficult Easy Easy Easy Easy Difficult Difficult Moderate Easy Easy Easy Easy Work Zone Congestion Moderate Moderate Severe Minor Minor Moderate Minor Moderate Severe Moderate Moderate Moderate Minor Moderate Moderate Moderate Minor Moderate Minor Moderate Moderate Severe Minor Minor Minor Moderate Work Zone Drainage Quickly Drained Moderate Easily Flooded Moderate Moderate Moderate Quickly Drained Quickly Drained Moderate Quickly Drained Moderate Quickly Drained Quickly Drained Quickly Drained Easily Flooded Quickly Drained Quickly Drained Quickly Drained Quickly Drained Moderate Moderate Quickly Drained Quickly Drained Quickly Drained Easily Flooded Quickly Drained Work Zone Clay Content High Moderate High Low Low Moderate Moderate Moderate Moderate Low Moderate Moderate Moderate Low Low High High Moderate High High High Moderate High Low High Moderate Soil Condition Stiff Stiff Stiff Rocky Rocky Stiff Stiff Stiff Stiff Rocky Stiff Stiff Stiff Rocky Rocky Rocky Stiff Stiff Stiff Loose Loose Stiff Stiff Rocky Stiff Stiff Work Area Quantity (CY) 2601.00 1200.00 970.00 1969.00 4004.00 2472.00 10673.00 1071.00 16798.00 2478.00 7377.00 1766.00 668.00 2198.00 1394.00 13924.00 4302.00 360.00 1064.00 4640.00 2600.00 1536.00 4560.00 7353.00 5237.00 995.00 Production Rate (CY/Crew Day) 1,300.50 600.00 242.50 787.60 2,002.00 618.00 3,557.67 535.50 2,799.67 619.50 922.13 883.00 267.20 628.00 557.60 2,784.80 860.40 360.00 709.33 1,546.67 1,300.00 1,536.00 1,140.00 1,935.00 1,540.29 199.00 234 Appendix H. Production Rates Data for this Study (Cont'd) Embankment DP ID 132001 132002 132003 132004 132005 132006 132007 132008 132009 132010 132011 132012 132013 132014 132015 132016 132017 132018 132019 132020 132021 132022 132023 132024 132025 132026 132027 132028 132029 132030 132031 132032 132033 132034 Project ID P12 P13 P8 P8 P7 P7 P8 P8 P8 P4 P4 P10 P9 P8 P8 P8 P8 P8 P7 P16 P19 P19 P23 P23 P25 P24 P24 P25 P24 P18 P30 P35 P30 P33 Work Zone Accessibility Moderate Difficult Moderate Moderate Difficult Difficult Moderate Moderate Difficult Moderate Moderate Easy Difficult Easy Moderate Moderate Easy Moderate Difficult Easy Difficult Difficult Moderate Moderate Moderate Moderate Moderate Moderate Moderate Moderate Easy Easy Easy Easy Work Zone Congestion Moderate Moderate Minor Minor Moderate Moderate Minor Moderate Moderate Moderate Moderate Minor Severe Moderate Minor Moderate Minor Minor Moderate Moderate Moderate Moderate Minor Minor Minor Minor Minor Moderate Moderate Moderate Minor Minor Minor Moderate Work Zone Drainage Quickly Drained Moderate Moderate Moderate Quickly Drained Quickly Drained Moderate Moderate Quickly Drained Quickly Drained Quickly Drained Moderate Quickly Drained Quickly Drained Quickly Drained Moderate Moderate Quickly Drained Quickly Drained Quickly Drained Moderate Moderate Quickly Drained Quickly Drained Moderate Quickly Drained Quickly Drained Easily Flooded Quickly Drained Quickly Drained Quickly Drained Quickly Drained Easily Flooded Quickly Drained Soil Condition Stiff Stiff Rocky Rocky Stiff Stiff Rocky Rocky Rocky Rocky Rocky Loose Stiff Rocky Rocky Rocky Rocky Rocky Stiff Stiff Loose Loose Stiff Stiff Loose Stiff Stiff Loose Stiff Loose Stiff Rocky Stiff Stiff Work Area Quantity (CY) 2601.00 1200.00 1969.00 4004.00 7377.00 1071.00 2478.00 2198.00 1394.00 10046.00 17838.00 2693.00 1243.00 5728.00 7920.00 2051.00 12936.00 6632.00 1766.00 1064.00 4640.00 2600.00 1536.00 1536.00 3000.00 18753.00 6447.00 1500.00 9261.00 28880.00 4280.00 33938.00 23674.00 3161.00 Production Rate (CY/Crew Day) 325.13 600.00 656.33 2002.00 819.67 535.50 619.50 549.50 464.67 2009.20 849.43 448.83 248.60 1145.60 1584.00 683.67 2156.00 1326.40 588.67 709.33 1546.67 1300.00 1536.00 1536.00 3000.00 1442.54 805.88 750.00 1157.63 1375.24 1070.00 1786.21 1392.59 287.36 235 Appendix H. Production Rates Data for this Study (Cont'd) Lime-Treated Sub-grade DP ID 260001 260002 260003 260004 260005 260006 260007 260008 260009 260010 260011 260012 260013 260014 260015 260016 260017 260018 260019 260020 260021 260022 260023 260024 260025 260026 260027 260028 260029 260030 260031 260032 Project ID P12 P14 P16 P7 P7 P12 P5 P5 P6 P5 P24 P24 P1 P1 P1 P19 P20 P18 P22 P18 P18 P21 P18 P19 P19 P6 P5 P25 P29 P32 P33 P30 Work Zone Congestion Moderate Severe Severe Moderate Moderate Moderate Minor Minor Moderate Moderate Moderate Moderate Moderate Minor Minor Moderate Minor Minor Minor Minor Minor Minor Moderate Moderate Moderate Moderate Moderate Moderate Minor Minor Moderate Minor Work Zone Clay Content High High High Moderate Moderate High Moderate Moderate Moderate Moderate High High Moderate Moderate Moderate High High High High High High High High High High Moderate Moderate Moderate High High Moderate High Work Zone Land Slope Flat Flat Moderate Flat Flat Flat Flat Flat Flat Moderate Flat Flat Flat Flat Flat Flat Flat Flat Flat Flat Flat Steep Flat Flat Flat Flat Moderate Flat Flat Flat Moderate Flat Soil Condition Stiff Rocky Stiff Stiff Stiff Stiff Stiff Stiff Loose Stiff Stiff Stiff Loose Loose Loose Loose Loose Loose Loose Loose Loose Loose Loose Loose Loose Loose Stiff Loose Stiff Stiff Stiff Stiff Work Area Length of Work Width of Work Quantity (SY) Area (LF) Area (LF) 23,010 1,632 409 3,291 5,180 18,449 3,701 3,730 11,026 9,361 31,019 11,165 7,041 10,458 5,553 50,490 17,007 5,758 5,583 7,239 10,167 6,848 5,490 13,104 13,601 26,645 18,463 3,033 7,275 15,569 7,380 12,558 6,472 708 200 456 304 5,301 558 974 3,135 1,758 6,647 3,588 1,647 2,445 990 9,621 4,449 1,151 450 1,303 1,830 1,284 1,247 2,106 2,106 6,002 4,226 900 1,169 3,357 3,087 2,512 32 21 18 53 53 31 60 34 32 48 42 31 38 38 50 47 34 45 111 50 50 48 40 56 56 45 39 30 56 42 22 45 Thickness 6" 6" 6" 6" 6" 6" 6" 6" 6" 6" 6" 6" 6" 6" 6" 6" 6" 6" 6" 6" 6" 6" 6" 6" 6" 6" 6" 6" 6" 6" 6" 6" Type C hydraulic Lime Application No No No No No No No No No No No No Yes Yes Yes No No No Yes No No Yes No No No No Yes No Yes Yes Yes Yes Production Rate (SY/Crew Day) 2,557 233 82 658 740 1,677 1,234 933 1,838 1,872 3,102 3,722 1,760 747 1,111 3,156 2,430 1,440 1,117 1,207 1,695 856 1,098 1,638 1,700 2,961 3,077 758 1,119 865 1,230 1,395 236 Appendix H. Production Rates Data for this Study (Cont'd) Aggregate Base Course DP ID 247001 247002 247003 247004 247005 247006 247007 247008 247009 247010 247011 247012 247013 247014 DP ID 247101 247102 247103 247104 247105 247106 247107 247108 247109 247110 247111 247112 247113 247114 247115 Project ID P25 P21 P18 P19 P19 P18 P18 P18 P20 P18 P24 P19 P24 P20 Project ID P13 P15 P4 P5 P6 P5 P6 P5 P5 P7 P10 P6 P34 P30 P30 Type of Base Material Cement Treated Base Cement Treated Base Cement Treated Base Cement Treated Base Cement Treated Base Cement Treated Base Cement Treated Base Cement Treated Base Cement Treated Base Cement Treated Base Cement Treated Base Cement Treated Base Cement Treated Base Cement Treated Base Type of Base Material Flexible Base Flexible Base Flexible Base Flexible Base Flexible Base Flexible Base Flexible Base Flexible Base Flexible Base Flexible Base Flexible Base Flexible Base Flexible Base Flexible Base Flexible Base Work Zone Congestion Moderate Minor Minor Moderate Moderate Moderate Moderate Minor Minor Minor Moderate Moderate Moderate Minor Work Zone Congestion Minor Minor Moderate Moderate Moderate Moderate Moderate Minor Minor Moderate Minor Minor Minor Minor Minor Work Zone Land Slope Flat Steep Flat Flat Flat Moderate Flat Flat Flat Steep Flat Flat Flat Flat Work Zone Land Slope Moderate Moderate Flat Moderate Flat Moderate Flat Flat Flat Flat Flat Flat Moderate Flat Flat Soil Condition Loose Loose Loose Loose Loose Loose Loose Loose Loose Loose Stiff Loose Stiff Loose Soil Condition Stiff Stiff Rocky Stiff Loose Stiff Loose Stiff Stiff Stiff Loose Loose Rocky Stiff Stiff Work Area Lift-Length of Width of Work Quantity (SY-Lift) Work Area (LF) Area (LF) 3087.00 6601.00 1416.00 16250.00 8211.00 6824.00 6431.00 7827.00 3916.00 4408.00 7319.00 35956.00 31266.00 17796.00 947.00 1324.00 250.00 2125.00 1275.00 770.00 2217.00 1409.00 1137.00 694.00 2114.00 6995.00 3250.00 4449.00 29.00 46.00 51.00 68.00 58.00 75.00 27.00 50.00 31.00 57.00 31.00 46.00 42.00 36.00 Lift-Height (Inch) 6.00 6.00 6.00 6.00 6.00 6.00 6.00 6.00 6.00 6.00 6.00 6.00 6.00 6.00 Lift-Height (Inch) 8.00 3.00 5.50 4.50 6.50 4.50 4.50 4.50 4.50 6.00 6.00 6.50 9.00 6.00 6.00 Production Rate (SYLift/Crew Day) 3087.00 2200.33 1416.00 6500.00 4105.50 3412.00 3215.50 3913.50 3916.00 4408.00 4879.33 5992.67 5211.00 4449.00 Production Rate (SYLift/Crew Day) 526.33 3185.33 2971.93 2505.00 4144.43 2965.89 4953.50 3597.00 2094.33 967.57 1087.43 1245.50 2778.00 5624.29 3178.29 Work Area Lift-Length of Width of Work Quantity (SY-Lift) Work Area (LF) Area (LF) 1579.00 9556.00 41607.00 10020.00 29011.00 26693.00 39628.00 28776.00 6283.00 6773.00 7612.00 2491.00 5556.00 19685.00 22248.00 263.00 1476.00 11371.00 2060.00 6150.00 5950.00 8492.00 7510.00 2520.00 850.00 1054.00 520.00 2000.00 2952.00 4450.00 54.00 58.00 33.00 43.00 42.50 40.00 42.00 34.50 37.50 72.00 65.00 43.00 25.00 60.00 45.00 237 Appendix H. Production Rates Data for this Study (Cont'd) Hot Mix Asphalt Pavement DP ID 