The George Washington University

School of Engineering & Applied Science

Department of Engineering Management and Systems Engineering
O-Campus Programs Oce
EMSE 6760-DDL Discrete Systems Simulation
CLASS TIMES AND DATES: 9:00am - 12:30pm ET, Jun. 21, 28; Jul. (not 5), 12,
19, 26; Aug. 2, 9, 16, 23, 2014.
INSTRUCTOR: Dr. C. J. Willy
email: [email protected]
phone: 301-904-6894 (7:30p.m. to 8:30p.m. EST, Tu, Th)
COURSE MATERIALS
Matlab Simulation Software (www.mathworks.com/academia/student version)
S.M. Ross, Simulation, Academic Press
Hanselman and B. Littleeld, Mastering Matlab, Prentice-Hall (recommended)
DESCRIPTION
EMSE 6760 covers at doctoral level the theory and practice of stochastic model sim-
ulation. The course is suitable for aspiring systems engineering Ph.D. candidates,
and others interested in the practical aspects of constructing computerized simula-
tion studies to analyze and interpret real phenomena. Students will learn to apply
results of these analyses to problems in a wide variety of elds to obtain eective,
accurate solutions and make predictions about future outcomes. The course aims to
provide a comprehensive, in-depth, and current guide for constructing probability
models and simulations. Potential example applications include numerical integra-
tion, insurance risk, and inventory control. Students will use the Matlab simulation
software, www.mathworks.com.
GRADING
Grading will be via weekly homework, Blackboard forum participation, and two
exams, with the following weights:
Blackboard Forum Participation 5%
Homework 20%
Mid-term exam 30%
Final exam 45%
HOMEWORK
Homework will be assigned weekly, and solutions must be delivered via the desig-
nated Blackboard dropbox by the start of the next class. As homework answers
will be covered in class and posted shortly thereafter, late homework will not be
accepted.
Collaboration on homework is permitted. However, students may NOT collaborate
to the extent that their answers are identical. Collaboration via the Blackboard
technical forum is encouraged.
1
EXAMS
Both examinations will be open book. Students may NOT collaborate on the ex-
aminations. Tentatively, the mid-term will be available after the 5th class and is
due the following Friday evening; and the nal examination will be available after
the last class and is due the next evening (Sunday). Exam solutions are due 8:00pm
ET on the due date, and must be delivered via the designated Blackboard dropbox.
ACADEMIC INTEGRITY
Compliance with the George Washington University Code of Academic Integrity
(see www.gwu.edu/ntegrity/code.html) will be monitored closely throughout the
course. Accordingly, the instructions provided with each exam will direct you to
submit the following statement with your solutions:
I, (your name), arm that I have completed this examination in accor-
dance with the Code of Academic Integrity. //signature//
LEARNING OBJECTIVES
Students will be able to use the following concepts in a variety of problem-solving
applications:
1. Elements of Probability Overview of relevant probability theory.
2. Random Numbers Pseudo-random number generation, using random num-
bers to evaluate integrals.
3. Generating Discrete Random Variables The inverse transform method,
generating a Poisson random variable, generating binomial random variables,
the acceptance-rejection technique, the composition approach, generating ran-
dom vectors.
4. Generating Continuous Random Variables The inverse transform algo-
rithm, the rejection method, generating a Poisson process, generating a non-
homogeneous Poisson process.
5. The Discrete Event Simulation Approach Selected examples of simula-
tion via discrete events.
6. Statistical Analysis of Simulated Data The sample mean and sample
variance, interval estimates of a population mean.
7. Variance Reduction Techniques The use of antithetic variables, the use of
control variates, variance reduction by conditioning.
8. Statistical Validation Techniques Goodness of t tests, the chi-square
goodness of t test for discrete data, the Kolmogorov-Smirnov test for con-
tinuous data.
9. Markov Chain Monte Carlo Methods Markov chains, the Hastings-Metropolis
algorithm, hidden Markov models, continuous-time Markov chains (as time
permits.)
2
GENERAL COURSE INFORMATION
1. Blackboard. It is the students responsibility to consult the Blackboard site for:
general course information
lecture notes and homework
announcements
ongoing assessment - students are expected to regularly monitor the gradebook
in Blackboard, and to report any errors in a timely fashion.
2. Discussion Fora. Technical and non-technical questions can be directed to the
appropriate Blackboard forum.
(a) Only students participation on the technical forum will contribute to their
assessment. To gain maximum credit students should ask and/or answer at
least 3 questions on the technical forum on a weekly basis.
(b) Please post any typographical or spelling errors in the typos forum.
(c) Please post any other general course discussion to the general forum.
3. File Naming. Students are to use the following naming convention when sub-
mitting Matlab .m les to the instructor: C3Q1.m (chapter 3 homework problem
1 solution, e.g.), MQ2a.m (mid-term exam problem 2a solution, e.g.) and FQ3.m
(nal exam problem 3 solution, e.g.). A .m le containing a function can be given
a more descriptive name.
4. Email. Students will be contacted using their George Washington University
(gwu.edu) email addresses. Note that email sent to this address can be automati-
cally forwarded to another address.
5. Acknowledgments. The assistance of W.J.J. Roberts in the preparation of the
course materials is gratefully acknowledged.
3
Student Information Sheet
Name
Nickname (if any)
Previous education
(yr/school/degree/major)
Brief description of your job,
if employed
How you can be reached Home:
Oce:
Email:
4