SEOR Fall 2009 Seminar Series
Date: November 20 (Friday)
Time: 11:00AM-12:00PM
Venue: Engineering Building, Room 4801
***********************************************
K. Preston
White, Jr., Ph.D.
Professor of
Systems and Information Engineering
University of
Virginia
***********************************************
Title: Acceptance Sampling as a Tool for Verifying Requirements using
Monte Carlo Simulation
Abstract: Developed by quality engineers as an efficient statistical
approach to verifying the acceptability of procured items within
production environments, acceptance sampling can be generalized to
other verification problems that rely on the outcomes of sampling experiments. This presentation illustrates
how the acceptance sampling can be adapted to the verification of
design requirements using Monte Carlo
simulation. We consider probabilistic requirements
expressed as limit standards, specifying the performance indicator
for conforming Monte Carlo trials; the minimum limiting
proportion of conforming trials; and the maximum risk of accepting
a nonconforming design. For such requirements,
an attributes acceptance plan prescribes the number of simulation
replications that must be run and the maximum number of nonconforming
replications allowable. The derivation single-sample attribute
acceptance plans for the lunar mass-delivery requirement for NASA’s
Constellation Program (CxP) is provided as an
example. The derivation of potentially more
efficient variables acceptance plans for applications with
limited computational budgets also is explored.
Speaker Information
K.
Preston White, Jr., is Professor of Systems and
Information Engineering at the University of Virginia. His
research interests include Monte Carlo and discrete-event simulation,
statistical analysis and data mining, probabilistic engineering design,
and process and quality control. He is particularly interested
in the integration of decision and information technologies and
applications in aerospace, manufacturing, distribution,
Instructor: Chun-Hung Chen
Email: cchen9@gmu.edu
Office: Engineering Building (Academic VI), Room 2213
Phone: 703-993-3572
Fax: 703-993-1521
Office Hours: Tuesday 5:00 - 6:00 PM and Wednesday 3:30 - 4:30
PM
Teaching Assistant: Mr. Ben Crain
Email: bcrain@gmu.edu
Office: Engineering Building (Academic VI), Room 2216
Office Hours: Monday 5:30 - 6:30 PM, or by appointment.
Course Description:
Examples of discrete-event systems are all around us: multiteller banks; computer networks; automated manufacturing systems; airport terminals; and traffic control systems. In order to efficiently manage and operate these systems, it is often necessary to apply simulation to study their performance since no closed-form analytical solutions exist for such problems. This course deals with this category of systems. Topics will include stochastic modeling, random number generators, discrete-event simulation approaches, simulated data analysis, and simulation variance reduction techniques. Simulation software packages are extensively used in this class. Each student is expected to produce successful simulations.
Prerequisites: OR 542, or STAT 346, or STAT 354, or equivalent, and one scientific programming language
Grading: Homework 35%; Midterm 30%; Term Project 35%.
Required Text: A. M. Law, "Simulation Modeling & Analysis," 4th Ed., 2007. (All earlier editions work well for this class.)
Recommended Text: W. D. Kelton, R. P. Sadowski,
and D. T. Sturrock, "Simulation With Arena," 5th Edition,
2010. You may have a question whether you need to
buy this book. ARENA is the major simulation software used in this class. Since
ARENA is very powerful, many earlier students used it to do their term
projects. It is highly recommended that each project team buys at least one
copy of this book. If you can not find the 5th Edition as suggested, the 2nd,
3rd, or 4th edition of this book is fine too. But the 1st Edition is too
obsolete for this class.
ARENA Software: ARENA is the major simulation software used in this class. The education version of Arena is free of charge if you use it for class homework or term project. You can download the software at the Arena Book Web Site. Please read the instructions in the appendix of the book carefully before installation. If you have a Windows-based computer, you can install Arena on your own PC. In addition, Arena Version 12.0 is available at the IT&E PC Lab. Please note that Arena version up to version 11 is not supported on Windows Vista and click "Issues about Installing Arena at Windows Vista" for details.
Professional Version of ARENA: The student version of Arena is essentially the same as professional version except the limit on the size of model you can run. There are a row of PCs (#23 ~ 30) in IT&E Computer Lab (Room 1506 in the Nguyen Engineering Building) installed with professional version (version 12.0). The professional version allows you to run much bigger models.
Exam:
In class, Wednesday, November 4.
There is no final exam. Make up exam for certified medical reason only. Make up
exam questions will be MUCH MORE DIFFICULT than regular exam questions.
General Rules:
Tentative Course Schedule & Reading
Assignment:
|
|
Topics |
Time (week) |
Reading Assignment |
|
1 |
Introduction and rationale |
0.5 |
|
|
2 |
Basic event scheduling simulation |
1 |
Sections 1.1~1.4.3 |
|
3 |
Review of basic probability and statistics |
0.5 |
Sections 4.1~4.2 |
|
4 |
Simulation Software |
3 |
Sections 3.1~3.3, Handouts on the simulation software ARENA. skim Sections 3.4~3.9. Must read: Chapters 3~ 4 of the recommended text (very useful!!) |
|
5 |
Uniform random numbers |
1 |
Sections 7.1~7.2, skim Section 7.3 |
|
6 |
Generating nonuniform random numbers |
1.5 |
Sections 8.1~8.5 |
|
7 |
Input Modeling |
1 |
Section 4.4 of Recommended Text, and Skim Chapter 6 of Main Text |
|
8 |
Simulation output analysis |
1.5 |
Sections 9.1~9.4, skim Section 9.5 |
|
9 |
|
1 |
Sections 1.8.3 |
|
10 |
Advanced Simulation Optimization |
1 |
Skim Chapter 10 and Handouts on Approximated Confidence Probability |
|
11 |
Term Project Presentation |
1 |
|
Term Project: More details about term project will be given during the semester. Here are some reminders.
Homework Assignments & Others
Excel Add-In for
Crystal Ball and @RISK are two useful packages for
There is another add-in package called SimulaAr
developed by Dr. Luciano Machain
at National University of Rosario in
SimulAr Web Site (Free Excel add-in for Monte Carlo Simulation)
Useful information for HW#1:
In the HW#1, you need to implement simulation code for a two-node system. For your convenience, a version of the C-code for a simplified one-node system is provided below. In addition, the results by running this code is provided. You are supposed to obtain the same results on your own computation platform.
Useful Links to the Manufacturers of Simulation
Software:
A good book for process modeling, simulation,
analysis, and design:
The book "Process Analysis and Improvement" by M. S. Seppanen, S. Kumar, C. Chandra, McGraw-Hill, 2005, gives an excellent introduction about how to use four useful software tools altogether for process modeling, simulation, analysis and design:
1. Microsoft Visio presents process logic as a visual diagram with necessary flows from one entity to the next.
2. Excel is a repository of process data.
3. Arena analyzes the process performance through simulation model.
4. Visual Basic for Application can be used to move data between the above applications.
Other Useful Links:
Go to Professor Chun-Hung Chen's Page