基于模糊时间窗的同时送取货选址路径规划模型研究

发布时间：2018-03-20 19:26

本文选题：LRP模型　切入点：同时送取货　出处：《华南理工大学》2014年硕士论文　论文类型：学位论文

【摘要】：选址-路径问题(LRP)同时解决了企业战略层的设施选址和运作层的车辆调度两种互相联系的复杂的决策问题，合理配置运力资源，令物流系统总成本最小，达到系统最优，在电子商务时代和绿色供应链环境下的集成化物流配送网络规划中具有重要意义。在实际物流配送中，顾客常同时存在送货和取货两种需求，客户弹性的预约配送时间偏好对客户满意度也存在影响。为进一步完善和优化集成物流配送系统，本文分析电子商务选址决策和运输管理存在的问题，结合物流供应链系统规划与设计的发展趋势，在选址路径规划模型中结合顾客对需求的多样性以及时间不确定性，以关注长期效益的物流成本最小和客户满意度最大为双目标要求。论文主要从以下方面对选址路径问题进行了分析和研究：运用非线性混合整数规划理论，在仓库容量限制和路径容量约束的基础上，考虑客户同时存在送货和取货需求的情况和客户可配送模糊时间范围约束的情况，建立基于模糊时间窗的同时送取货的多仓库、多车型选址路径模型。模型求解算法上，本文设计结合模拟退火算法的改进两阶段自适应遗传算法进行求解。第一阶段改进混合遗传算法分别对初始种群生成方式、遗传操作和重组策略进行改进，实现了模拟退火的良好局部搜索能力与遗传算法的全局搜索能力的有效结合。第二阶段设计模糊优化程序，以在最大可容忍时间窗内最大化客户服务水平。最后进行数值实验，，通过算例验证了模型和改进算法的可行性和有效性。仿真实验表明：两阶段改进混合遗传算法能够在短时间内给出满意解，模型有效平衡了客户服务水平和物流成本两个指标，可为实际的选址与运输决策提供重要参考依据。本研究对拓展选址路径问题理论和应用实际，改进启发式求解算法，指导决策者实践，促进选址与运输决策的科学化有一定的理论和现实意义。
[Abstract]:LRP) simultaneously solves the complex decision-making problems of facility location in strategic level and vehicle scheduling in operation layer. Rational allocation of transportation resources can minimize the total cost of logistics system and achieve the optimal system. It is of great significance in the planning of integrated logistics distribution network under the environment of electronic commerce and green supply chain. In order to perfect and optimize the integrated logistics delivery system, customers often have two kinds of demand, delivery and delivery, and customer's flexible preemptive delivery time preference also affects customer satisfaction in the actual logistics distribution. This paper analyzes the problems existing in E-commerce location decision and transportation management, combined with the development trend of logistics supply chain system planning and design, combined with the diversity of customer demand and uncertainty of time in the location path planning model. In order to focus on the long-term benefits of the minimum logistics costs and maximum customer satisfaction as a two-objective requirement. This paper mainly from the following aspects of the location path problem analysis and research:. Based on the constraints of storage capacity and path capacity, the nonlinear mixed integer programming theory is used to consider the situation that the customer has both the demand for delivery and the delivery of goods and the fuzzy time range constraint for customer distribution. A multi-warehouse and multi-vehicle location path model based on fuzzy time window is established. In the model solving algorithm, this paper designs an improved two-stage adaptive genetic algorithm combined with simulated annealing algorithm to solve the problem. In the first stage, the improved hybrid genetic algorithm improves the initial population generation method, genetic operation and recombination strategy, respectively. The good local search ability of simulated annealing is effectively combined with the global search ability of genetic algorithm. In the second stage, a fuzzy optimization program is designed to maximize the level of customer service in the maximum tolerable time window. Finally, numerical experiments are carried out to verify the feasibility and effectiveness of the model and the improved algorithm. The simulation results show that the two-stage improved hybrid genetic algorithm can give a satisfactory solution in a short time. The model effectively balances the customer service level and logistics cost, and can provide an important reference for the actual location selection and transportation decision. This study has certain theoretical and practical significance for expanding the theory and application of location routing problem, improving heuristic algorithm, guiding the practice of decision makers, and promoting the scientific decision of location and transportation.
【学位授予单位】：华南理工大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP301.6;F252

【参考文献】