产品需求预测演化神经网络算法MLPES.pdfVIP

第14 卷 第3 期 运 筹 与 管 理 Vol. 14, No. 3 2005 年6 月 OPERATIONS RESEARCH AND MANAGEMENT SCIENCE Jun. 2005 MLPES 王瑛 ( , 100084) : , , , , MLPES, BP , , , : ; ; ; ; ; : F2 2 :A : 100 3221(2005) Forecast of Manufacturer s Product Demand Based on MLPES Algorit hm WANG Ying ( Ts ing hua Univ ers ityH ong ta Gr oup Postdoct oral S tat ion , Beij ing 100084, Ch ina) Abstract: Product demand is the starting point and motivation in the demand- pulled supply chains. Because of its special position in the supply chain, the manufacturer becomes the decoupling point in the change of sup ply chain from demand driven to forecast driven. Accurate manufacturer- centric demand forecast can to some extent reduce the affect of the uncertainty of demand. Furthermore, on the basis of the multi layer perceptron model, a neural netw ork forecasting algorithm ( Multi Layer Perceptron based on Evolution Strategy, M LPES) combining an evolution strategy is proposed w hich improves the commonly used ones, back- propa gation ( BP) algorithms. The process of the learning algorithm and the model parameters obtained by several times of calculating are also presented. Finally, the forecasting result is analyzed and compared w ith that of a BP algorithm . Key words: enterprise management; supply chain management; demand forecasting; evolution strategy; de coupling point; multi layer perceptron ( M LP) 0 ( ) , , , , , : : 国家自然科学基金资助项目( 0028102) : 王瑛( 196 - ) , 女, 博士, 研究方向 供应链管理 6 运 筹 与 管 理 2005 年第14 卷 , : -