带软时间窗物流配送车辆路径问题的并行遗传算法

Parallel Genetic Algorithm for Vehicle Routing Problems with Soft Time Windows of Logistic Distribution

针对一般遗传算法在求解有时间窗车辆路径问题时初始种群的单一性,提出一种新的算法。该算法对不同的种群用不同的初始化方法——随机初始化法和构造初始化法,这种算法改变了过去那种种群内部的平衡。并将该算法所得结果与其他算法进行比较,表明该算法的合理性。

According to the oneness on the initialization of cluster in general genetic algorithms, a kind of new algorithm （parallel genetic algorithm） has been proposed. During initialization,it used two different methods on cluster i. e. randomization and construction method, so it can break the balance of inter-population thoroughly and made optimal solution evolving to high-level. In comparison with other algorithms,the parallel algorithm is an efficient method for vehicle routing problem.