三维装载约束下基于运输资源共享的车辆路径问题

Vehicle routing problem based on transportation resource sharing under three-dimensional loading constraints

针对车辆路径问题研究在三维装载方式和运输资源共享模式结合方面存在的不足,提出三维装载约束下基于运输资源共享的车辆路径优化策略。首先,结合客户点地理位置特征和服务时间窗属性确定多个服务周期,建立了多个服务周期内物流运营成本最小化和车辆使用数最小化的双目标优化模型。其次,设计了集成k-means时空聚类的Clarke-Wright—非支配排序遗传算法求解模型,该算法引入Clarke-Wright节约算法以提高初始解的质量,并结合非支配排序遗传算法提高了混合算法寻找优化解的全局和局部空间搜索能力。最后,结合实例数据对所提方法进行了计算验证,给出了三维装载约束下基于运输资源共享的车辆路径优化方案并探讨了不同车厢空间分区模式下物流运营总成本、车辆使用数、车辆平均装载率和车辆平均使用频次的变化情况。研究表明:根据客户需求货物种类数和货物规格划分配送车辆装载空间,可有效降低物流运营总成本,减少配送车辆使用数,增加车辆共享频次和有效提高车辆平均装载率,并可为基于运输资源共享的三维装载物流网络优化问题提供决策参考和方法支持。

To overcome the shortcomings of the vehicle routing problems incorporated with three-dimensional loading and transportation resource sharing,a vehicle routing optimization strategy with transportation resource sharing was proposed based on the three-dimensional loading constraint.The multiple service periods were determined based on the customers'geographical location and service time windows,and a bi-objective optimization model was established to achieve the minimization of the total operating cost and the number of vehicles in multiple service periods.A Clarke-Wright—fast elitist Non-dominated Sorting Genetic Algorithm(CW-NSGA-II)algorithm combined with three-dimension k-means clustering was designed to find the optimal solution.With the introduction of the Clarke Wright(CW)algorithm,the proposed hybrid algorithm could improve the quality of the initial generated solution and enhance the capacities of global and local searching.A real-world case was employed to verify the proposed method,and the vehicle routing optimization scheme based on the transportation resource sharing with the three-dimensional loading constraint was presented.The comparisons of the total logistics operating cost,the number of vehicles,the average vehicle loading rate and the average vehicle using frequency under different compartment space partition modes were analyzed.The results showed that dividing the compartment space of vehicles according to the number of types and specifications of goods required by customers could effectively reduce the total logistics operating cost,reduce the number of distribution vehicles,increase the frequency of vehicle sharing and effectively improve the average loading rate of vehicles.This study provided some related reference and method support for optimizing three-dimensional loading and logistics networks based on transportation resource sharing.