Loading...

Meta-Heuristic Methods for Solving Project Scheduling and Shop Scheduling Problems

Nasiri Khansari, Mohammad Mahdi | 2011

586 Viewed
  1. Type of Document: Ph.D. Dissertation
  2. Language: Farsi
  3. Document No: 42221 (01)
  4. University: Sharif University of Technology
  5. Department: Industrial Engineering
  6. Advisor(s): Kianfar, Farhad
  7. Abstract:
  8. The main purpose of this thesis is to propose effective meta-heuristic algorithms for solving the scheduling problems. We consider the problems that are associated with two areas: project scheduling and shop scheduling. In project scheduling area, only the resource constrained project scheduling problem (RCPSP) is considered. For solving this problem, a hybrid particle swarm optimization (PSO) algorithm is developed. The proposed algorithm outperforms other existing PSO algorithms for the problem. However, most of the contributions of this thesis are related to the second area. The shop scheduling problems of concern include: job shop, stage shop and partial job shop. We developed two algorithms for the job shop scheduling problem. The first algorithm uses the path relinking with N1 neighborhood and tabu search with N6 neighborhood and its guideposts. In addition, we proposed an effective method for updating the topological order and computation of heads and tails in N6 neighborhood. The first algorithm found six new upper bounds for Taillard benchmarks that are also available at his website. Moreover, we introduced new conditions for preventing the cycle generation in N6 neighborhood. The second algorithm applies these conditions in its tabu search procedure. The global equilibrium search is also used in the second algorithm. Furthermore, the second algorithm for job shop scheduling found new upper bounds for 16 problem instances. The other problem considered in this thesis is a new problem called stage shop. In this problem, the operations of each job are partitioned into some consecutive stages so that no precedence relation exists among the operations of a stage. In order to solve this problem, N6 neighborhood is extended and applied in a tabu search framework. Finally, the last considered problem is the partial job shop. For solving this problem, the scatter search is used in combination with path relinking and tabu search.

  9. Keywords:
  10. Project Scheduling ; Tabu Search Algorithm ; Variable Neighborhood Search ; Job Shop ; Particles Swarm Optimization (PSO) ; Stage Shop

 Digital Object List

 Bookmark

No TOC