基于聚类与改进蚁群搜索的配送中心车货匹配方法

A Method for Matching Vehicles and Goods in Distribution Centers Based on Clustering and Improved Ant Colony Search

文章针对物流配送中心的配送车辆在运输过程中的装卸货时长、配送车辆行驶时长以及车辆的装载率等问题,构建以配送中心的运营后成本为目标的车货匹配模型,在模型中考虑了装卸货时长、车辆行驶时长以及车辆的装载率等现实因素,通过对该模型的求解为配送中心提供最优匹配方案。该模型先将货物的配送点进行聚类,随后通过对蚁群算法蚂蚁选择下一节点的概率公式进行了改进,将下一节点需要配送货物的重量加入了考量,并采用了一种奖惩机制用作蚁群算法信息素的更新策略和一种随迭代次数而变化的信息素挥发因子,用于提高传统的蚁群算法的收敛速度。用于求解该车货匹配模型。然后通过仿真实验验证模型和算法的有效性,为物流配送中心的车货匹配提供了一个新的解决思路。

This article focuses on the issues of loading and unloading time,driving time,and loading rate of distribution vehicles in logistics distribution centers during transportation.A vehicle cargo matching model is constructed with the goal of post-operational costs of the distribution center.The model takes into account practical factors such as loading and unloading time,vehicle driving time,and vehicle loading rate.By solving the model,the optimal matching scheme is provided for the distribution center.The model first clusters the distribution points of goods,and then improves the probability formula of ant colony algorithm ants selecting the next node,taking into account the weight of goods to be distributed at the next node,and uses a reward and punishment mechanism as a new strategy of ant colony algorithm pheromone and a pheromone volatilization factor that changes with the number of iterations to improve the convergence speed of traditional ant colony algorithm and to solve the vehicle cargo matching model.Then,the effectiveness of the model and algorithm was verified through simulation experiments,providing a new solution for the vehicle cargo matching in logistics distribution centers.