基于改进离散粒子群禁忌算法的并行测试任务调度

针对并行测试任务调度问题特点,提出了一种解决并行测试任务调度的离散粒子群禁忌算法,设计了符合并行测试任务调度特点的粒子初始化编码方式、粒子位置转移公式以及禁忌搜索邻域。通过与禁忌搜索相结合,避免了离散粒子群算法早熟问题。通过实验以及算法对比,表明提出的算法是快速有效的,能够很好地解决并行测试任务调度问题,求得测试总时间最短的任务调度序列。

A Parallel Test Task Scheduling Algorithm Based on Improved Discrete Particle Swarm Optimization and Tabu Search

According to the characteristics of parallel test task scheduling,a novel discrete particle swarm optimization-tabu search algorithm is presented for optimizing the parallel test task scheduling.The method of initialization,the update model for new positions and neighbor set of tabu search are designed.Combining with the PSO and TS,the problem of PSO which is easy to be trapped in local optimal is avoided.The proposed algorithm is proved effectively after comparing with other algorithms.