基于滑动平均订单预测的半导体生产库存控制模型.docVIP

基于滑动平均订单预测的半导体生产库存控制模型 摘要：本文针对半导体生产企业的生产时间不确定性和订单到达不确定性，分别使用滑动平均法与指数平滑法建立客户订单预测模型，提出了基于这两种客户需求预测模型的离散反馈状态生产库存控制系统模型；讨论了这两种模型系统有界输入有界输出的稳定性，并给出了基于这两种模型的系统达到有界输入有界输出稳定的参数限制条件。进一步在满足系统稳定的条件下，使用某半导体制造企业的订单生产数据，对两种预测模型进行仿真运算，对不同模型下系统生产库存控制的具体效果进行了比较，结果显示该模型能使半导体企业生产库存控制获得小于3%的牛鞭效应，优于指数平滑法预测模型的结果。 关键词：生产控制系统，预报方法，稳定性，半导体制造 A Production and Inventory Control Model in the Semiconductor Manufacturing Processes Base on Moving Average Demand Forecasting Method Abstract：Covering the customs’ requirements and controlling the work in progress are the major challenges of applying simulation to short-term operation scheduling of semiconductor manufacturing factories, which are characterized by re-entrant process flows, stringent production control requirements and fast changing technology and business environments. In this paper, targeting the uncertainties of the semiconductor manufacturing process and the demands, a control model is proposed that is a discrete time case and derive state space feedback model that based on the moving average methods to predict the customs requirements to be representations of production and inventory control policies for semiconductor manufacturing, which estimates the production lead time and the uncertainty of customs’ demands. Then the stability of the two models that are MA(moving average) and ES(exponential smoothing) forecast methods has been analyzed, and the general stability criterion of the order policy parameters for a system, which is BIBO (Bound Input, Bound Output), with any lead time is given by the paper. Finally, we demonstrate the effectiveness of the proposed model by simulating a real world example taken from a wafer fabrication shop floor in an IC manufacturing factory. The result shows that the MA forecast method of produce control model would achieve less than 3% bullwhip effect in this semiconductor manufacturing factory. Key words: Produce Control System, Forecasting Methods, Stable, Semiconductor Manufacturing 0 引言 半导体生产