Research Of Auto Parts Enterprises Supply Chain Inventory Optimization Based On Model Predictive Control

Under the complex and volatile market environment,"flexible and responsive" hasbecome the common aspirations of automotive parts industry while facing the customerneeds changing all the time. This requires companies in automotive parts industry supplychain to fully grasp, real-time analysis, timely treatment all kinds of information duringproduction, manufacture, procurement and sales activities. In further, a quick and flexiblemanagement optimization could be done and the organization efficient and highly qualityservice to customers of them kept. Therefore, the introduction of advanced technologiesinto practical automotive parts supply chain inventory solutions have become urgentneeds.Based on the model predictive control algorithm, the auto parts supply chainoptimization methods of operation were studied. The main three aspects as follows:(1) The existing car parts supply chain inventory management model is analyses.Fourth-order dynamic model of the supply chain, including raw material suppliers,component manufacturers, OEMs, distributors, are presented. Combined with thecharacteristics of automotive parts supply chain, the optimization objective function isdesigned for the guidance in the supply chain supply chain procurement, inventory andother key sections.(2) Combined with the characteristics of the auto parts industry supply chain, theobjective function of traditional model predictive control algorithm, by analyzing andcombining the respective advantages of the linear objective function and quadraticobjective function to optimize. Optimization algorithm implementation steps are thengiven.(3) Under uncertain demand, safety stock is improved in the established model.Target safety stock is evolved into a correction value based on changes in uncertaindisturbance from the experience constant value thus to ensure minimized inventory costsin the whole supply chain and make auto parts supply chain inventory management closerto reality.In addition, a simulation platform based on Simulink is constructed, in which theeffectiveness of the algorithm is verified. Simulation results show that MPC controlstrategy can ensure that the system can approach set goals well and inhibited the bullwhipeffect in the supply chain in a large extent. It also shows that in the case of uncertaindemand forecast error, improved MPC optimal control strategy has brought a satisfactoryreduction of cost and upgrading customer satisfaction, and a good stability robustness is manifested.