alternative_system_design

alternative_system_design - SYSC4005/5001 Winter 2009...

This preview shows pages 1–7. Sign up to view the full content.

Professor John Lambadaris SYSC4005/5001 Winter 2009 1 Comparison and Evaluation of Alternative System Designs Winter 2009 Slides are based on the texts: -Discrete Event System Simulation, by Banks et al -Discrete Event Simulation: A first Course, by Leemis and Park

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

View Full Document
Professor John Lambadaris SYSC4005/5001 Winter 2009 2 Purpose ± Purpose: comparison of alternative system designs. ± Approach: discuss a few of many statistical methods that can be used to compare two or more system designs. ± Statistical analysis is needed to discover whether observed differences are due to: ² Differences in design or, ² The random fluctuation inherent in the models.
Professor John Lambadaris SYSC4005/5001 Winter 2009 3 Outline ± For two-system comparisons: ² Independent sampling. ² Correlated sampling (common random numbers). ± For multiple system comparisons: ² Bonferroni approach: confidence-interval estimation, screening, and selecting the best. ± Metamodels

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

View Full Document
Professor John Lambadaris SYSC4005/5001 Winter 2009 4 Comparison of Two System Designs ± Goal: compare two possible configurations of a system ² e.g., evaluation of two routing or scheduling schemes in a computer communication network! ± Approach: the method of replications is used to analyze the output data. ± The mean performance measure for system i is denoted by θ i ( i = 1,2 ). ± To obtain point and interval estimates for the difference in mean performance, namely 1 2 .
Professor John Lambadaris SYSC4005/5001 Winter 2009 5 Comparison of Two System Designs ± Vehicle-safety inspection example: ² The station performs 3 jobs: (1) brake check, (2) headlight check, and (3) steering check. ² Vehicles arrival: Poisson with rate = 9.5 /hour. ² Present system: ± Three stalls in parallel (one attendant makes all 3 inspections at each stall). ± Service times for the 3 jobs: normally distributed with means 6.5, 6.0 and 5.5 minutes, respectively. ² Alternative system: ± Each attendant specializes in a single task, each vehicle will pass through three work stations in series ± Mean service times for each job decreases by 10% ( 5.85, 5.4 , and 4.95 minutes). ² Performance measure: mean response time per vehicle (total time from vehicle arrival to its departure).

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

View Full Document
Professor John Lambadaris SYSC4005/5001 Winter 2009 6 Comparison of Two System Designs ± From replication r of system i , the simulation analyst obtains an estimate Y ir of the mean performance measure θ i . ± Assuming that the estimators Y ir are unbiased: 1 = E(Y 1r ), r = 1, … , R 1 ; 2 = E(Y 2r ), r = 1, … , R 2 ± Goal: compute a confidence interval for 1 2 to compare the two system designs ± Confidence interval for 1 2 (c.i.): ² If c.i. is totally to the left of 0 , strong evidence for the hypothesis that 1 2 < 0 ( 1 < 2 ). ² If c.i. is totally to the right of 0 , strong evidence for the hypothesis that 1 2 > 0 ( 1 > 2 ).
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}

Page1 / 37

alternative_system_design - SYSC4005/5001 Winter 2009...

This preview shows document pages 1 - 7. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online