340001 340002 340003 340004 340005 340006 340007 340008 340009 340010 340011 340012 340013 340014 340015 340016 340017 340018 340019 340020 340021 340022 340023 340024 340025 340026 340027 340028 340029 340030 340031 340032 Project ID P12 P13 P1 P7 P2 P4 P4 P4 P3 P8 P7 P12 P16 P20 P20 P19 P19 P26 P25 P26 P24 P27 P27 P28 P27 P31 P32 P31 P32 P32 P32 P34 Work Zone Accessibility Easy Easy Easy Easy Easy Easy Easy Easy Easy Easy Easy Easy Easy Easy Easy Easy Easy Moderate Moderate Moderate Easy Easy Moderate Easy Easy Moderate Easy Moderate Moderate Easy Easy Easy Work Zone Congestion Moderate Minor Minor Minor Severe Moderate Severe Moderate Minor Moderate Moderate Moderate Severe Moderate Moderate Minor Moderate Moderate Severe Moderate Moderate Minor Severe Minor Minor Moderate Minor Moderate Moderate Moderate Minor Minor Work Zone Land Slope Flat Flat Flat Flat Moderate Flat Flat Flat Steep Flat Flat Flat Flat Flat Flat Flat Flat Flat Flat Flat Flat Flat Flat Flat Flat Flat Flat Flat Flat Flat Flat Moderate Work Area Quantity (Ton) 5840.00 4500.00 2372.00 3011.00 397.00 694.00 1139.00 2964.00 963.00 1182.00 2723.00 2788.00 783.00 964.00 942.00 1318.00 482.00 614.00 316.00 1062.00 975.00 3010.00 2115.00 2800.00 1519.15 2996.00 1588.00 3268.06 940.00 5828.00 832.00 227.00 Course Type Base Course Base Course Base Course Base Course Surface Surface Surface Surface Base Course Base Course Surface and Base Base Course Base Course Base Course Base Course Base Course Base Course Surface Surface Base Course Base Course Base Course Base Course Surface Base Course Surface Base Course Surface Base Course Base Course Base Course Base Course Main Lane Application Yes Yes Yes Yes No Yes Yes Yes No No No Yes No Yes Yes Yes Yes Yes Yes No Yes No No Yes No Yes No Yes No Yes Yes No Production Rate (Ton/Crew Day) 1460.00 750.00 1186.00 602.20 397.00 347.00 379.67 988.00 963.00 1182.00 907.67 1394.00 783.00 642.67 628.00 659.00 482.00 307.00 158.00 531.00 975.00 1204.00 1410.00 1400.00 759.58 749.00 794.00 1089.35 940.00 777.07 832.00 454.00 238 Appendix H. Production Rates Data for this Study (Cont'd) Slip-form Concrete Pavement DP ID Project ID Type of Concrete Pavment CRCP CRCP JCP CRCP JCP CRCP CRCP CRCP CRCP CRCP JCP CRCP CRCP CRCP CRCP CRCP CRCP CRCP CRCP CRCP CRCP CRCP CRCP Work Zone Accessibility Moderate Moderate Difficult Easy Moderate Easy Easy Difficult Easy Easy Moderate Easy Moderate Moderate Easy Easy Easy Moderate Easy Easy Moderate Moderate Easy Work Zone Congestion Moderate Moderate Severe Minor Severe Moderate Minor Moderate Moderate Moderate Severe Minor Moderate Moderate Moderate Moderate Minor Minor Severe Moderate Severe Moderate Minor Work Zone Land Slope Flat Flat Flat Flat Flat Flat Flat Moderate Flat Moderate Flat Flat Flat Flat Flat Flat Flat Flat Flat Flat Flat Flat Flat Work Area Quantity (SY) 8940.00 10768.00 4088.00 740.00 2152.00 5153.00 38896.00 1600.00 6746.00 1156.00 1984.00 3321.00 4889.00 9873.00 7456.00 9820.00 2000.00 4119.00 18592.00 9754.00 4953.00 3996.00 8361.00 Length of Work Width of Work Lift-Thickness Area (LF) Area (LF) (Inch) 4013.00 4038.00 1533.00 303.00 1614.00 3303.00 15912.00 600.00 3795.00 473.00 1446.00 1624.00 2000.00 4040.00 3050.00 7615.00 1500.00 1685.00 7783.00 4083.00 3547.00 4045.00 6270.00 20.00 24.00 24.00 22.00 12.00 14.00 22.00 24.00 14.00 22.00 12.00 22.00 22.00 22.00 22.00 12.00 12.00 22.00 22.00 22.00 14.00 12.00 12.00 10 10 13 13 8 8 13 13 8 13 13 13 10 10 10 10 10 10 13 13 8 8 8 Production Rate (SY/Crew Day) 1625.45 2153.60 2044.00 740.00 2152.00 1288.25 1994.67 800.00 1124.33 462.40 992.00 1328.40 977.80 1410.43 1242.67 1227.50 1000.00 1373.00 1690.18 1625.67 707.57 999.00 1286.31 360101 360102 360103 360104 360105 360106 360107 360108 360109 360110 360111 360112 360113 360114 360115 360116 360117 360118 360119 360120 360121 360122 360123 P18 P18 P14 P13 P14 P12 P17 P16 P12 P13 P14 P17 P19 P19 P20 P20 P20 P20 P24 P24 P27 P27 P27 239 Appendix H. Production Rates Data for this Study (Cont'd) Conventional Form Concrete Pavement DP ID 360201 360202 360203 360204 360205 360206 360207 360208 360209 360210 360211 360212 360213 360214 360215 360216 360217 360218 360219 360220 Project ID P14 P14 P20 P26 P19 P19 P20 P19 P19 P20 P20 P18 P13 P13 P27 P27 P27 P32 P32 P32 Work Zone Accessibility Difficult Difficult Easy Moderate Easy Moderate Easy Easy Easy Easy Easy Easy Easy Easy Moderate Moderate Easy Easy Easy Easy Work Zone Congestion Severe Severe Moderate Moderate Moderate Moderate Moderate Moderate Moderate Moderate Minor Minor Moderate Minor Severe Severe Moderate Minor Minor Minor Work Zone Land Slope Flat Flat Flat Flat Flat Flat Flat Flat Flat Flat Flat Flat Flat Flat Flat Flat Flat Moderate Moderate Moderate Work Area Quantity (SY) 766.00 919.00 1002.00 2621.00 805.00 736.00 4320.00 614.00 3265.00 1560.00 581.00 423.00 47.00 15.00 914.00 1364.00 992.00 405 329 211 Configuration Typical Typical Non-typical Typical Typical Typical Typical Non-typical Typical Typical Typical Non-typical Non-typical Non-typical Non-typical Non-typical Non-typical Non-typical Non-typical Non-typical Thickness(in) 13.00 8.00 9.00 10.00 13.00 13.00 9.00 6.00 10.00 9.00 9.00 10.00 8.00 8.00 8.00 8.00 8.00 8.00 8.00 8.00 Production Rate (SY/Crew Day) 383.00 459.50 250.50 582.44 536.67 490.67 540.00 307.00 362.78 390.00 290.50 211.50 94.00 30.00 228.50 341.00 248.00 135 164.5 70.33 240 Production Rate (CY/Crew Day) 1000 A A 2000 3000 A A A A A A A Widen Nonfreeway Widen Freeway Rehailitate Existing Road Appendix I-1. Excavation: Scatter Plots of Observed Production Rates vs. Candidate Drivers Project Type Covert Nonfreeway to Freeway 241 A New Location Freeway A A Upgrade Freeway to Standards A A A Upgrade Nonfreeway To Standards A A A A A A A A A Interchanges Appendix I-1. Excavation: Scatter Plots of Observed Production Rates vs. Candidate Drivers (Cont'd) A Produc tion Rate (CY/Crew Day) 3 00 0 A A 2 00 0 A A A A A A A 1 00 0 A A A A A A A A A A A A A Rura l Urb a n M etro Location A Produc tion Rate (CY/Crew Day) 3 00 0 A A 2 00 0 A A A A A A 1 00 0 A A A A A A A A A A A A Rare l y Co ng e ste d M ost ho u rs ve ry co ng e ste d O nl y rush ho u rs co ng e ste d Traffic Flow 242 Appendix I-1. Excavation: Scatter Plots of Observed Production Rates vs. Candidate Drivers (Cont'd) A Produc tion Rate (CY/Cr ew Day) 300 0 A A 200 0 A A A A A A 100 0 A A A A A A A A A A A A A A S im pl e M od era te Com ple x Project Complexity A Produc tion Rate (CY/Cr ew Day) 3 00 0 A A 2 00 0 A A A A A A 1 00 0 A A A A A A A A A A A A None M ilestones with Incentives/Discentive S ubstantial Completion I/D Accelerated Construction Provision 243 Appendix I-1. Excavation: Scatter Plots of Observed Production Rates vs. Candidate Drivers (Cont'd) A Produc tion Rate (CY/Crew Day) 3 00 0 A A 2 00 0 A A A A A A 1 00 0 A A A A A A A A A A A A A A ve ra ge G oo d Contractor Management Skill A Production Rate (CY/Cr ew Day) 3 00 0 A A 2 00 0 A A A A A A A A 1 00 0 A A A A A A A A A A A A A A E asy M od e ra te Diffi cul t Accessibility_W Z 244 Appendix I-1. Excavation: Scatter Plots of Observed Production Rates vs. Candidate Drivers (Cont'd) A Produc tion Rate (CY/Cre w Day) 300 0 A A 200 0 A A A A A A A 100 0 A A A A A A A A A A A A M in or M ode ra te Severe Congestion_W Z A Produc tion Rate (CY/Cre w Day) 3 00 0 A A 2 00 0 A A A A A A A A A A A 1 00 0 A A A A A A A A A Q uickly Drai ns M oderate E asi l y Fl oode d Drainage_W Z 245 Appendix I-1. Excavation: Scatter Plots of Observed Production Rates vs. Candidate Drivers (Cont'd) A Produc tion Rate (CY/Cr ew Day) 3000 A A 2000 A A A A AA A 1000 A A A A A A A A A A A A A A 0 5000 10000 15000 W ork Area Quantity (CY) A Produc tion Rate (CY/Crew Day) 3 00 0 A A 2 00 0 A A A A A A A A A A A A A A A A 1 00 0 A A A L oo se S ti ff Rocky Soil Condition 246 Production Rate (CY/Crew Day) 1000 1500 2000 500 A A A A A A Widen Nonfreeway Widen Freeway Rehailitate Existing Road Appendix I-2. Excavation: Scatter Plots of Observed Production Rates (adjusted by crew size) vs. Candidate Drivers Project Type New Location Freeway 247 A A A A A Covert Nonfreeway to Freeway A A Upgrade Freeway to Standards A A A A Upgrade Nonfreeway To Standards A A A A A A A A A Interchanges Appendix I-2. Excavation: Scatter Plots of Observed Production Rates (adjusted by crew size) vs. Candidate Drivers (Cont'd) 2000 A Production Rate (CY/Crew Day) A 1500 A A A A A 1000 A A A A A A A A A A A A A A A A 500 A A A Rural Urban Metro Location 2000 A Production Rate (CY/Crew Day) A 1500 A A A A A 1000 A A A A A A A A A A A A A A 500 A A A A A Rarely Congested Most hours very congested Only rush hours congested Traffic Flow 248 Appendix I-2. Excavation: Scatter Plots of Observed Production Rates (adjusted by crew size) vs. Candidate Drivers (Cont'd) 2000 A Production Rate (CY/Crew Day) A 1500 A A A A A A 1000 A A A A A A A A A A A A A A A 500 A A Simple Moderate Complex Project Complexity 2000 A Production Rate (CY/Crew Day) A 1500 A A A A A A 1000 A A A A A A A A A A A 500 A A A A None Milestones with Incentives/Discentive Substantial Completion I/D Accelerated Construction Provision 249 Appendix I-2. Excavation: