粒子群算法求解任务可拆分项目调度问题

Particle swarm optimization for resource-constrained project scheduling problems with activity splitting

首先针对任务可拆分的项目调度问题,提出一种带有局部搜索的粒子群算法LSPSO;然后采用基于任务排列的粒子表示方法,将遗传算法中的定位交叉引入粒子的更新过程中,并采用局部搜索技术对更新后的粒子进行改进;最后对Patterson测试集中110个问题实例进行了测试,实验结果表明,算法LSPSO具有较快的速度,所给出的调度方案较优.

A local search particle swarm optimization （LSPSO） is proposed to solve the resource constrained project scheduling problem （RCPSP） with activity splitting. The LSPSO makes use of a permutation based particle representation and an updating mechanism with one-point crossover. Then, a local search technique is adopted to improve the quality of the updated particles. Finally, the algorithm is tested on the instance set Patterson, and the results show that the LSPSO is an alternative and efficient optimization methodology for solving the RCPSP.