ETM 607

1
ETM 607 Putting It All Together

• Review for MidTerm II
• Apply Lessons Learned in a Team Lab
• Input Modeling
• Absolute Output Analysis
• Relative Output Analysis

2
ETM 607 MidTerm Review

• Chapter 7 Random Number Generation
• Generation Methods
• Linear Congruential Method
• Tests
• Frequency (Chi Squared and Kolmogorov-Smirnov)
• Autocorrelation
• Chapter 8 Random Variate Generation
• Inverse transform method
• Discrete distributions
• Continuous distributions

3
ETM 607 MidTerm Review

• Chapter 9 Input Modeling
• Histograms and Selecting Distribution Families
• Parameter Estimation
• Goodness of Fit Tests
• Chi Squared
• Kolmogorov-Smirnov)
• Fitting Non-Stationary Poisson Process
• Chapter 10 Estimation of Absolute Performance
• Point estimation
• Confidence Intervals
• Initialization Bias
• Determining number of Replications (desired CI
  width)

4
ETM 607 MidTerm Review

• Chapter 12 Estimation of Relative
  Performance
• Comparison of Two systems
• Independent sampling
• CRN approach
• CI overlap or containing the value 0
• Logistics
• Monday 6-9pm, December 3
• Open book and notes
• No computer section
• Computer section - Excel

5
ETM 607 Putting It All Together

• Base Entrance with Guard Booths

2 guards in series
6
ETM 607 Putting It All Together

• Simulation Study Guard Booth Analysis
• Assume you have collected vehicle counts at 15
  minute intervals and the times for guards to
  process vehicles.
• This data has been tabulated in an Excel
  spreadsheet Lesson13.xls.
• A lane can either be manned by one or two
  guards. Times to process vehicles are slightly
  longer in the two guard case to allow for vehicle
  to advance to second guard.

7
ETM 607 Putting It All Together

• Simulation Study Guard Booth Analysis
• Objective Determine the best staffing levels
  throughout the day.
• Alternatives to evaluate how many lanes /
  guards, and if and when to use a one guard or
  two guard lane option.
• An Arena model with logic has been provided.
• As a team, evaluate alternatives and provide a
  recommendation. Back up your recommendation with
  data.

8
ETM 607 Putting It All Together

• Simulation Study Guard Booth Analysis
• Keep In Mind
• Do you need to model every 15 minute interval?
• How are you to develop your input models?
• What is your design of experiment (i.e.
  scenarios)
• How are you going to compare scenarios?
• What performance measures do you believe to be
  important?
• How wide do you want your confidence intervals?
• How will you present your recommendations?