时变需求下多周期车辆服务设施选址多目标优化

Multi-objective Multi-stage Automotive Service Facility Location Optimization under Time-dependent Demand

服务设施选址对企业运营管理至关重要,尤其是在物流领域。为确保物流运输车辆的正常运作,及时的车辆检测非常关键。影响设施选址决策的因素如需求、成本等通常是动态变化的。因此,考虑时变需求的多周期设施选址更加符合实际。针对时变需求下多周期车辆服务设施选址优化这一类新问题,构建了同时最大化投资回报与客户满意度的双目标混合整数规划模型,采用了基于模型的ε-约束法获得小规模问题的Pareto前沿,并结合问题结构特性,设计了两种迭代启发式算法快速获得大规模问题的高质量近似Pareto前沿。典型实例和大量随机算例数值实验验证了模型的正确性和算法的有效性,并为车辆服务企业提供了多周期设施选址的管理启示。

The location of service facilities is very important to the operation and management of enterprises,especially in the field of logistics.To ensure the operation of logistics transportation vehicles,timely vehicle detection is very important.Factors that affect the decision of facility location,such as demand and cost,usually change dynamically.Therefore,it is more practical to consider multistage facility location considering time-dependent demands.For a new time-dependent multi-stage automotive service facility location problem,a bi-objective mixed-integer linear programming model to simultaneously maximize investment return and customer satisfaction was developed.A MILP-basedε-constraint method was first proposed to obtain the Pareto frontiers for small-sized instances,and then two iterative heuristic algorithms were devised to quickly obtain the high-quality approximate Pareto frontiers for large-sized problems based on the structural properties of the problem.Numerical experimental results for a real case and extensive randomly generated instances verify the correctness of the model and the effectiveness of the algorithm and provide management insights for automotive enterprises'decision-makers in optimally locating service facilities under time-dependent demand.