基于PSO_H优化算法的单机E_T调度问题.pdfVIP

19 9 © Vol. 19 No. 9 2007 5 Journal of System Simulation May, 2007 PSO-H E/T , 100083 E/T(Earliness/Tardiness) (Particle Swarm Optimization integrated with Heuristic PSO-H) OR-Library E/T : TP391.9 A 1004-731X (2007) 09-1933-04 Common Due Date Scheduling Problem on Single Machine Based on PSO-H Algorithm ZHOU Hong, WU Xue-jing, TAN Xiao-wei (Faculty of Economy Management, Beijing Aeronautics and Astronautics University, Beijing100083, China) Abstract: The classical Particle Swarm Optimization (PSO) is a powerful method to find the minimum for a function optimization problem, especially with a continuous solution space. So far, it is seldom used to solve those problems with discrete featuresThe problem of scheduling jobs on a single machine against common due dates with respect to earliness and tardiness (E/T) penalties was dealt. At the same time, a PSO strategy integrated with a kind of heuristic algorithm was proposed, where the heuristic information is composed of the processing time and tardiness penalty for each job. It is indicated that such strategy can significantly improve the performance of the solutions. Benchmark testing from OR-Library demonstrates that the algorithm is both effective and efficient in achieving satisfied solutions for scheduling problems with earliness and tardiness penalties. Key words: PSO; heuristic; optimization; scheduling; E/T 1 H)E/T OR Libra