Research And Implementation Of Inventory Optimization System For Automotive Parts Based On Demand Forecasting

Demand forecasting plays a crucial role in enterprise management processes and serves as the foundation for planning and decision-making in production and business activities.Accurate forecasting of product demand can assist companies in adjusting production and procurement plans to better meet market demands.Inventory management is a critical link in supply chain management,and effective inventory management can help companies reduce costs,improve service quality,and enhance competitiveness.Establishing accurate demand forecasting models to guide inventory management,has a positive impact on the development of enterprise operations.This thesis focuses on demand forecasting and inventory optimization of automotive parts.After investigating the current status of inventory management among suppliers of a certain car sales company,it was found that most automotive parts sales companies have the following shortcomings:(1)Single inventory part classification,with a rough classification,making it difficult to manage in a refined manner.(2)Most companies still rely on manual and traditional methods for demand forecasting and have not established scientific and effective demand forecasting methods.Inaccurate forecasting of parts demand affects the company’s production and procurement planning.(3)Inventory management is crude,without an effective inventory optimization plan,resulting in out-of-stock situations and stagnant inventory,greatly affecting the company’s profitability and market competitiveness.Based on the above shortcomings,this thesis collected relevant data on the sales and inventory of automotive parts suppliers from the car sales company,combined with local automotive production and parts factory price index,constructed an automotive parts demand forecasting model,proposed a stock optimization plan based on a combination model,and implemented a warehousing collaborative management and inventory optimization system.The main work of this thesis includes:(1)Based on automotive parts sales data,the inventory parts were categorized according to their importance.Using the ABC classification method and K-shape clustering algorithm to optimize the parts classification and further divide the original Aclass,and B-class parts based on their sales trends,making it easier for companies to manage in a refined manner.Finally,the classified key parts were selected as the research objects for demand forecasting.(2)Using data warehouse technology(Extract-Transform-Load,ETL)to collect historical sales and inventory data for automotive parts and conducting pre-processing on the data.The exploratory analysis was conducted to obtain factors that influence parts demand and extract key features from them.Multiple single models were constructed to predict parts demand,and the models with better predictive effects were screened to establish a combination model.Experiments have proved that the combined model based on Light GBM and PSO-LSTM is at least 4.9 percentage points and 2.6 percentage points lower than the single prediction model root mean square error(RMSE)and mean absolute error(MAE),which effectively improves the accuracy of demand forecasting and reliability.(3)Based on the demand forecasting results and inventory optimization theory,the safety stock of the parts was calculated,and the inventory alert line was set.By comparing it with the current inventory,a reasonable optimization plan was obtained,which achieved the goal of optimizing enterprise inventory.(4)Based on the demand forecasting model and inventory optimization plan,using popular B/S architecture and Web development framework,the inventory optimization module was added to the inventory management system to assist enterprise inventory management personnel in inventory optimization management.This thesis has constructed an accurate demand forecasting model,proposed a reasonable inventory optimization plan,and implemented an inventory optimization system based on it,which can help companies reduce inventory costs,improve service quality,and enhance competitiveness.It has certain application and promotion value.