Towards Developing an AI Random Forest Model Approach Adopted for Sustainable Food Supply Chain under Big Data

Abstract

Big data presents a transformative solution for addressing operational challenges and emerging risks in the food industry while unlocking new opportunities. It enables the analysis and integration of complex, large-scale datasets that often suffer from poor quality and unstructured formats. Although big data is a well-established technique in supply chain management, several areas remain unexplored, particularly in the global food supply chain, which faces significant limitations such as environmental impact, resource wastage, and operational inefficiencies. Achieving sustainability requires enhancing food supply chain operations through data-driven methods. The integration of big data with artificial intelligence models, such as Random Forest, offers a more efficient and sustainable approach to optimizing resource utilization, minimizing waste, and improving overall efficiency. This study develops and implements an artificial intelligence-based Random Forest model, demonstrating its effectiveness in improving sustainability in the food supply chain. The model achieves an accuracy of 96%, outperforming traditional Linear Regression, which records 91% accuracy. Additionally, the F1-score for Random Forest is 0.89, compared to 0.84 for Linear Regression, highlighting its superior balance between precision and recall. The model also improves waste reduction by 17% and optimizes resource utilization by 22%, contributing to more efficient food supply chain operations. These findings underscore the potential of integrating big data analytics and AI-driven approaches to enhance sustainability and decision-making in global food supply chains.