不确定环境下车辆配送路径鲁棒优化及求解算法

Robust Distribution Vehicle Routing Under Uncertain Environment

为了提高配送方案在需求信息或运输时间等不确定因素相关数据无法完全获知时的可靠性,以最小化运输成本为基本目标,在保证较低物资配送延迟率的前提下,提出了基于场景分布式鲁棒优化的车辆配送路径规划模型。在分析传统随机优化模型的基础上,构建了多种不确定集合下的确定型鲁棒对等式模型。同时,为了提高该模型在大规模问题下的可行性,提出了基于免疫遗传算法的求解算法。最后,论文以某零售商品配送路径优化问题为例对模型及算法进行验证,结果表明该模型比一般模型降低配送延误28.6%,因此,鲁棒优化模型在提高客户满意度方面比传统风险中性模型具有明显优势。

In order to improve the reliability of the distribution scheme when the data of demand or transportation time cannot be fully known,this paper develops a robust vehicle routing planning model,to minimize transportation cost and ensure the delay rate.Based on the analysis of traditional stochastic optimization models,this paper presents an equivalent deterministic robust model under multiple uncertain sets.At the same time,in order to improve the feasibility of the model under large-scale problems,an algorithm based on immune genetic algorithm is proposed.Finally,the paper validates the model and algorithm by taking a retail commodity distribution routing problem as an example.The results show that the model reduces the delivery delay by 28.6%compared with the general model.Therefore,the robust routing model has obvious advantages than traditional models in improving customer satisfaction.