Scatter Plots of Observed Production Rates (adjusted by crew size) vs. Candidate Drivers (Cont'd) 2000 A Production Rate (CY/Crew Day) A 1500 A A A A A 1000 A A A A A A A A A A A A A A A 500 A A A Average Good Contractor Management Skill 2000 A Production Rate (CY/Crew Day) A 1500 A A A A A 1000 A A A A A A A A A A A A A A 500 A A A A Easy Moderate Difficult Accessibility_WZ 250 Appendix I-2. Excavation: Scatter Plots of Observed Production Rates (adjusted by crew size) vs. Candidate Drivers (Cont'd) 2000 A Production Rate (CY/Crew Day) A 1500 A A A A A A 1000 A A A A A A A A A A A A A A A 500 A A Minor Moderate Severe Congestion_WZ 2000 A Production Rate (CY/Crew Day) A 1500 A A A A A 1000 A A A A A A A A A A A A A 500 A A A A Quickly Drains Moderate Easily Flooded Drainage_WZ 251 Appendix I-2. Excavation: Scatter Plots of Observed Production Rates (adjusted by crew size) vs. Candidate Drivers (Cont'd) 2000 A Production Rate (CY/Crew Day) A 1500 A A A A A A A A 1000 A A A A A A A A A A A A 500 A A A A 0 5000 10000 15000 Work Area Quantity (CY) 2000 A Production Rate (CY/Crew Day) A 1500 A A A A A 1000 A A A A A A A A A A A A A A 500 A A A A Loose Stiff Rocky Soil Condition 252 Appendix J. Results of Testing Assumptions of the Regression Analysis for Excavation: Production Rates vs. Work Area Quantity Normal Q-Q Plot Production Rate (CY/Crew Day) 3000 2000 Expected Normal Value 1000 0 -1000 -1000 0 1000 2000 3000 Observed Value Normal Q-Q Plot Log (Work Area Quantity (CY)) 10 9 8 Expected Normal Value 7 6 5 5 6 7 8 9 10 Observed Value 253 Appendix J. Results of Testing Assumptions of the Regression Analysis for Excavation: Production Rates vs. Work Area Quantity (Cont'd) 8 00 A A A A Unstandardized Residuals A 4 00 A A AA A A A A A A A A A A 0 -4 00 A A A -8 00 A 0 5 00 1 00 0 1 50 0 2 00 0 Predicted Production Rates Normal Q-Q Plot Unstandardized Residuals 1000 Expected Normal Value 0 -1000 -1000 0 1000 Observed Value 254 Production Rate (CY/Crew Day) 1 00 0 A 2 00 0 3 00 0 A Bridge W idening/R ehibilitatio n A A A A A A A A Widen Nonfreewa y Widen Freew ay Rehailitate Existing Road A A A A A A A A Appendix K. Embankment: Scatter Plots of Observed Production Rates vs. Candidate Drivers Project Type 255 Covert Nonfreewa y to Freeway New Location Freew ay A A A A A A Upgrade Freew ay to Standards A A A A A Upgrade Nonfreewa y To Standards Interchang es A A A A Appendix K. Embankment: Scatter Plots of Observed Production Rates vs. Candidate Drivers (Cont'd) 3 00 0 A Produc tion Rate (CY/Cr ew Day) A 2 00 0 A A A A A A A A A A A A A A A A A 1 00 0 A A A A A A A A A A A A A Rura l Urb a n M etro Location 3 00 0 A Produc tion Rate (CY/Cr ew Day) A 2 00 0 A A A A A A A A A A A A A A A A A A A A A A A A A A A A 1 00 0 A A Rare l y Co ng e ste d M ost ho u rs ve ry co ng e ste d O nl y rush ho u rs co ng e ste d Traffic Flow 256 Appendix K. Embankment: Scatter Plots of Observed Production Rates vs. Candidate Drivers (Cont'd) 3 000 .0 A Produc tion Rate (CY/Cre w Day) A 2 000 .0 A A A A A A A A A A A A 1 000 .0 A A A A A A A A A A A A A A A A A A S im p l e M od e ra te Com p le x Project Complexity 3 000 A Produc tion Rate (CY/Cr ew Day) A 2 000 A A A A A A A A A A A A A A A A A A A A A A A A A 1 000 A A A Non e M il e sto n es wi th Ince n ti ves/Di sce nti ve Sub stan tial Com pl e ti on I/D Accelerated Construction Provision 257 Appendix K. Embankment: Scatter Plots of Observed Production Rates vs. Candidate Drivers (Cont'd) 3 00 0 .0 A Produc tion Rate (CY/Cre w Day) A 2 00 0 .0 A A A A A A A A A A A A A A 1 00 0 .0 A A A A A A A A A A A A A A A A A ve ra ge G oo d Contractor Management Skill 3000 A P roduc tion R ate (CY /C rew Da y) A 2000 A A A A A A A A A A A A A A A A A A A A A 1000 A A A A A A A A E asy M od era te Diffi cul t W o rk Z one Acce ssib ilit y 258 Appendix K. Embankment: Scatter Plots of Observed Production Rates vs. Candidate Drivers (Cont'd) 3 00 0 A P roduc tion R ate (CY /C rew Da y) A 2 00 0 A A A A A A A A A A A A A A A A A A A A A A A A A A A A 1 00 0 M in o r M od e ra te S eve re W ork Z on e Co ng est ion 3 00 0 A Produc tion Rate (CY/Crew Day) A 2 00 0 A A A A A A A A A A A A A A A A A A A A A A 1 00 0 A A A A A A A Qui ckly Drai n s M od e ra te E asi l y Fl o od e d Drainage_W Z 259 Appendix K. Embankment: Scatter Plots of Observed Production Rates vs. Candidate Drivers (Cont'd) 3000 A P roduc tion R ate (CY /C rew Da y) A 2000 A A A A A A A A A A A A A A A 1000 A AA A AA A AA AA A A A AA 0 10000 20000 30000 W o rk A re a Qu an tit y ( CY) 3 00 0 A Production Rate (CY/Crew Day) A 2 00 0 A A A A A A A A A A A A A A A A A A A A A A A A A A A A 1 00 0 L oo se Sti ff Rocky Soil Condition 260 Appendix L-1. Results of Testing Assumptions of the Regression Analysis for Embankment: Production Rates and Work Area Quantity Normal Q-Q Plot Production Rate (CY/Crew Day) 3000 2000 Expected Normal Value 1000 0 -1000 -1000 0 1000 2000 3000 4000 Observed Value Normal Q-Q Plot LOG (Work Area Quantity) 11 10 9 Expected Normal Value 8 7 6 6 7 8 9 10 11 Observed Value 261 Appendix L-1. Results of Testing Assumptions of the Regression Analysis for Embankment: Production Rates and Work Area Quantity (Cont'd) 2 00 0 A Unstandardized Residual 1 00 0 A A A A A A A A A A A A AA A A A A A AA A A A A A A A 0 A AA A 8 00 1 00 0 1 20 0 1 40 0 1 60 0 Unstandardized Predicted Value Normal Q-Q Plot Unstandardized Residual 2000 1000 Expected Normal Value 0 -1000 -2000 -2000 -1000 0 1000 2000 3000 Observed Value 262 Appendix L-2. Results of Testing Normality of Variables for Embankment: Production Rates by Work Zone Congestion Normal Q-Q Plot Production Rate (CY/Crew Day) Minor Congestion 3000 Expected Normal Value 2000 1000 0 0 1000 2000 3000 4000 Observed Value Normal Q-Q Plot Production Rate (CY/Crew Day) Moderate Congestion 2000 Expected Normal Value 1000 0 -1000 -1000 0 1000 2000 3000 Observed Value 263 Production Rate (SY/Crew Day) 1 00 0 2 00 0 3 00 0 0 A Bridge R eplacem ent/New Bridge A Bridge W idening/R ehibilitatio n A A A A A A Widen Nonfreewa y A Appendix M. Lime-Treated Sub-grade: Scatter Plots of Observed Production Rates vs. Candidate Drivers 264 Widen Freeway A A A A A A A A Covert Nonfreewa y to Freeway A Project Type New Location Freeway A A A A A A A A A Upgrade Freeway to Standards Upgrade Nonfreewa y To Standards A A A A Interchange s Appendix M. Lime-Treated Sub-grade: Scatter Plots of Observed Production Rates vs. Candidate Drivers (Cont'd) A P roduc tion R ate (SY /C rew D ay) 3 00 0 A A A A A A 2 00 0 A A A A A A A A A A A A A A A A A A A 1 00 0 A 0 Rura l Urb an A M etro Lo ca t io n A Produc tion Rate (SY/Cr ew Day) 3 000 A A A A A 2 000 A A A A A A A A A A A A A 1 000 A A A A A A A A 0 S im pl e M od e ra te Com ple x Project Complexity 265 Appendix M. Lime-Treated Sub-grade: Scatter Plots of Observed Production Rates vs. Candidate Drivers (Cont'd) A Produc tion Rate (SY/Crew Day) 3 00 0 A A A A A 2 00 0 A A A A A A A A A A A A A A A A A 1 00 0 0 Non e A M il e ston e s wi th In ce n tive s/Di scen ti ve Accelerated Construction Provision A Produc tion Rate (SY/Crew Day) 3 00 0 A A A A A A 2 00 0 A A A A A A A A A A A 1 00 0 A A A A A A A A 0 A ve ra ge A G oo d Contractor Management Skill 266 Appendix M. Lime-Treated Sub-grade: Scatter Plots of Observed Production Rates vs. Candidate Drivers (Cont'd) A Produc tion Rate (SY/Cr ew Day) 3 00 0 A A A A A 2 00 0 A A A A A A A A A A A A A A A A A A A A A A 1 00 0 0 M ino r M od era te Severe Congestion_W Z A Production Rate (SY/Cr ew Day) 3 00 0 A A A A A A 2 00 0 A A A A A A A A A A A 1 00 0 A A A A A A 0 M od e ra te A A Hig h Clay Content_W Z 267 Appendix M. Lime-Treated Sub-grade: Scatter Plots of Observed Production Rates vs. Candidate Drivers (Cont'd) A Produc tion Rate (SY/Crew Day) 3 00 0 A A A A A A 2 00 0 A A A A A A A A A A A A A A A A A A A 1 00 0 0 Fl at A M od e ra te S te e p Land Slope_W Z A P roduc tion R ate (SY /C rew D ay) 3000 A A A A A A 2000 A A A A A AA AA A A A A A 1000 A A AA A A A A A 0 0 10000 20000 30000 40000 50000 W o rk A re a Qu an tit y ( SY) 268 Appendix M. Lime-Treated Sub-grade: Scatter Plots of Observed Production Rates vs. Candidate Drivers (Cont'd) A P roduc tion R ate (SY /C rew D ay) 3 000 A A A A A A 2 000 A A A A A A A A A A A AA A A A A A A A A A A A A 1 000 0 0 1 000 0 2 00 0 0 3 000 0 Le ng t h of W or k Ar ea ( LF ) A Production Rate (SY/Cr ew Day) 3 00 0 A A A A A A 2 00 0 A A A A A A A A A A A A A A A A A A A A 1 00 0 0 No Y es Type C Lime Used 269 Appendix M. Lime-Treated Sub-grade: Scatter Plots of Observed Production Rates vs. Candidate Drivers (Cont'd) A Production Rate (SY/Crew Day) 3000 A A A A A 2000 A A A A A A A A A A A A A A A A A A 1000 0 6 Lift-Thickness A Produc tion Rate (SY/Cr ew Day) A 3 00 0 A A A A 2 00 0 A A A A A A A A A A A A A A A A A A A 1 00 0 0 L oo se A Sti ff Rocky Soil Condition 270 Appendix N-1. Results of Testing Assumptions of the Regression Analysis for Lime-Treated Sub-grade: Production Rates and Work Area Quantity Normal Q-Q Plot Production Rate (SY/Crew Day) 3000 2000 Expected Normal Value 1000 0 -1000 -1000 0 1000 2000 3000 4000 Observed Value Normal Q-Q Plot