生鲜农产品车辆和无人机组合配送路径优化及效果测试  

Optimization and effect testing of the combined distribution route of fresh agricultural products vehicles and drones

作　　者：张荣 庞梦荻 刘斌[2] ZHANG Rong;PANG Mengdi;LIU Bin(Institute of Logistics Science and Engineering,Shanghai Maritime University,Shanghai 201316,China;Business School,University of Shanghai for Science and Technology,Shanghai 200093,China)

机构地区：[1]上海海事大学物流科学与工程研究院,上海201316 [2]上海理工大学管理学院,上海200093

出　　处：《河南农业大学学报》2024年第1期96-105,共10页Journal of Henan Agricultural University

基　　金：国家自然科学基金项目(71971134)。

摘　　要：【目的】为提升生鲜农产品的配送效率,提出生鲜农产品车辆与无人机组合配送路径优化方案。【方法】按照生鲜农产品冷链配送体系要求,采用模糊K-means聚类方法确定生鲜农产品配送中心;定义配送中心位置为车辆和无人机组合配送的起点,以生鲜农产品配送总成本最小为目标函数,利用蚁群算法和无人机物流车协调配送算法求解上述构建的目标函数,获取最低配送总成本的配送路径规划结果;引入安全启发函数优化蚂蚁转移规则,获取安全最高的无人机最优配送路径。再按照该方法,将生鲜农产品按照质量划分为重件和轻件,并计算重件和轻件在配送过程中产生的成本,获取在不同的总配送生鲜农产品快件数量下,优化前与优化后的配送成本。【结果】效果测试表明,该方法具有良好的聚类效果,簇类凝聚度(Davies-Bouldin, DB)指数值均低于0.15,可合理选取配送中心,保证配送中心的覆盖程度。并且通过对比该方法应用前后的数据对比可知,应用前重件和轻件的配送成本明显更高于应用后的成本。【结论】该方法可以更高质量地完成车辆和无人机组合配送路径规划,能够显著降低生鲜农产品的配送总体成本,应用效果满足应用需求,提升配送效率。【Objective】In order to improve the delivery efficiency of fresh agricultural products,an optimized scheme of combining vehicles and unmanned aerial vehicles(UAV)was proposed for the distribution of fresh agricultural products.【Method】Based on the cold chain distribution system requirements for fresh agricultural products,the fuzzy K-means clustering method was used to determine the distribution center of fresh agricultural products,and define the location of the distribution center as the starting point for the combined distribution of vehicles and drones.With the objective function of minimizing the total cost of fresh agricultural product distribution,the ant colony algorithm and drone logistics vehicle coordination distribution algorithm were used to solve the above constructed objective f unction,and obtain the distribution path planning result with the lowest total cost of distribution;a security heuristic function to introduced to optimize ant transfer rules and obtain the safest optimal delivery path for unmanned aerial vehicles.According to this method,fresh agricultural products were divided into heavy and light items by weight,and the costs incurred during the distribution process of heavy and light items were calculated to obtain the distribution costs before and after optimization under different total delivery quantities of fresh agricultural products.【Result】The test results showed that this method had good clustering performance,with cluster cohesion Davies-Boldin(DB)index values below 0.15.It can reasonably select distribution centers to ensure their coverage.And by comparing the data before and after the application of the method,it can be seen that the distribution cost of heavy and light items before the application is significantly higher than that after the application.【Conclusion】The method proposed in this article can achieve higher quality delivery path planning for vehicle and UAV combinations,significantly reducing the overall delivery cost of fresh agricultural products,

关 键 词：蚁群算法 生鲜农产品 无人机 组合配送 路径规划 配送中心位置 

分 类 号：S24[农业科学—农业电气化与自动化] TP18[农业科学—农业工程]