A Nature Inspired Optimisation Method for Supply Chain Management Problem

Supply chain management has emerged as one of the most important fields to the industry. It demands reliable models for prediction and highly accurate optimisation techniques. It has been shown that mathematical models based on artificial intelligence consistently make sufficiently accurate predictions. In this paper we improve upon a novel method of using Support Vector Regression (SVR) along with Particle Swarm Optimisation (PSO) and hyperparameter tuning. Regularisation constant (RC) was set to its optimal value to reduce error and more accurate predictions. We used the retail sales data from the United States Census Bureau based on the monthly sales report for retail trade and food services report (2002) to test the accuracy for different values of regularization coefficient. We achieved successfully lower error rates in forecasts which results in better sales predictions.