结合电商物流特点,探讨混合时间窗车辆路径问题,设计了混合时间窗惩罚函数和模糊预约时间函数,建立了基于混合时间窗约束的多目标车辆路径模型。设计了改进智能水滴算法对模型进行求解,改进节点概率选择方式来保护优秀水滴,设置路径泥土量最大和最小限制以防止算法提前进入收敛。最后,运用实际案例模拟计算,与传统智能水滴算法计算结果对比分析。结果表明,改进智能水滴算法求解混合时间窗下多目标电商物流路径优化问题,能够以很高的概率获得更优的全局最优解,是求解这类问题有效算法。

Combining the characteristics of e-commerce logistics, studying the problem of hybrid time window vehicle routing , taking into account the constraint conditions of vehicle capacity, mixing time window and satisfaction degree, The multi-objective programming model is established to minimize the sum of vehicle fixed costs and vehicle operating costs and the minimize penalty cost. Based on the basic principle of the traditional intelligent water droplet algorithm, a fast algorithm is designed to solve the problem of multi-target vehicle routing problem. In the end, The practical case is used to simulate and compare with the calculation results of the traditional water droplet algorithm. The results showed, Using intelligent water droplets algorithm for solving mixed time window path optimization problems, multi-objective electric business logistics can get better global optimal solution with high probability, is an effective algorithm to solve the problem.

F252

国家自然科学基金项目（面上项目，重点项目，重大项目）；上海市自然科学基金项目