Modeling and Simulation

Modeling and Simulation
(CS - 503)

Department of Computer Engineering

Bilkent University

Fall 2008

Dr.Çağatay ÜNDEĞER

Part Time Instructor

Department of Computer Engineering - Bilkent University

room : 514

e-mail : cagatay@undeger.com , cagatay@cs.bilkent.edu.tr

web-site : www.cs.bilkent.edu.tr/~cagatay

1. Objective

In this course, modeling and simulation (M&S) methodologies considering both practical and theoretical aspects will be studied in details. By taking the lecture, you will be supported with sufficient knowledge about a wide range of M&S concepts that will lead you to develop your own M&S applications. Also you will be able to learn about some real-word practices in the context of defense industry and game programming.

2. Scope

M&S is a discipline of understanding and evaluating the interaction of parts of a real or theoretical system by designing its representation (model) and executing (running) the model including the time and space dimension (simulation). In that definition, a system is an entity (unit, process, event...), which exists and operates in time and space through the interaction of its parts; a model is a simplified representation of a real or theoretical system at some particular point in time or space intended to provide understanding of the system; and a simulation is the manipulation of a model in such a way that it operates in time or space to summarize it.

In recent decades, M&S has become a very popular research and development area, which can be applied in many areas such as medical research, training & support, industrial engineering designs & presentations, civil engineering designs & presentations, mechanical engineering designs & presentations, nature sciences (e.g. physic, chemistry, biology, meteorology, astronomy), geographic information systems, military decision support, training and entertainment. The focus of this lecture will be theoretical aspects of M&S and real-world practices primarily towards the defense industry and game programming; which includes earth modeling, entity modeling, behavior modeling, sensor & weapon systems modeling, distributed simulations, simulation based optimization and analysis.

3. News

16.10.2008 – Source code of experimental game development environment can be downloaded here.

Source code is for Borland C++ Builder 6.0, but it can be compiled without modification in Visucal C++ except the user interface.

OpenGL component given in the directory “components” should be registered to IDE with options:

Advance Compiler / Instruction Set = Pentium

Advance Compiler / Data Alignment = Quad Word

03.11.2008 – The due dates for the projects are as follows:

1) Proposals (28.10.2008)

2) Introductory Presentations (18.11.2008) ( 10 minutes per presentation )

3) Progress reports (25.11.2008) ( literature survey and draft design/algorithms )

5) Final Presentations (16.12.2008) ( (1) Emre Akatürk, Denizhan Güçer, Cihan Halit, (2) Oğuz Oğuzcan, (3) Engin Bayrak, (4) Gizem Gürcüoğlu, Tuğba Yıldız ) ( 25-35 minutes each )

6) Final Presentations (23.12.2008) ( (5) Zeki Yanlız, (6) Abdullah Bülbül, Mustafa Kabak, (7) İskender Yakın, (8) Bertan Gündoğdu, Adil Yalçın) ( 25-35 minutes each )

7) Final Presentations (30.12.2008) ( (9) Merve Sağlam, Mustafa Sefünç, (10) Erhan Okuyan, (11) Ferhat Kutlu, (12) Tolga Özaslan) ( 25-35 minutes each )

8) Final Reports (30.12.2008 – hard copy to be given in lecture hour) ( literature survey, final design & algorithms, important implementation details, and instructions for demo )

9) Final Demos (06.01.2009) ( all ) ( 8-12 minutes each, do rehearsal and be well-prepared since no delays are accepted )

Grading criteria for presentations are:

Technical Quality
Slide Quality
Clarity of presentation
Usage of voice
Timing
Questions answered

Attandance to presentations will provide 6 points bonus.

Grades for Attandance to Seminar:

No	Name	Last name	Total
1	Emre	Akatürk	33,3
2	Ahmet Engin	Bayrak	33,3
3	Muhammed Abdullah	Bülbül	33,3
4	Denizhan	Güçer	33,3
5	Ramazan Bertan	Gündoğdu	50
6	Gizem	Gürcüoğlu	0,0
7	Cihan	Halit	33,3
8	Mustafa	Kabak	66,7
9	Ferhat	Kutlu	100,0
10	Oğuzcan	Oğuz	66,7
11	Erhan	Okuyan	33,3
12	Tolga	Özaslan	66,7
13	Merve	Sağlam	100,0
14	Mustafa Akın	Sefünç	66,7
15	İskender	Yakın	0,0
16	Mehmet Adil	Yalçın	50
17	İsmet Zeki	Yalnız	66,7
18	Tuğba	Yıldız	33,3

Introductory Presentations

Name	Last name	Total
Emre	Akatürk	97
Ahmet Engin	Bayrak	97
Muhammed Abdullah	Bülbül	73
Denizhan	Güçer	97
Ramazan Bertan	Gündoğdu	98
Gizem	Gürcüoğlu	63
Cihan	Halit	96
Mustafa	Kabak	73
Ferhat	Kutlu	97
Oğuzcan	Oğuz	100
Erhan	Okuyan	100
Tolga	Özaslan	90
Merve	Sağlam	92
Mustafa Akın	Sefünç	92
İskender	Yakın	65
Mehmet Adil	Yalçın	95
İsmet Zeki	Yalnız	93
Tuğba	Yıldız	63

Grades for Progress Reports:

