The effects of artificial intelligence-based demand forecasting and inventory management on supply chain performance

This study aims to examine the impact of artificial neural network (ANN)-based demand forecasting and inventory management on business performance in supply chain management. Sales data from a business for the years 2021-2024 were utilized to develop demand forecasting models. ANN models were applied to product groups selected through ABC analysis, and the accuracy of the forecasts was evaluated based on historical data. During the application process, the dataset was preprocessed, and ANN models were employed to generate forecasts for different demand levels. The results indicated that the models have the potential to optimize inventory costs and improve customer satisfaction. Moreover, the effects of exchange rate fluctuations and customer behavior variability on model performance were analyzed. It was observed that incorporating these variables into the models positively influenced the forecasting results. In this study, the accuracy performance of demand forecasting using Artifical Neural Networks (ANN) was evaluated. The results indicate that ANN models achieve higher accuracy with lower error rates (MAPE, PAE) compared to clasical methods. Based on these findings, it has been demonstrated that the integration of the artificial intelligence enhances the efficiensy of supply chain processes and business operations while improving decision-making processes.