Transparent Crop Price Forecasting Framework Using Explainable AI for Multi-Factor Driver Analysis and Stakeholder Decision Support

Reliable open crop price forecasts are needed for the future of agriculture, food availability, and making informed decisions by the stakeholders. The uncertain nature of the markets of farmers and traders, the fluctuation of their incomes, are typical results of the difficult work of prediction, which is caused by the interacting effect of meteorological, economic, and policy factors. Problem Statement: The conventional approaches to market movement forecasting are not always effective due to their incoherent nature and lack of transparency to the stakeholders. This renders it hard to ascertain the nature of the variables that are driving price changes. A predictability and responsiveness system that is explainable and data-driven is becoming more and more important, and this paper introduces the Explainable AI-based Crop Price Transparency Framework to Factor Analysis and Resilient Management (EXACT-FARM), one of the Crop Price Transparency Frameworks that is driven by Explainable AI. EXACT-FARM is a hybrid modelling (SARIMAX-XGBoost-LSTM) strategy coupled with feature-level explanations based on SHAP values and counterfactual reasoning. The model takes into account a number of factors in order to come up with forecasts and driver attributions that are easy to comprehend. These are the production trends, events in the policy, weather indices, and trade dynamics. The use of multiple-year datasets in experiments indicates that EXACT-FARM increases predictive accuracy by 1520 percent, and visually explains the influence of drivers. EXACT-FARM is an open and trusted decision-support system that gives stakeholders actionable pricing, planning, and sustainable agriculture management information.