关键词: 交货延迟, 自适应控制, APIOBPCS, 动态库存优化

Abstract: Considering the problems of inventory overstock, stockout and bullwhip effect caused by random demand fluctuation during the time of delay, a multi-stage inventory dynamic optimization model based on model reference adaptive control（MRAC） is established. Firstly, based on APIOBPCS theory, the multi-stage inventory dynamic model is established by applying Taylor expansion and Laplace transformation. Secondly, based on the Lyapunov asymptotic stability theory, a model reference adaptive controller suitable for multi-stage inventory control is designed. In this MRAC system, a reference inventory model with smaller time of delay is set as the target. By adjusting the linear compensator function, the output error between reference inventory model and actual inventory model is narrowed, and the inventory system achieves dynamic gradual stability. Finally, through numerical simulation, the dynamic optimization of MRAC on a three-stage supply chain inventory system is tested. The simulation results show that the average residual inventory of the multi-stage inventory systems without information sharing under adaptive control decreases slightly, the stockout has completely improved, and the bullwhip effect decreases by 40.7%. Under the premise of not increasing the operation cost, this paper optimizes the resource allocation through the adaptive control algorithm, and most importantly, dynamically weakens the impacts of time of delay on multi-stage inventory and improves the operation efficiency of the whole supply chain.

Key words: delivery delay, adaptive control, APIOBPCS, dynamic inventory optimization