Work Area Quantity (SY) 30000 20000 Expected Normal Value 10000 0 -10000 -10000 0 10000 20000 30000 Observed Value 271 Appendix N-1. Results of Testing Assumptions of the Regression Analysis for Lime-Treated Sub-grade: Production Rates and Work Area Quantity (Cont'd) A A Unstandardized Residual 500 A A A A A A A A A A A A A A A A A A A A A A 0 A A A -500 -1000 1000 A 2000 3000 Unstandardized Predicted Value Normal Q-Q Plot Unstandardized Residual 1000 Expected Normal Value 0 -1000 -2000 -1000 0 1000 Observed Value 272 Appendix N-2. Results of Testing Assumptions of the Regression Analysis for Lime-Treated Sub-grade: Production Rates vs. Length of Work Area Normal Q-Q Plot Production Rate (SY/Crew Day) 4000 3000 2000 Expected Normal Value 1000 0 -1000 -1000 0 1000 2000 3000 4000 Observed Value Normal Q-Q Plot Log (Length of Work Area (LF)) 10 9 8 Expected Normal Value 7 6 5 5 6 7 8 9 10 Observed Value 273 Appendix N-2. Results of Testing Assumptions of the Regression Analysis for Lime-Treated Sub-grade: Production Rates vs. Length of Work Area (Cont'd) 100 0 .00 000 A Unstan dardized Residuals A 500 .0 00 00 A A A A A A A A A A A A A A A A A A A 0 .00 0 00 AA A A -500 .00 0 00 A A A -100 0 .0 0 000 0 .00 0 00 1 00 0.000 00 A A 200 0 .00 000 Predicted Production Rate Normal Q-Q Plot Unstandardized Residuals 2000 1000 Expected Normal Value 0 -1000 -2000 -2000 -1000 0 1000 2000 Observed Value 274 Appendix O. Cement-Treated Base: Scatter Plots of Observed Production Rates vs. Candidate Drivers Production Rate (SY-LIFT/Crew Day) A 6000 A A 5000 A A A 4000 A A A A 3000 A A 2000 A Bridge Replacement/New Bridge Widen Freeway Bridge Widening/Rehibilitation Upgrade Freeway to Standards Project Type A Produc tion Rate (SY-LIF T/Crew Day) 6 000 A A 5 000 A A A 4 000 A A A A 3 000 A A 2 000 A Rura l Urb a n Location 275 Appendix O. Cement-Treated Base: Scatter Plots of Observed Production Rates vs. Candidate Drivers (Cont'd) A Produc tion Rate (SY-LIFT/Crew Day) 6 000 A A 5 000 A A A 4 000 A A A A 3 000 A A 2 000 A S im p l e M od e ra te Project Complexity Production Rate (SY-LIFT/Crew Day) A 6000 A A 5000 A A A 4000 A A A A A 3000 A 2000 A None Accelerated Construction Provision 276 Appendix O. Cement-Treated Base: Scatter Plots of Observed Production Rates vs. Candidate Drivers (Cont'd) Production Rate (SY-LIFT/Crew Day) A 6000 A A 5000 A A A 4000 A A A A 3000 A A 2000 A Average Good Contractor Management Skill A Production Rate (SY-LIF T/Crew Day) 6 00 0 A A 5 00 0 A A A 4 00 0 A A 3 00 0 A A A A 2 00 0 A M in o r M od e ra te Congestion_W Z 277 Appendix O. Cement-Treated Base: Scatter Plots of Observed Production Rates vs. Candidate Drivers (Cont'd) A Production Rate (SY-LIFT/Crew Day) 6 00 0 A A 5 00 0 A A A 4 00 0 A A A 3 00 0 A A A 2 00 0 A Fl at M od e ra te S te e p Land Slope_W Z A Produc tion Rate (SY-LI FT/Cre w Day) 6000 A A 5000 A A A A A A A 4000 A 3000 A A 2000 A 10000 20000 30000 Work Area Quanti ty (SY-LIF T) 278 Appendix O. Cement-Treated Base: Scatter Plots of Observed Production Rates vs. Candidate Drivers (Cont'd) A Production Rate (SY-LIFT/Crew Day) 600 0 A A 500 0 A A A A AA A 400 0 300 0 A A A 200 0 A 0 200 0 400 0 6 000 Lift-Length of W ork Area (LF) Production Rate (SY-LIFT/Crew Day) A 6000 A A 5000 A A A A A 4000 A A A A 3000 A 2000 A 30 40 50 60 70 Width of Work Area (LF) 279 Appendix O. Cement-Treated Base: Scatter Plots of Observed Production Rates vs. Candidate Drivers (Cont'd) Production Rate (SY-LIFT/Crew Day) A 6000 A A 5000 A A A 4000 A A A A A 3000 A 2000 A 6 Lift-Thickness (Inch) 280 Appendix P. Flexible Base: Scatter Plots of Observed Production Rates vs. Candidate Drivers A Production Rate (SY-Lift/Crew Day) 5000 A 4000 A A A A A 3000 A A A 2000 A 1000 A A A A Widen Nonfreeway Upgrade Nonfreeway To Standards Covert Nonfreeway to Freeway Interchanges New Location Freeway Project Type A Production Rate (SY-Lift/Crew Day) 5000 A 4000 A A A 3000 A A A 2000 A 1000 A A A A Rural Urban Metro Location 281 Appendix P. Flexible Base: Scatter Plots of Observed Production Rates vs. Candidate Drivers (Cont'd) A Produc tion Rate (SY-Lift/Crew Day) 5 00 0 A 4 00 0 A A A A A A A A 3 00 0 2 00 0 A 1 00 0 A A A A S im p l e M od e ra te Com p le x Project Complexity A Produc tion Rate (SY-Lift/Crew Day) 5 00 0 A 4 00 0 A A A A A A 3 00 0 A 2 00 0 A 1 00 0 A A A A Non e S ub stan ti al Com ple ti o n I/D M il e ston e s with In ce n tive s/Di scen ti ve Accelerated Construction Provision 282 Appendix P. Flexible Base: Scatter Plots of Observed Production Rates vs. Candidate Drivers (Cont'd) A Produc tion Rate (SY-Lift/Crew Day) 5 00 0 A 4 00 0 A A A A A A 3 00 0 A 2 00 0 A 1 00 0 A A A A A ve ra ge G oo d Contractor Management Skill A Production Rate (SY-Lift/Crew Day) 5 00 0 A 4 00 0 A A A 3 00 0 A A A 2 00 0 A 1 00 0 A A A A M in o r M od e ra te Congestion_W Z 283 Appendix P. Flexible Base: Scatter Plots of Observed Production Rates vs. Candidate Drivers (Cont'd) A Production Rate (SY-Lift/Crew Day) 5 00 0 A 4 00 0 A A A A A A A A 3 00 0 2 00 0 A 1 00 0 A A A A Fl at M od e ra te Land Slope_W Z A Production Rate (SY-Lift/Crew Day) 5000 A 4000 A A A A 3000 A A A A 2000 A A 1000 A A A 0 10000 20000 30000 40000 Work Area Quantity (SY-Lift) 284 Appendix P. Flexible Base: Scatter Plots of Observed Production Rates vs. Candidate Drivers (Cont'd) A Produc tion Rate (SY-Lift/Crew Day) 5 00 0 A 4 00 0 A A A A A A 3 00 0 A A 2 00 0 A A 1 00 0 A A A 0 2 50 0 5 00 0 7 50 0 1 00 0 0 Lift-Length of W ork Area (LF) A Produc tion Rate (SY-Lift/Crew Day) 5 00 0 A 4 00 0 A A 3 00 0 A A A A A A 2 00 0 A A 1 00 0 A A A 30 40 50 60 70 W idth of W ork Area (LF) 285 Appendix P. Flexible Base: Scatter Plots of Observed Production Rates vs. Candidate Drivers (Cont'd) A Production Rate (SY-Lift/Crew Day) 5000 A 4000 A A A A 3000 A A A A 2000 A 1000 A A A A 4 6 8 Lift-Thickness (Inch) 286 Appendix Q-1. Results of Testing Assumptions of the Regression Analysis for Cement-Treated Base: Production Rates vs. Work Area Quantity Normal Q-Q Plot Production Rate (SY-LIFT/Crew Day) 7000 6000 5000 Expected Normal Value 4000 3000 2000 1000 1000 2000 3000 4000 5000 6000 7000 Observed Value Normal Q-Q Plot Log (Work Area Quantitty) 11.0 10.5 10.0 9.5 Expected Normal Value 9.0 8.5 8.0 7.5 7.0 7.0 7.5 8.0 8.5 9.0 9.5 10.0 10.5 11.0 Observed Value 287 Appendix Q-1. Results of Testing Assumptions of the Regression Analysis for Cement-Treated Base: Production Rates vs. Work Area Quantity (Cont'd) A Unstandardiz ed Residual 1000 .00 000 A A A A 0.00 0 00 A A A A A A A A -1000 .0 0 00 0 A 2 00 0.00 0 00 3 00 0.00 0 00 4 00 0.000 00 5 00 0.000 00 6 00 0.000 00 Unstandardized Predicted Value Normal Q-Q Plot Unstandardized Residual 2000 1000 Expected Normal Value 0 -1000 -2000 -2000 -1000 0 1000 2000 Observed Value 288 Appendix Q-2. Results of Testing Assumptions of the Regression Analysis for Cement-Treated Base: Production Rates vs. LiftLength of Work Area Normal Q-Q Plot Production Rate (SY-LIFT/Crew Day) 7000 6000 5000 Expected Normal Value 4000 3000 2000 1000 1000 2000 3000 4000 5000 6000 7000 Observed Value Normal Q-Q Plot Lift-Length (LF) 3000 2000 Expected Normal Value 1000 0 0 1000 2000 3000 4000 Observed Value 289 Appendix Q-2. Results of Testing Assumptions of the Regression Analysis for Cement-Treated Base: Production Rates vs. Lift-Length of Work Area (Cont'd) 2 000 .000 00 A Unstandardized Residuals A 1 000 .000 00 A A A A A A 0.000 00 A -1 000 .0 000 0 A A A 3 000 .000 00 4 00 0.000 00 5 00 0 .00 000 Predicted Production Rates Normal Q-Q Plot Unstandard Residuals 2000 1000 Expected Normal Value 0 -1000 -2000 -2000 -1000 0 1000 2000 3000 Observed Value 290 Appendix R-1. Results of Testing Assumptions of the Regression Analysis for Flexible Base: Production Rates vs. Work Area Quantity Normal Q-Q Plot Production Rate (SY-Lift/Crew Day) 6000 5000 4000 Expected Normal Value 3000 2000 1000 0 0 1000 2000 3000 4000 5000 6000 Observed Value Normal Q-Q Plot Log (Work Area Quantity) 11.5 11.0 10.5 10.0 Expected Normal Value 9.5 9.0 8.5 8.0 7.5 7 8 9 10 11 12 Observed Value 291 Appendix R-1. Results of Testing Assumptions of the Regression Analysis for Flexible Base: Production Rates vs. Work Area Quantity (Cont'd) A 2 00 0.00 000 Unstandardiz ed Residuals 1 00 0.00 000 A A A A A A 0.00 000 A A A A A -1 00 0.0 0 000 AA A 1 000 .000 00 2000 .000 00 3 00 0.00 000 4 00 0.00 000 Predicted Production Rate Normal Q-Q Plot Unstandardized Residuals 2000 1000 Expected Normal Value 0 -1000 -2000 -2000 -1000 0 1000 2000 3000 Observed Value 292 Appendix R-2. Results of Testing Assumptions of the Regression Analysis for Flexible Base: Production Rates vs. Lift-Length of Work Area Normal Q-Q Plot Production Rate (SY-Lift/Crew Day) 6000 5000 4000 Expected Normal Value 3000 2000 1000 0 0 1000 2000 3000 4000 5000 6000 Observed Value Normal Q-Q Plot Log (Lift-Length of Work Area) 10 9 8 Expected Normal Value 7 6 5 5 6 7 8 9 10 Observed Value 293 Appendix R-2. Results of Testing Assumptions of the Regression Analysis for Flexible Base: Production Rates vs. Lift-Length of Work Area (Cont'd) A Unstandardiz ed Residuals 2 00 0 .00 0 00 1 00 0 .00 0 00 A A A A A A A 0 .00 0 00 