基于多目标规划的原材料订购与运输决策研究

Research on Raw Material Ordering and Transportation Decision-Making Based on Multi-Objective Planning Model

文章针对生产企业的原材料订购与运输决策问题,以减少转运及仓储成本(即尽量多地采购A类和尽量少地采购C类原材料)和保证转运损耗最低为目标函数,以企业接收的原材料满足每周的生产需求、每家转运商的运输能力有限和每家供应商的原材料尽量由一家转运商转运等为约束条件,建立多目标规划模型,采用粒子群方法求解,并分析了模型的效果。在建立模型之前需要进行两个准备,一是企业在计划未来的订购和转运方案时,总是会优先考虑对自己重要的供应商,因此需要对已知的供应商进行评价和选择,文章采用熵权topsis方法,对供应商的选择更为客观,二是采用LSTM对未来24周的订货量进行预测,准确率为75.96%,效果较好。该模型具有一定的合理性和普适性。

This paper addresses the raw material ordering and transportation decision problem of a manufacturing enterprise,and establishes a multi-objective planning model with the objective function of reducing the cost of transshipment and storage(purchasing as much as possible of Class A and as little as possible of Class C raw materials)and ensuring the lowest loss of transshipment,with the constraints that the raw materials received by the enterprise meet the weekly production demand,the limited transportation capacity of each transshipment provider,and the raw materials from each supplier are transshipped by one transshipment provider as much as possible.The objective programming model is solved by the particle swarm method,and the effect of the model is analyzed.Two preparations are needed before building the model.First,when planning future ordering and forwarding solutions,enterprises always give priority to suppliers that are important to them,so they need to evaluate and select known suppliers,and this paper uses the entropy weight topsis method,which is more objective for the selection of suppliers.Second,the LSTM is used to predict the order quantity for the next 24weeks,and the accuracy rate is 75.96%,and the effect is better.The model has certain rationality and universality.