{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

CHAPTER10a - CHAPTER Duxbury Thomson Learning Making Hard...

Info iconThis preview shows pages 1–6. Sign up to view the full content.

View Full Document Right Arrow Icon
• A. J. Clark School of Engineering •Department of Civil and Environmental Engineering Third Edition CHAPTER 10 Making Hard Decision Duxbury Thomson Learning ENCE 627 – Decision Analysis for Engineering Department of Civil and Environmental Engineering University of Maryland, College Park Using Data FALL 2003 By Dr . Ibrahim. Assakkaf CHAPTER 10. USING DATA Slide No. 1 ENCE 627 ©Assakkaf Introduction Source for information about probabilities and historical data
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
CHAPTER 10. USING DATA Slide No. 2 ENCE 627 ©Assakkaf Introduction So far we have seen two ways to calculate probability for decision models. – Subjective probabilities (chapter 8) – Theoretical probability models (chapter 9) In chapter 9, given a particular probability model, we had just assumed certain parameter values for the any particular distribution. CHAPTER 10. USING DATA Slide No. 3 ENCE 627 ©Assakkaf Introduction Example: – Poisson model for tornadoes occur in a particular area an average of two times a year. In this case, λ = 2/year. – The parameters u n and α n in the example of the Extreme Value Distribution, Type I for maximum wind velocity V n were assumed as ( ) 9157 . 57 17055 . 0 5772 . 0 3 . 61 α γ μ and 17055 . 0 7.52 6 σ 6 α 2 2 2 2 = = = = = = n X n X n n n u π π
Background image of page 2
CHAPTER 10. USING DATA Slide No. 4 ENCE 627 ©Assakkaf Introduction Now we will learn to calibrate models to data finding the best parameters CHAPTER 10. USING DATA Slide No. 5 ENCE 627 ©Assakkaf Using Data to Construct Probability Distributions Imagine that you are in charge of a manufacturing plant, and you are trying to develop a maintenance policy for your machines. You may collect the following data over 260 days: 11 days Two failures 32 days One failure 217 days No failures
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
CHAPTER 10. USING DATA Slide No. 6 ENCE 627 ©Assakkaf Using Data to Construct Probability Distributions These data lead to the following relative frequencies, which could be used as estimates in your analysis: Data collected: out of 260 days (= 52 weeks × 5 days/week): 0.042 = 11/260 Two failures 1.000 0.123 = 32/260 One failure 0.835 = 217/260 No failures Basically one year’s worth of working days CHAPTER 10. USING DATA Slide No. 7 ENCE 627 ©Assakkaf Using Data to Construct Probability Distributions The only serious consideration to keep in mind is that you should have enough data to make a reliable estimate of the probabilities.
Background image of page 4
CHAPTER 10. USING DATA Slide No. 8 ENCE 627 ©Assakkaf Using Data to Construct Probability Distributions Relative frequency histogram for machine failure: 32/260 11/260 1.00 0.75 0.50 0.25 0.00 0 1 2 Machine Failures Relative Frequency 217/260 CHAPTER 10. USING DATA Slide No. 9 ENCE 627 ©Assakkaf Decision-tree Representation of Uncertainty Regarding Machine Failures Note: – The data requirements depend on the particular problem, but the minimum should be approximately five observations in the least likely category. The other categories, of course, will have more observations.
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Image of page 6
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}