A A A A A A -1 00 0 .0 0 00 0 A 1 00 0 .00 0 00 2 00 0 .00 0 00 3 00 0 .00 0 00 4 00 0 .00 0 00 Predicted Production Rates Normal Q-Q Plot Unstandardized Residuals 2000 1000 Expected Normal Value 0 -1000 -2000 -2000 -1000 0 1000 2000 3000 Observed Value 294 Appendix S. Results of Testing Assumptions of the Multiple Regression Analysis for Embankment: 1 00 0 A A Unsta ndardi zed Residu al A A A A A A A A A A 5 00 0 A A A A A A A A A AA A A A A A A A A A -5 00 8 00 1 20 0 1 60 0 2 00 0 Unstandardized Predicted Value 295 Production Rate (Ton/C rew Day) 1 00 0 1 50 0 5 00 A A A A A A A A A A A A A A A A Bridge W idening/R ehibilitation Bridge Replacem ent/New Bridge Rehailitate Existing Road Widen Nonfreewa y Widen Freeway A Appendix T. Hot Mix Asphalt Pavement: Scatter Plots of Observed Production Rates vs. Candidate Drivers Project Type 296 Covert Nonfreewa y to Freeway A A A A New Location Freeway A A A A Upgrade Freew ay to Standards A A A Upgrade Nonfreewa y To Standards Interchang es A A A A Appendix T. Hot Mix Asphalt Pavement: Scatter Plots of Observed Production Rates vs. Candidate Drivers (Cont'd) 1 500 A A A A Produc tion Rate (Ton/C rew Day) A A A 1 000 A A A A A A A A A A A A A A A 5 00 A A A A A A A A Rural Urb an M etro Location 1 50 0 A A A A Produc tion Rate (Ton/Crew Day) A A A 1 00 0 A A A A A A A A A A A A A A A A A A A 5 00 A A A A A Rare l y Co ng e ste d M ost ho u rs ve ry co ng e ste d Onl y rush ho u rs co ng e ste d Traffic Flow 297 Appendix T. Hot Mix Asphalt Pavement: Scatter Plots of Observed Production Rates vs. Candidate Drivers (Cont'd) 1500 A A A A Produc tion Rate (Ton/C rew Day) A A A A 1000 A A A A A A A A A A A A A A A A 500 A A A A A A A A S im p l e M od erate Com plex Project Complexity 1500 A A A Production Rate (Ton/Crew Day) A A A 1000 A A A A A A A A A A A A A A A A 500 A A A A A A A A Substantial Completion I/D Incentive using contract administrative cost Milestones with Incentives/Discentive None Accelerated Construction Provision 298 Appendix T. Hot Mix Asphalt Pavement: Scatter Plots of Observed Production Rates vs. Candidate Drivers (Cont'd) 1 50 0 A A A A Produ ction Rate (Ton/C rew Day) A A A A 1 00 0 A A A A A A A A A A A A A A A A A A A A A A A 5 00 A A ve ra ge G oo d Contractor Management Skill 1 50 0 A A A A Production Rate (Ton/C rew Day) A A A 1 00 0 A A A A A A A A A A A A A A A 5 00 A A A A A A A A E asy M od e ra te Accessibility_W Z 299 Appendix T. Hot Mix Asphalt Pavement: Scatter Plots of Observed Production Rates vs. Candidate Drivers (Cont'd) 1 50 0 A A A A Produ ction Rate (Ton/C rew Day) A A A A 1 00 0 A A A A A A A A A A A A A A A A A A A A A A 5 00 A A M in o r M od e ra te S eve re Congestion_W Z 1 50 0 A A A A Produc tion Rate (Ton/C rew Day) A A A 1 00 0 A A A A A A A A A A A A A A 5 00 A A A A A A A A Fl at M od e ra te S te e p Land Slope_W Z 300 Appendix T. Hot Mix Asphalt Pavement: Scatter Plots of Observed Production Rates vs. Candidate Drivers (Cont'd) 1500 A A A A Production Rate (Ton/Crew Day) A A A A 1000 A A A A A A A A A A A A A A A A 500 A A A A A A A A 1000 2000 3000 4000 5000 6000 Work Area Quantity (Ton) 1500 A A A A Production Rate (Ton/C rew Day) A A A A 1000 A A A A A A A A A A A A A A A 500 A A A A A A A A B ase Co u rse S urfa ce S urfa ce a nd B ase Course Type 301 Appendix T. Hot Mix Asphalt Pavement: Scatter Plots of Observed Production Rates vs. Candidate Drivers (Cont'd) 1 50 0 A A A A Produc tion Rate (Ton/C rew Day) A A A A 1 00 0 A A A A A A A A A A A A A A A 5 00 A A A A A A A A M ain La ne Non -ma i n Lane Main Lane vs. Non-main Lane Apllication 302 Appendix U-1. Results of Testing Assumptions of the Regression Analysis for Hot Mix Asphalt Pavement: Production Rates vs. Work Area Quantity Normal Q-Q Plot Production Rate (Ton/Crew Day) 1600 1400 1200 1000 Expected Normal Value 800 600 400 200 0 0 200 400 600 800 1000 1200 1400 1600 Observed Value Normal Q-Q Plot Log (Work Area Quantity (Ton)) 9.0 8.5 8.0 7.5 7.0 6.5 6.0 5.5 5 6 7 8 9 10 Expected Normal Value Observed Value 303 Appendix U-1. Results of Testing Assumptions of the Regression Analysis for Hot Mix Asphalt Pavement: Production Rates vs. Work Area Quantity (Cont'd) 5 00 A A A Unstandardized Residuals 2 50 A A A A A A A A A A 0 A A A A A A A A A -2 50 A A AA A A A A A 4 00 6 00 8 00 1 00 0 1 20 0 Predicted Production Rates Normal Q-Q Plot Unstandardized Residuals 600 400 200 Expected Normal Value 0 -200 -400 -600 -600 -400 -200 0 200 400 600 Observed Value 304 Appendix U-2. Results of Testing Normality of Variables for Hot Mix Asphalt Pavement: Production Rates by Course Types Normal Q-Q Plot Production Rate (Ton/Crew Day) Surface 1400 1200 1000 Expected Normal Value 800 600 400 200 0 -200 -200 0 200 400 600 800 1000 1200 1400 1600 Observed Value Normal Q-Q Plot Production Rate (Ton/Crew Day) Base Course 1600 1400 Expected Normal Value 1200 1000 800 600 400 200 200 400 600 800 1000 1200 1400 1600 Observed Value 305 Appendix V. Slip-form Concrete Pavement: Scatter Plots of Observed Production Rates vs. Candidate Drivers A Produc tion Rate (SY/Crew Day) 2 00 0 A A A A 1 50 0 A A A A A A A A 1 00 0 A A A A A A 500 A Reh a i li ta te Existi n g Roa d Covert No nfre eway to Fre eway Wide n Fre eway Upg ra de Free way to Stan dards Wid en Nonfreeway In tercha nge s Project Type A Produc tion Rate (SY/Crew Day) 2 000 A A A A 1 500 A A A A A A A 1 000 A A A A A 5 00 Rura l A Urb a n M etro Location 306 Appendix V. Slip-form Concrete Pavement: Scatter Plots of Observed Production Rates vs. Candidate Drivers (Cont'd) A Produc tion Rate (SY/Crew Day) 2 00 0 A A A 1 50 0 A A A A A A A A 1 00 0 A A A A A A 5 00 A S im p l e M od e ra te Com p le x Project Complexity A Produc tion Rate (SY/Cr ew Day) 2000 A A A A 1500 A A A A A A A 1000 A A A A A 500 A Rare l y Cong este d M ost ho urs very co ngested O nl y rush hou rs co ngested Traffic Flow 307 Appendix V. Slip-form Concrete Pavement: Scatter Plots of Observed Production Rates vs. Candidate Drivers (Cont'd) A Production Rate (SY/Cr ew Day) 2 00 0 A A A 1 50 0 A A A A A A A A 1 00 0 A A A A A A 5 00 A Non e In ce n ti ve usin g co ntra ct a dm i nistrative co st M ile ston e s with In ce n tive s/Di sce n tive S ub stan ti al Com p le tion I/D Accelerated Construction Provision A Produc tion Rate (SY/Cr ew Day) 2 00 0 A A A A 1 50 0 A A A A A A A 1 00 0 A A A A A A 5 00 A ve ra ge A G oo d Contractor Management Skill 308 Appendix V. Slip-form Concrete Pavement: Scatter Plots of Observed Production Rates vs. Candidate Drivers (Cont'd) A Produc tion Rate (SY/Cr ew Day) 2 000 A A A A 1 500 A A A A A A A 1 000 A A A A A A 5 00 A E asy M ode rate Diffi cul t Accessibility_W Z A Production Rate (SY/Crew Day) 2 00 0 A A A 1 50 0 A A A A A A A A 1 00 0 A A A A A A 500 M in or A M oderate S evere Congestion_W Z 309 Appendix V. Slip-form Concrete Pavement: Scatter Plots of Observed Production Rates vs. Candidate Drivers (Cont'd) A Produc tion Rate (SY/Cr ew Day) 2000 A A A 1500 A A A A A A A 1000 A A A A A 500 Fl at A M oderate Land Slope_W Z A P roduc tion R ate (SY /C rew D ay) 2000 A A A A 1500 A A A A A A A A 1000 A AA A A A 5 00 0 A 1 000 0 2 00 00 3 000 0 4 000 0 W o rk A re a Qu an tit y ( SY) 310 Appendix V. Slip-form Concrete Pavement: Scatter Plots of Observed Production Rates vs. Candidate Drivers (Cont'd) A P roduc tion R ate (SY /C rew D ay) 2 00 0 A A A A 1 50 0 A A A A A A A A 1 00 0 A A AA A A 5 00 0 A 500 0 1 000 0 1 500 0 Le ng t h of W or k Ar ea ( LF ) A Produc tion Rate (SY/Cr ew Day) 2 00 0 A A A A 1 50 0 A A A A A A A A 1 00 0 A A A A A 5 00 12 16 20 A 24 W idth of W ork Area (LF) 311 Appendix V. Slip-form Concrete Pavement: Scatter Plots of Observed Production Rates vs. Candidate Drivers (Cont'd) A Produc tion Rate (SY/Crew Day) 2 00 0 A A A A 1 50 0 A A A A A A A 1 00 0 A A A A A A 5 00 8 9 10 11 12 A 13 Thickness of Concrete Pavement (Inch) 312 Appendix W-1. Results of Testing Assumptions of the Regression Analysis for Slip-form Concrete Pavement Construction: Production Rates vs. Work Area Quantity Normal Q-Q Plot Log (Production Rates) 7.8 7.6 7.4 Expected Normal Value 7.2 7.0 6.8 6.6 6.4 6.6 6.8 7.0 7.2 7.4 7.6 7.8 Observed Value Normal Q-Q Plot Log (Work Area Quantity) 10.5 10.0 9.5 Expected Normal Value 9.0 8.5 8.0 7.5 7.0 7.0 7.5 8.0 8.5 9.0 9.5 10.0 10.5 11.0 Observed Value 313 Appendix W-1. Results of Testing Assumptions of the Regression Analysis for Slip-form Concrete Pavement Construction: Production Rates vs. Work Area Quantity (Cont'd) A 5 00 .0 00 00 Unstandardized Residuals A 2 50 .0 00 00 A A A A A A A 0 .00 000 A A A A A AA A -2 50 .00 0 00 -5 00 .00 0 00 A 7 50 .0 00 00 1 000 .00 0 00 1250 .000 00 1 50 0.00 000 1750 .000 00 Predicted Production Rates Normal Q-Q Plot Unstandardized Residuals 600 400 200 Expected Normal Value 0 -200 -400 -600 -600 -400 -200 0 200 400 600 800 Observed Value 314 Appendix W-2. Results of Testing Assumptions of the Regression Analysis for Slip-form Concrete Pavement Construction: Production Rates vs. Length of Work Area Normal Q-Q Plot Production Rate (SY/Crew Day) 2000 1800 1600 1400 Expected Normal Value 1200 1000 800 600 400 0 1000 2000 3000 Observed Value Normal Q-Q Plot Log (Length of Work Area) 9.5 9.0 8.5 Expected Normal Value 8.0 7.5 7.0 6.