Heuristic Optimization Methods
- explain the basic underlying principles of heuristic search as an optimization method to solve complex problems
- calculate problem complexity and algorithm complexity
- distinguish between exact methods and heuristic methods, and when to apply which methods
- identify the need for heuristics
- explain the methodology of the most commonly used heuristics (Greedy, Simmulated Annealing, Tabu Search, Evolutionary algorithms, Ant Colony optimization, PSO)
- explain the advantages and disadvantages of various heuristic methods
- develop new (hybrid) heuristic methods by applying existing heuristic methods
- evaluate the quality of the solutions obtained with heuristic methods
Forms of Teaching
Lectures are held in two cycles. The first cycle contains 7 weeks of lectures followed by the second cycle lasting 6 weeks, with a teaching load of 2 hours per week.Laboratory
Students are assigned 2 laboratory assignments in which they will independently design, implement and test heuristic algorithms for solving given combinatorial optimization problems.Other
Project: design, implementation and testing of heuristic algorithms for solving a given combinatorial optimization problem.
|Type||Threshold||Percent of Grade||Threshold||Percent of Grade|
|Laboratory Exercises||25 %||25 %||25 %||25 %|
|Seminar/Project||50 %||25 %||50 %||25 %|
|Mid Term Exam: Written||0 %||25 %||0 %|
|Final Exam: Written||0 %||25 %|
|Exam: Written||0 %||50 %|
Week by Week Schedule
- Introduction to optimization; optimization models, combinatorial optimization.
- Complexity theory; categorization of optimization methods; why and when to use heuristics.
- Exact methods: integer programming, branch and bound, dynamic programming.
- Greedy algorithms. Basic concepts of metaheuristics.
- Local search. Greedy randomized adaptive search procedure (GRASP).
- Variable neighborhood search. Large neighborhood search.
- Tabu search.
- Midterm exam
- Simulated annealing.
- Ant colony optimization.
- Particle swarm optimization. Project assignment.
- Evolutionary algorithms.
- Case studies involving hybrid and meta heuristics.
- Project presentations.
- Final exam