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. The research aims to evaluate the effectiveness of these models in improving accuracy, reducing costs, and mitigating supply chain inefficiencies. A systematic review and meta-analysis approach was employed, analyzing data from 2020 to 2024 through regression analysis, correlation analysis, and chi-square tests. The findings indicate a strong positive correlation (r = 0.85, p < 0.01) between forecasted demand and actual sales, with a regression model achieving an R² value of 0.92, demonstrating that 92% of future demand variability can be predicted based on historical data and economic indicators. The chi-square test (χ² = 18.34, p = 0.002) confirmed that predictive models significantly enhance inventory accuracy. These improvements led to a 10% annual sales growth, a reduction in stockouts from 50 in 2020 to 10 in 2024, and over $10M in cost savings. Despite challenges such as data integration and model adaptability, hybrid forecasting techniques and blockchain integration emerged as best practices to enhance performance. This study recommends investing in centralized data systems, leveraging AI-driven models, and fostering a data-driven decision-making culture to maximize the benefits of predictive analytics in supply chain management.