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 Observed Value 315 Appendix W-2. Results of Testing Assumptions of the Regression Analysis for Slip-form Concrete Pavement Construction: Production Rates vs. Length of Work Area (Cont'd) 8 00.0000 0 A Unsta ndardiz ed Residuals 4 00.0000 0 A A A A A 0 .00 0 00 A A A A AA A A A A A -4 00.00 000 A 7 50.0000 0 1000 .00 0 00 1 250 .000 00 1 50 0.000 00 Predicted Production Rates Normal Q-Q Plot Unstandardized Residuals 800 600 400 200 Expected Normal Value 0 -200 -400 -600 -800 -800 -600 -400 -200 0 200 400 600 800 1000 Observed Value 316 Appendix X. Conventional Form Concrete Pavement: Scatter Plots of Observed Production Rates vs. Candidate Drivers 600 A A A Produc tion Rate (SY/Cr ew Day) 500 A A 400 A A A A A A 300 A A A A 200 100 A A Wid e n Fre ewa y Cove rt Nonfreewa y to Free wa y Reh ai li tate E xisti ng Roa d Upgra de Fre e wa y to S ta nda rd s Project Type 6 00 A A Produc tion Rate (SY/Cr ew Day) 5 00 A A 4 00 A A A A A A A A A A 3 00 2 00 1 00 A A Urb a n Location 317 Appendix X. Conventional Form Concrete Pavement: Scatter Plots of Observed Production Rates vs. Candidate Drivers (Cont'd) 6 00 A A Produc tion Rate (SY/Crew Day) 5 00 A A 4 00 A A A A A A A 3 00 2 00 A A A A A 1 00 A A A S im p l e M od e rate Com p le x Project Complexity 6 00 A A A Produc tion Rate (SY/Cr ew Day) 5 00 A A 4 00 A A A A A A A A A 3 00 2 00 1 00 A A Rare l y Co ng e ste d M ost ho u rs ve ry co nge ste d O nl y rush ho u rs co ng e ste d Traffic Flow 318 Appendix X. Conventional Form Concrete Pavement: Scatter Plots of Observed Production Rates vs. Candidate Drivers (Cont'd) 6 00 A A Produc tion Rate (SY/Crew Day) 5 00 A A 4 00 A A A A A A A 3 00 2 00 A A A A A 1 00 A A A Non e Sub stan ti al Com pl e ti on I/D In ce n ti ve usi ng con tract a dm i nistrative co st M il e sto n e s wi th Ince n ti ve s/Di sce n ti ve Accelerated Construction Provision 6 00 A A Produc tion Rate (SY/Crew Day) 5 00 A A 4 00 A A A A A A 3 00 A A A A 2 00 1 00 A A A ve rage G oo d Contractor Management Skill 319 Appendix X. Conventional Form Concrete Pavement: Scatter Plots of Observed Production Rates vs. Candidate Drivers (Cont'd) 6 00 A A Produc tion Rate (SY/Cr ew Day) 5 00 A A 4 00 A A A A A A A A A 3 00 2 00 A 1 00 A A E asy M od e ra te Diffi cul t Accessibility_W Z 6 00 A A Produc tion Rate (SY/Crew Day) 5 00 A A 4 00 A A A A A A A 3 00 A 2 00 A A 1 00 A A M in o r M od e ra te S eve re Congestion_W Z 320 Appendix X. Conventional Form Concrete Pavement: Scatter Plots of Observed Production Rates vs. Candidate Drivers (Cont'd) 6 00 A A Production Rate (SY/Cr ew Day) 5 00 A A 4 00 A A A A A A A A A A 3 00 2 00 1 00 A A Fl at Land Slope_W Z 6 00 A A A P roduc tion R ate (SY /C rew D ay) 5 00 A A 4 00 A A A A 3 00 A A A A A 2 00 A 1 00 A A 0 1 00 0 2 00 0 3 00 0 4 00 0 W o rk A re a Qu ant it y ( SY ) 321 Appendix X. Conventional Form Concrete Pavement: Scatter Plots of Observed Production Rates vs. Candidate Drivers (Cont'd) 600.0 A A Production Rate (SY/Crew Day) 500.0 A A 400.0 A A A A A A A A 300.0 A 200.0 A A 100.0 A A A None Yes Configuration (Curve or Sharp Angle) 6 00 A A A A A Produ ction Rate (SY/Cr ew Day) 5 00 4 00 A A A A 3 00 A A A A A A 2 00 1 00 A A 6 8 10 12 Thickness of Concrete Pavement (Inch) 322 Appendix Y-1. Results of Testing Assumptions of the Regression Analysis for Conventional Form Concrete Pavement Construction: Production Rates vs. Work Area Quantity Normal Q-Q Plot Production Rate (SY/Crew Day) 600 500 400 Expected Normal Value 300 200 100 0 0 100 200 300 400 500 600 Observed Value Normal Q-Q Plot Production Rate (SY/Crew Day) 700 600 500 Expected Normal Value 400 300 200 100 0 0 100 200 300 400 500 600 700 Observed Value 323 Appendix Y-1. Results of Testing Assumptions of the Regression Analysis for Conventional Form Concrete Pavement Construction: Production Rates vs. Work Area Quantity (Cont'd) A 200 A Unstandardized Residuals A 100 A A A A 0 A A A A A A A -100 A A A A 200 300 400 500 Predicted Values Normal Q-Q Plot Unstandardized Residuals 200 100 Expected Normal Value 0 -100 -200 -200 -100 0 100 200 300 Observed Value 324 Appendix Y-2. Results of Testing Normality for Conventional Form Concrete Pavement Construction: Observed Production Rates by Configuration of Concrete Pavement Normal Q-Q Plot Production Rate (SY/Crew Day) Typical Configuration 700 600 Expected Normal Value 500 400 300 200 100 200 300 400 500 600 700 Observed Value Normal Q-Q Plot Production Rate (SY/Crew Day) Non-Typical Configuration 400 300 Expected Normal Value 200 100 0 0 100 200 300 400 Observed Value 325 Appendix Z. Results of Testing Assumptions of the Multiple Regression Analysis for Hot Mix Asphalt Pavement: 500 A A Unstandardized Residual A A 250 A A A A A A A A A A A A 0 A A AA A A A A A A A A -250 -500 400 600 800 1000 A A A 1200 Unstandardized Predicted Value 326 GLOSSARY Candidate Driver: Driver that is known at the design stage Contract Time: Maximum time allowed for completion of all work described in contract documents (Herbsman and Ellis 1995). Data Point: An observation or a series of observations that document the Production Rate information of a Work Item including total quantity, total working days, employed resources characteristics of Production Rate Factors and disruptions in a Work Area Operation: Combination of one or several tasks employed to complete a particular Work Item Production Rate Factor: A factor that causes fluctuation of Production Rate Production Rate: An average quantity of output produced within a working day by a group of resource, where the quantity can be in the form of Cubic Yard (CY), Square Yard (SY) and Ton. Significant Driver: Driver that is found to have statistically significant effect(s) on Production Rate Task: A single work process in an Operation Work Area: A designated area where an operation of a Work Item is being performed and is only limited to the observed working phase Work Area Quantity: Total quantity of a Work Item in a Work Area 327 Work Item: A single item of construction activity usually in combination with products, or materials, and construction aids undertaken by one person or a team, such as Excavation and Slip-from Concrete Pavement (2004 www.buildingcatalogue.com.au) Work Zone: A zone where an operation of a Work Item is being performed and may consist of one or several Work Areas depending on number of work phases being constructed in the zone 328 BIBLIOGRAPHY Abdelhamid, T. S., and Everett, J. G. (1999). "Time Series Analysis for Construction Productivity Experiments." J. Constr. Engrg.and Mgmt., ASCE, 125(2), 87-95. AbouRizk, S., Knowles, P., and Hermann, U. R. (2001). "Estimating Labor Production Rates For Industrial Construction Activities." J. Constr. Engrg. and Mgmt., ASCE, 127(6), 502-511 Albright, S. C., Winston, W. L., and Zappe, C. (1999). Data Analysis & Decision Making, Pacific Grove, CA Allmon, E., Haas, C. T., Borcherding, J. D., and Goodrum, P. M. (2000). "U.S. Construction Labor Productivity Trends, 1970~1998." J. Constr. Engrg. and Mgmt., ASCE, 126(1), 97-104 Borcherding, J. D. (1976). "Improving Productivity in Industrial Construction." Journal of Construction Devision, ASCE, 102(CO4), 599-614 Borcherding, J. D., and Alarcon, L. F. (1991). "Quantitative Effects on Construction Productivity." The Construction Lawyer, American Bar Association, 11(1), 35-48 Borcherding, J. D., and Garner, D. F. (1981). "Work Force Motivation and Productivity on Large Projects." Journal of Construction Division, ASCE, 107(CO3), 443-453 Borcherding, J. D. and Oglesby, C. H. (1974). "Construction Productivity and Job Satisfaction." Journal of the Construction Division, ASCE, 100(CO3), 413-431 Borcherding, J. D., Sebastian, S. J., and Samelson, N. M. (1980). "Improving Motivation and Productivity on Large Projects." Journal of the Construction Division, ASCE, 106(CO1), 73-89. Bhurisith, L., and Touran, A. (2002). "Case Study of Obosolescence and Equipment Productivity." J. Constr. Engrg. and Mgmt., ASCE, 128(4), 357-361. 329 The Business Roundtable (1980). "Scheduled Overtime Effect on Construction Project." Construction Industry Cost Effectiveness Task Force Report, November. The Business Roundtable (1982). "Measuring Productivity in Construction." Report A-1, September. Chambers, John, William Cleveland, Beat Kleiner, and Paul Tukey, (1983), Graphical Methods for Data Analysis, Wadsworth. Chang. L. M., and Borcherding, J. D. (1986). "Craftsman Questionnaire Sampling." J. Constr. Engrg. and Mgmt., ASCE, 112(4), 543-556. Cho, S. (2000). "Development of the Project Definition Rating Index (PDRI) for Building Projects." Doctoral Dissertation, The University of Texas at Austin, Austin, TX, May 2000. Christian J., and Hachey D. (1995). "Effects of Delay Times on Production Rates in Construction." J. Constr. Engrg. and Mgmt., ASCE, 121(1), 20-26. Christian J., and Hachey D. (1996). "A Computer-aided System to Improve Production Rates in Constructiion." Advances in Engineering Software., V. 25, 207-213. The Construction Industry Institute. (1988). "The Effects of Scheduled Overtime and Shift Schedule on Construction Craft Productivity." Construction Industry Institute, The University of Texas Austin The Construction Industry Institute. (1990). "Productivity Measurement: An Introduction." Construction Industry Institute, The University of Texas Austin El-Rayes, K., and Moselhi, O. (2001). "Impact of Rainfall On the Productivity of Highway Construction." J. Constr. Engrg. and Mgmt., ASCE, 127(2), 125-131. Gidado, K. I. (1996). "Project Complexity: The Focal Point of Construction Production Planning." J. Construction Management and Economics, V.14, 213225. Goodrum, P. M. (2001). "The Impact of Equipment Technology on Productivity in the U.S. Construction Industry." Doctoral Dissertation, The University of Texas at Austin, Austin, TX, May 2001. 