摘要
多配送中心物流车辆调度问题是一类实用性很高的NP难解问题。针对标准差分进化算法进化过程中缺乏动态调整,进化后期由于种群多样性的降低,算法容易陷入早熟收敛的问题,提出了一种改进的差分进化算法。该算法在变异过程中动态自适应地调整缩放因子,在交叉过程中通过高斯扰动增加种群的多样性,在变异操作之后,加入新的选择机制。将该算法应用于多配送中心物流车辆调度问题,建立了数学模型,介绍了该算法的详细实现过程。仿真通过和遗传算法和标准差分进化算法比较,表明该算法具有更好的寻优效果,从而证明了该算法应用于该问题的可行性和有效性。
Multi distribution center logistics vehicle scheduling problem is a kind of NP combination problem which possesses important practical value. Standard Differential Evolution（DE）algorithm is lack of dynamic adjustment in the evolutionary process. The diversity of species is decreased in the later stage of evolution and the algorithm can easily fall into premature convergence problem. In order to overcome these problems, an Improved Differential Evolution（IDE）algorithm is put forward. The algorithm dynamically adjusts the zoom factor in the process of mutation, increases the diversity of population through Gaussian disturbance in the process of cross and joins a new selection mechanism after the mutation operator. The algorithm is applied to multi distribution center logistics vehicle scheduling problem. The mathematical model is established and the detailed implementation process of the algorithm is introduced. The simulation results show that the algorithm has better optimization effect than GA and DE, which proves the feasibility and validity of the algorithm applied to the problem.
出处
《计算机工程与应用》
CSCD
2014年第3期232-235,共4页
Computer Engineering and Applications
基金
国家自然科学基金(No.61261029)
甘肃政法学院重点科研资助项目(No.GZF2013XZDLW008)
甘肃省财政厅2012年度高校基本科研业务费项目(甘财教[2012]129号)
关键词
多配送中心
物流车辆调度问题
NP问题
改进差分进化
高斯扰动
multi distribution centers
logistics vehicle scheduling problem
NP problem
improved differential evolution
Gaussian disturbance
