Time Series Forecasting and Modelling of Food Demand Supply Chain Based on Regressors Analysis

Abstract

Efficient management of food demand and supply chains is critical for ensuring food security, minimizing waste, and optimizing resource utilization. This study presents a comprehensive approach to time series forecasting and modelling of food demand within supply chains using regressor-based analysis. The proposed framework integrates historical demand data with multiple influencing factors such as seasonal variations, price fluctuations, population growth, and climatic conditions. Advanced statistical and machine learning techniques, including regression models and time series methods (e.g., ARIMA, SARIMA, and hybrid models), are employed to capture both temporal patterns and external impacts on demand. The analysis highlights the significance of incorporating exogenous regressors to improve forecasting accuracy compared to traditional univariate models. The model is evaluated using standard performance metrics such as Mean Absolute Error (MAE) and Root Mean Square Error (RMSE), demonstrating enhanced predictive capability. The results support better decision-making in inventory management, procurement planning, and distribution logistics. This research contributes to developing a robust, data-driven framework for sustainable and responsive food supply chain management.