COURSE SYLLABUS

ADVANCED SIMULATION TECHNIQUES

1	Course Title:	ADVANCED SIMULATION TECHNIQUES
2	Course Code:	END 6112
3	Type of Course:	Compulsory
4	Level of Course:	Third Cycle
5	Year of Study:	2
6	Semester:	4
7	ECTS Credits Allocated:	7,5
8	Theoretical (hour/week):	3
9	Practice (hour/week) :	0
10	Laboratory (hour/week) :	0
11	Prerequisites:	END5115 SIMULATION ANALYSIS or any equivalent course
12	Recommended optional programme components:	None
13	Language:	Turkish
14	Mode of Delivery:	Face to face
15	Course Coordinator:	Prof. Dr. ERDAL EMEL
16	Course Lecturers:
17	Contactinformation of the Course Coordinator:	erdal@uludag.edu.tr Tel: 0224 294 2080 Endüstri Mühendisliği Bölümü, Mühendislik Fakültesi Bursa Uludağ Üniversitesi, Görükle, Bursa
18	Website:	http://endustri.uludag.edu.tr
19	Objective of the Course:	Determination of the optimal static and dynamic operating conditions for the stochastic production and service systems and establishment and analysis of their parametric and control simulation models
20	Contribution of the Course to Professional Development	Gives the ability to use reinforcement learning method using neural networks for optimal solution of problems such as maintenance planning, header pricing, stock management under stochastic demand.

Learning Outcomes:

1	Gain the ability to create, verify, and validate simulation models;
2	Have an understanding the principles of simulation system implementation and have knowledge on advanced simulation methods;
3	Be able to develop new simulation methods and have knowledge on when to apply known methods;
4	Gain the ability to comprehend simulation analyses and interpret outputs from a simulation model;
5	Be able to simulate a complex model on a computer environment and have knowledge on up-to-date simulation software;

22	Course Content:

Week	Theoretical	Practical
1	Simulation optimization: an overview Stochastic parametric optimization Stochastic control optimization
2	Parametric Optimization: Response surfaces and neural nets RSM: an overview RSM: details Neuro-response surface methods
3	Parametric optimization: Continuous optimization Discrete optimization (Ranking and Selection, Meta-Heuristics:Simulated Annealing, GeneticAlgorithm)
4	Parametric Optimization: .- Stoch.Grad. and Adap. Search: Discrete Optimization (Stochastic Adaptive Search)
5	Dynamic programming Stochastic processes Markov processes, Markov chains and semi-Markov processes Markov decision problems How to solve an MDP using exhaustive enumeration Dynamic programming for average reward
6	Dynamic programming and discounted reward The Bellman equation: an intuitive perspective Semi-Markov decision problems Modified policy iteration Miscellaneous topics related to MDPs and SMDPs
7	Control Opt.- Reinforcement learning: Curses of DP, RL: Fundamentals, MDPs
8	7. Control Opt.- Reinforcement learning: Semi-Markov decision problems and RL , RL algorithms and their DP counterparts Actor-critic algorithms
9	Control Opt.- Reinforcement learning: Model-building algorithms
10	Control Opt.- Reinforcement learning: Finite Horizon Problems, Function Approximation
11	Control Opt.- Stochastic Search: MCAT Framework, Actor Critics
12	Case studies: Airline Revenue Management. Preventive maintenance. Production line buffer optimization
13	RL Applications in MATLAB
14	RL Applications in PYTHON

23	Textbooks, References and/or Other Materials:	Simulation-Based Optimization, Abhijit Gosavi, Springer, 2015. Discrete Event System Simulation, 4th ed., J.Banks, J.S. Carson, B.L. Nelson, D.M. Nicol, Prentice Hall, 2005. Simulation Modeling and Analysis, 4th ed., Averill M. Law, McGraw-Hill, Inc., 2007. Simulation Using Promodel with CD-Rom, Charles R. Harrell, Biman K. Ghosh, Royce O. Bowden, McGraw-Hill, 2003. Approximate Dynamic Programming: Solving the Curses of Dimensionality, Warren B. Powell, Wiley-Interscience; 1st edition, 2007 Markov Decision Processes: Discrete Stochastic Dynamic Programming, Martin L. Puterman, Wiley-Interscience; 1st edition, 2005
24	Assesment

TERM LEARNING ACTIVITIES	NUMBER	PERCENT
Midterm Exam	1	20
Quiz	0	0
Homeworks, Performances	4	50
Final Exam	1	30
Total	6	100
Contribution of Term (Year) Learning Activities to Success Grade		70
Contribution of Final Exam to Success Grade		30
Total		100
Measurement and Evaluation Techniques Used in the Course	Measurement and evaluation are performed according to the Rules & Regulations of Bursa Uludağ University on Undergraduate Education.
Information

ECTS / WORK LOAD TABLE

Activites	NUMBER	TIME [Hour]	Total WorkLoad [Hour]
Theoretical	14	3	42
Practicals/Labs	0	0	0
Self Study and Preparation	13	8	104
Homeworks, Performances	4	18	54
Projects	1	24	24
Field Studies	0	0	0
Midtermexams	1	2	2
Others	0	0	0
Final Exams	1	2	2
Total WorkLoad			228
Total workload/ 30 hr			7,6
ECTS Credit of the Course			7,5

CONTRIBUTION OF LEARNING OUTCOMES TO PROGRAMME QUALIFICATIONS

	PQ3	PQ6	PQ7	PQ8	PQ9	PQ10	PQ12
LO1	3	5	0	0	0	0	0
LO2	0	0	5	3	4	0	0
LO3	0	0	0	5	4	0	0
LO4	0	0	0	0	5	0	4
LO5	0	0	0	3	3	5	0

LO: Learning Objectives

PQ: Program Qualifications

Contribution Level:

1 Very Low

2 Low

3 Medium

4 High

5 Very High