Fashion Products Demand Forecasting And Inventory Control Research Under Non-stationary Demand

With the progress of science and technology and the development of economic globalization,the enterprises are facing more and more uncertain factors,the supply chain emergency more and more specific performance fluctuations in demand lead time is longer,the product is increasing,which caused by the supply and demand aspects.It is very important for the enterprise to make the production and inventory decision,and how to forecast the product demand more accurately.This thesis regards the apparel industry nonstationary demand as the background,combined with the product life cycle theory to predict the demand of apparel products;from theory and practice,to study the non-stationary stochastic demand inventory control strategy.The main research contents of this thesis include:Firstly,the thesis introduces the nonstationary demand inventory problem research background and significance,summarizes the current research achievements about this problem and the problems,finishing up the present research situation of domestic and foreign scholars demand forecasting,put forward the research goal of the thesis.Secondly,according to the survey results of the visited enterprise the apparel industry sales features,combined with the theory of product life cycle in different periods of product sales changes,proposed two piecewise prediction models of rolling forecast,through a set of enterprise actual production data for empirical analysis,compared to get closer to the actual production prediction model.Finally,according to the forecast results,the inventory management model is constructed in the case of not zero lead time,variable purchase cost and no loss of sales,and the method is used to solve the problem of static and dynamic uncertainty.Through the simulation,the static and dynamic uncertain method can be compared with the static method,which can significantly reduce the cost.