USING PREDICTIVE STATISTICAL MODELS TO FORECAST DEMAND AND IMPROVE INVENTORY MANAGEMENT ACROSS SUPPLY CHAINS

This study examines the application of predictive statistical models in forecasting demand and optimizing inventory management across supply chains from 2020 to 2024. The research aimed to evaluate model accuracy, identify key inventory optimization factors, and propose best practices for integrating predictive analytics into supply chain operations. A mixed-methods approach was employed, utilizing quantitative analysis of industry reports and time-series data alongside qualitative insights from supply chain professionals. The results indicate that hybrid models outperformed traditional forecasting methods, achieving a 95% accuracy rate by 2024. Regression analysis showed a strong relationship (R² = 0.92) between inventory optimization and cost savings, with businesses experiencing a 25% reduction in holding costs and a 30% increase in inventory turnover efficiency. Correlation analysis confirmed a robust relationship (r = 0.85) between predictive accuracy and inventory turnover. The study concludes that predictive analytics significantly enhance supply chain efficiency, reduce stockouts, and improve financial performance. However, challenges such as data integration and scalability persist. It is recommended that businesses invest in hybrid predictive models, real-time data integration, and workforce training to maximize the benefits of predictive analytics in inventory management.