330 Goodrum, P. M., and Haas, C. T. (2002). "Partial Factor Productivity and Equipment Technology Change at Activity Level in U.S. Construction Industry." J. Constr. Engrg. and Mgmt., ASCE, 128(6), 463-472. Gransberg, D. D. (1996). "Optimizing Haul Unit Size and Number based on Loading Facility Characteristics." J. Constr. Engrg. and Mgmt., ASCE, 122(3), 248-253. Green, S. B. (1999). "How Many Subjects Does It Take to Do A Regression Analysis?" Multivariate Behavioral Research, Lawrence Erlbaum Associates, 26(3), 499-510 Grimm, C. T., and Wagner, N. K. (1974). "Weather Effects on Mason Productivity." Journal of Construction Division, ASCE, 100(3), 319-335. Hancher, D. E., McFarland, W. F., and Alabay, R. T. (1992) "Construction Contract Time Determination" Texas Department of Transportation Research Report Halligan, D. W., Demsetz, L. A., and Brown, J. D. (1994). "Action-Response Model and Loss of Productivity in Construction." J. Constr. Engrg. and Mgmt., ASCE, 120(1), 47-64. Hanna, A. S., Camlic, R., Peterson P. A. and Nordheim, E. V. (2002). "Quantitative Definition of Projects Impacted by Change Orders." J. Constr. Engrg. and Mgmt., ASCE, 128(1), 57-64. Hendrickson, C., Martinelli, D. and Rehak, D. (1987). "Hierarchical Rule-based Activity Duration Estimation." J. Constr. Engrg. and Mgmt., ASCE, 113(2), 288301. Herbsman, Z. J. and Ellis, R. (1995). "Determination of Contract Time for Highway Construction Projects." National Cooperative Highway Research Program Synthesis of Highway Practice 215, Transportation Research Board, National Research Council, Washington D. C. (October), 43PP Jiang, Y. (2003). "The Effects of Traffic Flow Rates at Freeway Work Zones on Asphalt Pavement Construction Productivity." Transportation Quarterly, V.57, n.3, Summer, P83-103 Jonasson, S., Dunston, P. S., Ahmed, K., and Hamilton, J. (2002). "Factors in Productivity and Unit Cost for Advanced Machine 331 Guidance." J. Constr. Engrg. and Mgmt., ASCE, 128(5), 367-374. Koehn, E. and Brown, G. (1985). "Climatic Effects on Construction." Journal of Construction Engineering and Management, ASCE, 111(2), 129-137. Koehn, E. and Ahmed, F. (2001). "Production Rates for Urban/Rural Projects in Developing Areas" International Transaction, AACE, INT.03, 01-06. Lee, E.-B., Lbbs, C. W., Harvey, J. T. and Roesler, J. R. (2000). "Construction Productivity and Constraints for Concrete Pavement Rehabilitation in Urban Corridors." Transportation Research Record, n.1712, P13-22. Liou, F. S., and Borcherding, J. D. (1986). "Work sampling Can Predict Unit Rate Productivity." J. Constr. Engrg. and Mgmt., ASCE, 112(1), 90-103. Marzouk, M., and Moselhi, O. (2003).. "Object-oriented Simulation Model for Earthmoving Operations." J. Constr. Engrg. and Mgmt., ASCE, 129(2), 173-181. National Cooperative Highway Research Program Synthesis of Highway Practice 79. (1981). "Contract Time Determination." Transportation Research Board, National Research Council, Washington D. C. (October), 45PP National Electrical Contractors Association. (1974). "The Effect of Temperature on Productivity." Test Report, Washington, D. C. Oglesby, C. H., Parker, H. W., and Howell, G. A. (1989). Productivity improvement in construction. McGraw-Hill Publishing Inc., New York, N.Y. Ovararin, N., and Popescu, C. M. (2001). "Field Factors Affecting Masonry Productivity." International Transaction, AACE, EST-09. Peurifoy, R. L., and Schexnayder, C. J. (2002). Construction Planning, Equipment, and Methods. 6th Edition, McGraw-Hill Publishing Inc., New York, N.Y. Proverbs, G. D., Holt, G. D., and Olomolaiye, P. O. (1997). "European Construction Contractors: A Productivity Appraisal of in-situ Concrete Operations." J. Construction Management and Economics, V.17, 221-230. R.S. Means Company, Inc. (2002). R.S. Means Heavy Construction Cost Data, 59th Annual Edition, R.S. Means Company, Inc., Kingston, MA 332 Sanders, S. R., and Thomas, H. R. (1991). "Factors Affecting Masonry-Labor Productivity." J. Constr. Engrg. and Mgmt., ASCE, 117(4), 626-644. Sanders, S. R., and Thomas, H. R. (1993). "Masonry Productivity Forecasting Model." J. Constr. Engrg. and Mgmt., ASCE, 119(1), 163-179. Schexnayder, C., and Webber, S. L. (1999). "Effect of Truck Payload Weight on Production." J. Constr. Engrg. and Mgmt., ASCE, 125(1), 1-7. Smith, S. D. (1999) "Earthmoving Productivity Estimation Using Linear Regression Techniques." J. Constr. Engrg. and Mgmt., ASCE, 125(3), 133-141. Smith, S. D., Wood, G. S., and Gould, M. (2000). "A New Earthworks Estimating Methodology." J. Construction Management and Economics, V.18, 219-228. Thomas, H. R. (1991). "Labor Productivity and Work Sampling: The Bottom Line." J. Constr. Engrg. and Mgmt., ASCE, 117(3), 423-444. Texas Department of Transportation (1993). "Standard Specifications for Construction of Highways, Streets and Bridges." TX, USA Thomas, H. R. (2000). "Schedule Acceleration, Work Flow, and Labor Productivity." J. Constr. Engrg. and Mgmt., ASCE, 126(4), 261-267. Thomas, H. R., Mathews, C. T. and Ward, J. G. (1986). "Learning Curve Models of Construction Productivity." J. Constr. Engrg. and Mgmt., ASCE, 112(2), 245258. Thomas, H. R., Maloney, W. F., Horner, R. M. W., Smith, G. R., Handa, V. K., and Sanders, S. R. (1990). "Modeling Construction Labor Productivity" J. Constr. Engrg. and Mgmt., ASCE, 116(4), 705-726. Thomas, H. R., and Raynar, K. A. (1997). "Schedules Overtime and Labor Productivity: Quantitative Analysis." J. Constr. Engrg. and Mgmt., ASCE, 123(2), 181-188. Thomas, H. R., Riley, D. R., and Sanvido, V. E. (1998). "Loss of Labor Productivity Due to Delivery Methods and Weather." J. Constr. Engrg. and Mgmt., ASCE, 125(1), 39-46. Thomas, H. R., and Yiakoumis I. (1987). "Factor Model of Construction Productivity." J. Constr. Engrg. and Mgmt., ASCE, 113(4), 623-639. 333 Thomas, H. R. and Zavrski, I. (1999). "Construction Baseline Productivity: Theory and Practice." Journal of Construction Engineering and Management, ASCE, 125(5), 295-303. Tucker, R. L. et al. (1982). "Implementation of Foreman-Delay Surveys." J. Construction Devision, ASCE, 108(4), 577-591. Wannacott, T. H. and Wannacott, R. J. (1987). Regression: A Second Course in Statistics, Malabar, Florida Werkmeister, R. F., Luscher, B. L. and Hancher, D. E. (2000). "Kentucky Contract Time Determination System." Transportation Research Record, n.1712, P185-195 Werkmeister, R. F., Luscher, B. L. and Hancher, D. E. (2000). "Kentucky Contract Time Determination System." Transportation Research Record, n.1712, P185-195 Winch, G., and Carr, B. (2001). "Benchmarking on-site productivity in France and the UK: A CALIBRE Approach." J. Construction Management and Economics, V.19, 577-590. 334 VITA Yao-Chen Kuo was born in Taichung, Taiwan, on September 1st, 1970. He is the first son of Fu-Hsiung Kuo and Chiung-Hua Tseng. He attended the National Chung-Hsin University in 1988 and received his Bachelor of Civil Engineering in 1992. After he graduated from college, he worked as a research assistant of the National Science Association for eight months before he entered the Ohio State University for his master program. In November of 1994, he received his Master of Science from the civil engineering department of the Ohio State University. After he completed his Master study, Yao-Chen went back to During Taiwan and worked in the Highway Construction Industry for six years. the first four years, he worked for the second largest general contractors in Taiwan and received the training on many aspects such as managing field works, cost estimation, scheduling and contracting. In 1999, he worked as a consultant engineer in the Sinotech Engineering Consultant Company which is one of the best consultant companies in Taiwan. Yao-Chen received the training on schedule and quality control of highway construction projects during the two years working in the Sinotech. He worked for the Sinotech until he entered the University of Texas to study for his Ph.D. program in January 2001. Permanent Address: 280-1 Chungcheng Rd Wufeng Hsang, Taichung, Taiwan This dissertation was typed by the Author. 335

Find millions of documents on Course Hero - Study Guides, Lecture Notes, Reference Materials, Practice Exams and more. Course Hero has millions of course specific materials providing students with the best way to expand their education.

