Demand Forecast Of The Automobile Parts For Auto A 4S Shop

In recent years,the earnings outlook is not optimistic about the new car sales and the new brand is worried by publics.The automotive aftermarket business gradually developed and became an important core of automotive 4S shop.In this paper,a single new brand automobile 4S shop is made to be a starting point.It is found that a series of problems is caused by inaccurate forecast of spare parts demand in 4S shop.The purpose of this paper is to select the aftermarket spare parts classification method and demand forecast model for this car 4S shop,solve the problem of component forecasting and provide data and model support for the component management of the 4S store.Through field research,I found that the 4S store customer service parts is improper about the inventory management of spare parts procurement.Empiricism is serious.The classification method is not reasonable and the demand forecasts is inaccurate.They often stored a lot of unnecessary parts and buy some unnecessary parts.The phenomenon about out of stock of some parts occur ususlly.In a long term,it can lead to a vicious inventory stagnation,reduce customer satisfaction,reduce the 4S store profitability and affect the development of the 4S shop.Considering the randomness of automobile parts and the fluctuation of demand,the fluctuation of demand is the key to inventory decision,therefore,this article hopes to optimize the scientific classification of the after-sales parts and using demand forecasting to improve the inventory management status of A 4S shop automotive spare parts.Based on the relevant theoretical research,the study of demand forecast is divided into two stages.The first stage is using data envelopment analysis method combined with clustering analysis method to clssify the 4S store parts which can find out the key spare parts that have predictive significance and make the classification more scientific and reasonable.Compared with the original classification method,the validity and rationality of the classification method is proved.The second stage is using the least squares support vector machine to predict the demand of the 4S store parts and using particle swarm optimization(pso)to determine the parameters which can raise the forecasting accuracy about A 4S shop and provide more scientific data support for the after-sale spare parts inventory management,reducing inventory and logistics costs,reducing the lag and the backlog of inventory,improving the service quality and customer satisfaction.Compared with the BP neural network model and the multivariate regression analysis,and analyzed the error,the relative accuracy of the least squares support vector machine is verified to predict the demand of car parts.The results show that the method can be used for the prediction of parts demand in 4S stores and then improve the status of parts inventory management.This paper includes 16 figures,13 tables and 54 references.