Instructor: Chun-Hung Chen
Email: cchen9@gmu.edu
Office: Engineering Building, Room 2213
Phone: 703-993-3572
Fax: 703-993-1521
Office Hours: Tuesday 2:30 - 4:30PM
Course Description:
This class basically is an advanced version and an extension of the basic simulation class OR 635 Discrete System Simulation. The extension includes both depth and breadth. In the depth part, we will cover the advanced materials which are not included in OR 635 course, with focus on how to make simulation faster without getting a faster computer. Examples include some variance reduction techniques. In the breadth part, we will study several useful simulation topics beyond the basics in OR 635. Examples include rare-event simulation, importance sampling, bootstrapping, Quasi Monte Carlo simulation, agent-based modeling, etc.
Since this is a doctoral-level class, in addition to regular lectures, this class will include extensive literature study and research project. Students will get a bit taste of doctoral study. For Master students, this class gives you a chance to see what a Ph.D. study looks like. For Ph.D. students, this class should better prepare you for doing research.
Students will conduct a small-scale research project. The focus of these projects is "simulation-based optimization". Simulation is a popular tool for designing large, complex, stochastic systems, since closed-form analytical solutions generally do not exist for such problems. While the advance of new technology has dramatically increased computational power, efficiency is still a big concern when using simulation for stochastic optimization, in which case many alternative designs must be simulated. A decision maker is forced to compromise on simulation accuracy, modeling accuracy, and the optimality of the selected design. This class will discuss different approaches to address this issue. Students will investigate and/or develop efficient simulation-based optimization techniques in the term projects.
Prerequisite: Students in this class are assumed to have the background of an introductory simulation class such as OR 635 Discrete System Simulation, or OR 645 Stochastic Processes, or permission from the instructor.
Grading: Homework 15%; Special Topic Study 30%; Project Proposal 5%; Project Presentations 15%; Term Project Report 30%; Class Participation 5%.
Required Text: C. H. Chen
and L. H. Lee, "Stochastic Simulation Optimization: An Optimal Computing
Budget Allocation," 2010 (same as OR 635). You can order it from the
university book store or Amazon.com.
A copy is available at Fenwick Library of GMU. The library call number is TA168 .C475 2011.
Recommended Text 1: A. M. Law, "Simulation Modeling & Analysis," 4th Ed., 2007 (same as OR 635, any earlier edition is fine).
Recommended Text 2: C. H. Chen, Q. S. Jia, and L. H. Lee, "Stochastic Simulation Optimization for Discrete Event Systems - Perturbation Analysis, Ordinal Optimization, and Beyond," 2013.
General Rules:
Course Outlines
This class includes three parts: 1) Depth; 2) Breadth; and 3) Research.
1. Lectures in Advanced Simulation (Depth):
|
|
Topics |
Reading Assignment |
|
A |
Review of Stochastic Simulation |
Appendix A |
|
B |
Simulation and Optimization |
Chapter 1 |
|
C |
Computing Budget Allocation |
Chapter 2 |
|
D |
Selecting the Best Alternative |
Chapters 3 & 4 |
|
E |
Selecting an Optimal Set |
Chapter 5 |
|
F |
Simulation-based Optimization |
Chapter 7 |
|
G |
Variance Reduction Techniques |
Chapter 11 of Law |
|
H |
Advanced Simulation Methodologies - Standard Clock Method - Monte Carlo Integration |
Handout-SC & Section 1.8 of Law |
2. Special Topic Study & Presentation (Breadth):
Some possible topics are listed below; but not limited to the list. Each student has to select a group of topics to study and present in the class. The schedule of 2014 presentations is as follow.
|
Date |
Topics |
Presenter |
|
3/18 |
-
Design of Experiments -
Validation & Verification |
Checco |
|
3/18 |
Continuous
Optimization for Stochastic Simulation |
Nicholas |
|
3/25 |
Markov Chain Monte Carlo |
Matsumoto |
|
3/25 |
Simulation
in Healthcare |
Neyshabouri |
|
4/1 |
Faster Simulation |
Guharay |
|
4/1 |
Agent-based Modeling
& Simulation |
McEligot |
|
4/8 |
-
Model Fidelity Evaluation and Selection -
Surrogate Model Based Optimization |
Mahoney |
|
4/8 |
Surrogate Model Based
Optimization |
Taghiyeh |
A good starting point for your study is our text book and the Winter Simulation Conference Proceedings.
Please identify a paper (or prepare a report) which can give a good introduction and overview of the topic you study, and email the paper to the class at least one week before your presentation. You can also consider to send another paper which give more in-depth discussions.
Each student gives a presentation. The length of presentation is around 45 minutes not including Q&A. Please email your presentation to the instructor at least 24 hours in advance. In your presentation, please consider to include the following items:
It is very important to show rigorous and
quantitative results in the presentation. However, it is even important to well
explain the ideas and intuitions of the fundamentals. It is not a good idea to
give too much extensive mathematical formula. Figures and animation are always
welcome. We want both depth and breath. Backups are useful too.
The paper and presentation will be graded by the instructor and the class. All students are required to read the paper before presentation and so will be able to ask good in-depth questions at the presentation.
3. Term Project (Research):
Theme: Efficient Simulation-based
Decision Making
In this research project,
students are expected to investigate a technique for efficient simulation-based
decision making. One possible approach is to integrate the efficient simulation
techniques with optimization methodology. Students have to meet with the
instructor personally in the projects to ensure right progress and discuss
potential research questions. Students will give presentations to the class
about their techniques at the end of the semester. Ideally, the project can lead into a possible publication in a
conference or journal.
The length of presentation is 20 ~ 30 minutes not including Q&A. Please upload your presentation to the computer before the class starts. In your presentation, please consider to include the following items:
Homework Assignments & Handouts:
Useful Links:
Go to Professor Chun-Hung Chen's Page