Below is a small sample set of documents:

University of Texas - BESTMD - 026
Copyright by Michael Douglas Best 2002Fluorescent Sensors For the Detection of Analytes In Solutionby Michael Douglas Best, B.S.DissertationPresented to the Faculty of the Graduate School of The University of Texas at Austin in Partial Fulfill
University of Texas - CHANGW - 026
Copyright by Wen-Jong Chang 2002The Dissertation Committee for Wen-Jong Chang Certifies that this is the approved version of the following dissertation:Development of an In Situ Dynamic Liquefaction TestCommittee: Ellen M. Rathje, Supervisor Ke
University of Texas - TOWRYK - 022
Copyright by Kristy Lynne Towry 2002The Dissertation Committee for Kristy Lynne Towry Certifies that this is the approved version of the following dissertation:Control in a Teamwork Environment The Impact of Social Ties on the Effectiveness of M
University of Texas - BAIS - 18804
University of Texas - HWANGS - 029
344 '3 $ 0 ( &amp;$ ! #! 21)'% #&quot; &amp; F&amp; t n 5 #&quot; Qv ! ! } n n ! # &quot;|{z6 @ $ y $ n % 5 xm # w ! &quot; l ! $ &quot;7dF t v u s R X b $ T q r 88 m n m 8 V o E @ &amp; # %pC $ $ k%HB R $ l &amp; C C c F&amp; EADD2h u g g i g u h u h g kj%ri7d1di
University of Texas - LYONDC - 026
Copyright by Donna Carolyn Lyon 2002The Dissertation Committee for Donna Carolyn Lyon Certifies that this is the approved version of the following dissertation:Achievement Through Small-Group Discussion Sessions in Large General Chemistry Lecture
University of Texas - KULASM - 042
Copyright by Mary-Anne Kulas 2004The Dissertation Committee for Mary-Anne Kulas certies that this is the approved version of the following dissertation:Mechanical and Microstructural Characterization of Commercial AA5083 Aluminum AlloysCommitte
University of Texas - GUOJ - 02612
Copyright By Jessica Yingchieh Guo 2004The Dissertation Committee for Jessica Yingchich Guo certifies that this is the approved version of the following dissertation:ADDRESSING SPATIAL COMPLEXITIES IN RESIDENTIAL LOCATION CHOICE MODELSCommittee
University of Texas - SMITHD - 022
Copyright by Daniel James Smith 2002The Dissertation Committee for Daniel James Smith Certifies that this is the approved version of the following dissertation:PATTERNS OF VARIATION IN SPANISH/ENGLISH BILINGUALISM IN NORTHEAST GEORGIACommittee:
University of Texas - KIMJ - 86337
Copyright by Jin Seok Kim 2004The Dissertation Committee for Jin Seok Kim certifies that this is the approved version of the following dissertation:NEIGHBORHOOD EFFECTS ON THE ETIOLOGY OF CHILD MALTREATMENT: A MULTILEVEL STUDYCommittee:A. Jam
University of Texas - MINB - 57653
Copyright by Byung-Soon Min 2004The Dissertation Committee for Byung-Soon Min certifies that this is the approved version of the following dissertation:Analysis of the Fuel Economy Potential of a Direct Injection Spark Ignition Engine and a CVT i
University of Texas - HOKEJK - 042
Copyright by Julia Kathleen Hoke 2004The Dissertation Committee for Julia Kathleen Hoke certifies that this is the approved version of the following dissertation:Parents as Coping Resources for Adolescents with Learning DisabilitiesCommittee:
University of Texas - GAIX - 85157
Copyright by Xiuli Gai 2004The Dissertation Committee for Xiuli Gai certies that this is the approved version of the following dissertation:A Coupled Geomechanics and Reservoir Flow Model on Parallel ComputersCommittee:Mary F. Wheeler, Superv
University of Texas - KIMJ - 87338
Copyright by Jinman Kim 2004The Dissertation Committee for Jinman Kim Certifies that this is the approved version of the following dissertation:PARAMETRIC DESIGN METHODOLOGY AND VISUALIZATION FOR SINGLE CURVATURE TENSEGRITY STRUCTURESCommittee:
University of Texas - HAOY - 96662
Copyright by Yung-Wei Hao 2004The Dissertation Committee for Yung-Wei Hao Certifies that this is the approved version of the following dissertation:Students Attitudes toward Interaction in Online Learning: Exploring the Relationship between Attit
University of Texas - LEEJ - 99355
Copyright by June Hui Min Lee 2004The Dissertation Committee for June Hui Min Lee certifies that this is the approved version of the following dissertation:PREDICTING HOW EARLY AND HOW MUCH YOUNG CHILDREN USE TELEVISION AND COMPUTERS: THE ROLE OF
University of Texas - PARDOD - 042
Copyright by David Pardo 2004The Dissertation Committee for David Pardo certies that this is the approved version of the following dissertation:INTEGRATION OF HP -ADAPTIVITY WITH A TWO GRID SOLVER: APPLICATIONS TO ELECTROMAGNETICSCommittee:L.
University of Texas - PUTZKW - 042
Copyright by Karl William Putz 2004The Dissertation Committee for Karl William Putz Certifies that this is the approved version of the following dissertation:The Dynamic Mechanical Response of Polymer-Based Nanocomposites and Network GlassesCom
University of Texas - CHOHSS - 042
Copyright by Suk-Joo Stephen Choh 2004The Dissertation Committee for Suk-Joo Stephen Choh certifies that this is the approved version of the following dissertation:Microfacies and Depositional Environments of Selected Pennsylvanian Calcareous Alg
University of Texas - LEEB - 06041
Copyright by Byung-Kwan Lee 2004The Dissertation Committee for Byung-Kwan Lee Certifies that this is the approved version of the following dissertation:The Effects of Product Knowledge on Product Memory and Evaluation in Competitive versus Non-Co
University of Texas - SHONOS - 042
Copyright by Sarah Shono 2004The Dissertation Committee for Sarah Shono Certifies that this is the approved version of the following dissertation:Good ESL Teaches: From the Perspectives of Teachers &amp; Adult LearnersCommittee: David Schwarzer, Su
University of Texas - RUEDAD - 022
Copyright by David Joseph Rueda 2002The Dissertation Committee for David Joseph Rueda Certifies that this is the approved version of the following dissertation:Career Perspectives of Mexican-American Male Superintendents in obtaining the position
University of Texas - QUINNL - 022
Copyright By Laura Ann Quinn 2002The Dissertation Committee for Laura Ann Quinn Certifies that this is the approved version of the following dissertation:Examining Community Stakeholder Relationships From a Communication PerspectiveCommittee: _
University of Texas - VANOJA - 026
Copyright by John Andrew Vano 2002The Dissertation Committee for John Andrew Vano certifies that this is the approved version of the following dissertation:A Nash-Moser Implicit Function Theorem with Whitney Regularity and ApplicationsCommittee
University of Texas - POWIST - 022
Copyright by Terry George Powis 2002The Dissertation Committee for Terry George Powis Certifies that this is the approved version of the following dissertation:An Integrative Approach to the Analysis of the Late Preclassic Ceramics at Lamanai, Be
University of Texas - TATEJA - 026
Copyright by Jennifer Alane Tate 2002The Dissertation Committee for Jennifer Alane Tate certifies that this is the approved version of the following dissertation:Systematics and Evolution of Tarasa Philippi (Malvaceae): An enigmatic Andean polypl
University of Texas - SONGMK - 026
Copyright by Moo Kyoung Song 2002The Evolution of Sonata-Form Design in Ludwig van Beethoven's Early Piano Sonatas, WoO 47 to Opus 22byMoo Kyoung Song, M. M.Dissertation Presented to the Faculty of the Graduate School of the University of Tex
University of Texas - TIUCI - 026
011 0 )'(&amp;%&quot; ! # &quot; $ # r g # jhF &quot; n 3U 3AqP39p &quot; @ ' &quot; n g o2W W&quot; jhF 9&quot; f g k i g m2 hF ed ljhF 9&quot; f W2 d ) 72 7 F eP3&quot; W R 2 ) ) W ) &quot; W &quot; 873333I3&quot;92I' a22 @ @ 79AAg e g w dyAs d i b cc r%hu i e g p i
University of Texas - BOCKRD - 029
Copyright by Robert Davis Bock 2002The Dissertation Committee for Robert Davis Bock Certies that this is the approved version of the following dissertation:Gravitation and ElectromagnetismCommittee:William C. Schieve, Supervisor Larry Horwitz
University of Texas - KERNJT - 029
Copyright by Jonathan Thurston Kern 2002The Dissertation Committee for Jonathan Thurston Kern Certifies that this is the approved version of the following dissertation:Studies on 3,4,9,10-Perylenetetracarboxylic Acid Diimide Based Ligands as G-Qu
University of Texas - LOVEDJ - 042
Copyright by David James Love 2004The Dissertation Committee for David James Love certifies that this is the approved version of the following dissertation:Feedback Methods for Multiple-Input Multiple-Output Wireless SystemsCommittee:Robert W