Design And Implementation Of Production Delivery Capability Improvement System With Customer's Demand Forecasting

The automobile industry has become one of the pillar industries of new China since the Economic Reform and open up.How to improve production efficiency has become an urgent problem for the automobile industry to solve.To produce the orders in advance,NFFaurecia needs to predict the number of customers' orders in the next few days.This thesis does research on the order-forecasting algorithm for the time series of the number of customer's order.The method used in this thesis is a combination of the forecasting algorithm for time series based on the most similar polyline segment and the ARIMA(p,d,q)model.This thesis firstly used the forecasting algorithm for time series based on the most similar polyline segment to forecast the number of customer's order.Then this thesis used the ARIMA(p,d,q)model in the statistics to forecast the number of customer's order.Finally,this thesis merged two forecasting algorithm for a better performance.NFFaurecia had no system for real-time display of production process,the number of inventory and delivery process,this thesis developed the E-Leveling system to improve the production delivery capability of NFFaurecia.This thesis finished a comprehensive and complete system testing process for the E-Leveling system to ensure the reliable working of the E-Leveling system.Compared with the previous production delivery process,the E-Leveling system with delivery capability enhancement is a comprehensive upgrade.Because most tasks of NFFaurecia are driven by the scanning actions of workers on the assembly line,how to simulate the generation process of time series data in integration testing process is a problem which needs to be solved in this thesis.Focusing on how to separate the test code from the test data,how to generate the simulation data one by one according to the test case,and how to make these more open,this thesis proposed an open time series-testing tool.With simple configuration,it can be used effectively for any similar integration-testing task for time series data-driven business.Combined with the QAD production traceability system of NFFaurecia,the E-Leveling system greatly improve the production delivery efficiency of NFFaurecia.