Machine learning–enabled decision support for planning and control in sustainable agricultural supply chains
本文提出一个概念性决策支持框架，将机器学习嵌入农业供应链的规划、协调与控制过程，以提升产量规划、资源分配、物流协调、质量控制和废物减少，并关联环境效率、经济韧性和透明度等可持续性成果。
The development of sustainable agricultural supply chains is critical for improving operational efficiency, coordination, and resilience under increasing demand uncertainty, climate variability, and resource constraints. These supply chains involve complex planning and control decisions across interconnected stages, including pre-production, production, post-harvest processing, and distribution. However, existing research on artificial intelligence (AI) and machine learning (ML) remains fragmented and lacks an integrated planning and control perspective. This study addresses this gap by developing a conceptual decision-support framework that positions ML as a decision-intelligence layer embedded within planning, coordination, and control processes. Using a structured literature synthesis, the study maps ML techniques to key operational decision contexts across the supply chain. The framework demonstrates how ML-enabled decision support can enhance yield planning, resource allocation, logistics coordination, quality control, and waste reduction, while linking these decisions to sustainability outcomes such as environmental efficiency, economic resilience, and transparency. The study also identifies key implementation challenges, including data availability, interoperability, infrastructure constraints, and organizational readiness. The study contributes a decision-centric framework that integrates ML with supply chain planning and control, providing actionable insights for improving performance and sustainability in agricultural supply chains.