Multi-source information classification optimization based spare parts demand prediction model

• 1、State Key Laboratory for Manufacturing Systems Engineering, Xi’an Jiaotong University, Xi’an 710049, China
• 2、Xi’an Shaangu Power Co., Ltd., Xi’an 710611, China

Abstract：In order to solve the difficult demand prediction problem of main key spare parts in large power equipment manufacturing supply enterprises, the multi-source heterogeneous information from multisectoral departments was trimmed, classified and analyzed，and a spare parts demand prediction model based on multi-source information classification optimization was proposed. This model mainly included the establishment of the basic spare parts inventory, the model optimization of customer satisfaction rate, spare parts reserve strategy and the product service status. The spare parts results from hierarchical optimization prediction model, combined with time series forecasting method and enterprise actual forecasting methods respectively were analyzed by an actural example. Model actual satisfied rate is improved from 90.32% and 98.81% respectively to 98.87%. Meanwhile, practical feasibility and economical efficiency were verified for large equipment main key spare parts demand prediction.

Keywords： large power equipment multi-source information spare parts hierarchical optimization and prediction