一种集成最后一公里的四方物流网络设计问题启发式算法

A heuristic algorithm for fourty logistics network design problem with last-mile delivery

最后一公里分销网络可以帮助企业达成高响应性的供应链管理目标,集成最后一公里四方物流网络设计问题成为网络设计的一个重要研究方向.解决该问题需要对分销中心的位置,三方物流的选择、分配以及其车辆路径规划进行决策.在满足车辆路径规划、流守恒等约束条件下,以最小化网络构建费用为目标建立混合整数规划模型.由于该问题的NP-难特性,可将该问题分解成两个子问题并设计两阶段启发式算法,通过迭代算法解决两个子问题.在数值实验中,将启发式算法分别与CPLEX和粒子群优化算法求出的解进行比较,实验结果验证了启发式算法的有效性;同时,将提出的启发式算法成功地应用到实际规模的问题中,表明所提出的算法能够为解决集成最后一公里四方物流网络设计问题提供有效的工具.

A distributor storage network with last-mile delivery can be used to achieve the supply chain management objective of high responsiveness. The fourth party logistics network design problem with last-mile delivery is an important research in the network design problem, which needs to be solved by simultaneously making decisions on issues of the distribution center location, the selection, allocation and routing programming of the third party logistics. A mix-integer programming model is established to minimize the total system cost with satisfying the constraints of vehicle routing programming and fiow conservation and so on. Because the problem is NP-hard, this paper presents a two-phase heuristic algorithm for the fourth party logistics network design problem with last-mile delivery by dividing the problem into two subproblems. The algorithm solves the subproblems in an iterative manner. In the numerical tests, by comparing with the solutions calculated by the CPLEX and the particle swarm optimization algorithm, the heuristic algorithm outperforms the other two algorithms, which illustrates that the heuristic algorithm can provide an efficient solution to the fourth party logistics network design problem with last-mile delivery;meanwhile, the heuristic algorithm is successfully applied in the experiment of practical scale, which also shows that it can offer an effective tool in the field.