Prj No	Project	No	Name	Last name	Progress Report
1	Design & Simulation of City Flow	12	Tolga	Özaslan	80
2	Building and City Modeling	10	Oğuzcan	Oğuz	80
3	Distributed Weapon/Platform Modeling	2	Ahmet Engin	Bayrak	75
4	Sketching Character Motion	6	Gizem	Gürcüoğlu	75
4	Sketching Character Motion	18	Tuğba	Yıldız	75
5	Fish Modeling	17	İsmet Zeki	Yalnız	70
6	Shooting Range Simulator	3	Muhammed Abdullah	Bülbül	85
6	Shooting Range Simulator	8	Mustafa	Kabak	85
7	Dynamic Path Planning	15	İskender	Yakın	35
8	Terrain and Tree Simulation	5	Ramazan Bertan	Gündoğdu	80
8	Terrain and Tree Simulation	16	Mehmet Adil	Yalçın	95
9	Modeling mountains, water and bushes	13	Merve	Sağlam	95
9	Modeling mountains, water and bushes	14	Mustafa Akın	Sefünç	95
10	Flight Simulator	11	Erhan	Okuyan	85
11	Sensor Modeling and Simulation	9	Ferhat	Kutlu	75
12	Car Simulator	1	Emre	Akatürk	90
		4	Denizhan	Güçer	90
		7	Cihan	Halit	90

Bonus for Attandance to Presentations:

Name	Last name	Total
Tolga	Özaslan	5
Oğuzcan	Oğuz	2
Ahmet Engin	Bayrak	3
Gizem	Gürcüoğlu	4
Tuğba	Yıldız	3
İsmet Zeki	Yalnız	5
Muhammed Abdullah	Bülbül	4
Mustafa	Kabak	2
İskender	Yakın	6
Ramazan Bertan	Gündoğdu	4
Mehmet Adil	Yalçın	5
Merve	Sağlam	6
Mustafa Akın	Sefünç	6
Erhan	Okuyan	4
Ferhat	Kutlu	2
Emre	Akatürk	6
Denizhan	Güçer	6
Cihan	Halit	6

4. Course Outline

• Introduction To Modeling & Simulation

– What is Modeling and Simulation?

– Complexity Types

– Model Types

– Simulation Types

– M&S Terms and Definitions

• Input Data Analysis

– Simulation Input Modeling

– Input Data Collection

1. Data Collection Problems

2. Practical Suggestions

3. Effect of Period of Time

– Input Modeling Strategy

1. Histograms

2. Probability Distributions

3. Selecting a Probability Distribution

4. Evaluating Goodness of Fit

• Random Variate Generation

– Random Numbers

– Random Number Generators

– Random Variate Generation

1. Factors to be considered

2. General principles

1. Inverse Transform Method

2. Acceptance-Rejection Method

3. Composition Method

4. Relocate and Rescale Method

3. Specific distributions

• Output Data Analysis

– Introduction

1. Types of Simulation With Respect to Output Analysis

2. Stochastic Process and Sample Path

3. Sampling and Systematic Errors

4. Mean, Standard Deviation and Confidence Interval

– Analysis of Finite-Horizon Simulations

1. Single Run

2. Independent Replications

3. Sequential Estimation

– Analysis of Steady-State Simulations

1. Removal of Initialization Bias (Warm-up Interval)

2. Replication-Deletion Approach

3. Batch-Means Method

• Comparing Systems via Simulation

– Introduction

– Comparison Problems

1. Comparing Two Systems

2. Screening Problems

3. Selecting the Best

4. Comparison with a Standard

5. Comparison with a Fixed Performance

• Discrete Event Simulations

– Introduction

1. Next-Event Time Advance

2. Arithmetic and Logical Relationships

3. Discrete-Event Modeling Approaches

– Event-Scheduling Approach

– Process-Interaction Approach

– Processes and Resources

• A Discrete Event Simulation Language: General Purpose Simulation System (GPPS)

– Simulation Languages

– GPSS Simulation Language

– A Sample Simulation Model

• A Distributed Simulation Standard: High Level Architecture (HLA)

– Introduction to Distributed Simulations

– Introduction to HLA

– HLA Interface Specification:

1. Federation Management

2. Declaration Management

3. Object Management

– Conclusion

• Entity Behavior Modeling

– Introduction to Artificial Intelligence (AI)

– AI Techniques & Architectures

– Some AI Algorithms/Techniques

1. Finite State Machines

2. Decision Trees

3. Artificial Neural Networks

4. Logic Programming

5. Production Systems

6. Genetic Algorithms

7. Path Planning

8. Script Programming

– Conclusion

5. Course Requirements

The students are expected to develop a small-scale M&S project with a team of preferably at most three people. The project will include documentation, implementation and a class presentation. The project topic can be either proposed by the team or selected from a list of topics provided by the lecturer, which will be towards the development of an experimental game environment on Windows platform. Microsoft Word and C++ programming language ((1) Borland C++ Builder or (2) Microsoft Visual Studio) are preferred for the reporting and development of the projects.

6. Proposed Project Topics

The topics proposed below will be towards the development of an experimental game environment, basic implementation of which (see Figure 1) will be provided with the source code by the lecturer in Borland C++ Builder 6.0 or Microsoft Visual Studio 2005.