Supply Chain Management Optimization and Cost Reduction Based on Data Mining Algorithm

Abstract

Supply Chain Management (SCM) is an evolving industry that constantly evolves, but one constant is the significance of effective inventory control in optimizing profits and customer satisfaction. The primary objective of the research is to identify the most effective method for a business to supply several products while maintaining customer service and keeping inventory costs low. The main aim of this paper's Stochastic Gradient Descent (SGD)-enhanced Support Vector Machine (SVM) approach is to predict ideal order and lack quantities, considering into account historical demand, level of inventory, vendor performance, and cost data. The key goal is to reduce the total cost of inventory (which includes holding, placing orders, and supply-chain costs) while maintaining excellent customer service. In order to accomplish this, the proposed system combines SGD's efficiency to enhance predictions and SVM's predictability. A consumer products business chain's real-world dataset was employed to train and test the algorithm. The dataset provides three years' worth of revenue, stock levels, vendors, and costs. Parameters like Mean Absolute Error (MAE), Inventory Turnover Ratio (ITR), and Service Level Achievement (SLA) have been employed to test the methodology, and parameters were optimized using a 5-fold cross-validation. Based on the study findings, the simulation model obtains an average MAE of 1.81 following 100 training epochs, substantially boosting ITR and SLA. As it effectively and accurately balances inventory levels with customer needs, the findings demonstrate that the SVM model with SGD is an excellent match for IMO.