Smart Agriculture Economics and Engineering: Unveiling the Innovation Behind AI-Enhanced Rice Farming

Abstract

Food security challenges in Southeast Asia, across all income brackets, have been growing, according to the Food and Agriculture Organization (FAO) of the United Nations. This article introduced innovative Artificial Intelligence-based (AI-based) predictive algorithms for short-term rice production, utilizing the Cross Industry Standard Process for Data Mining (CRISP-DM) data science framework. The predictive algorithms integrated features addressing three food security dimensions: availability, accessibility, and stability, and identified key determinants in three clusters: atmospheric, socioeconomic, and farming practices. By employing the proposed innovative modified stacked Multiple Linear Regression-Support Vector Regression-based (MLR-SVR-based) algorithms, and ranking them utilizing the modified Taguchi-based VIseKriterijumska Optimizacija I Kompromisno Resenje (Taguchi-based VIKOR) multi-criteria decision-making algorithm, the analysis demonstrated high predictive accuracy even with limited data. The proposed AI-based predictive algorithm was utilized to forecast 5-year future rice production for Southeast Asia nations, yielding generally accurate results except for Cambodia (KHM). This research has significant implications for agriculture, food production, data analytics, and policymaking, potentially enhancing efficiency and innovation in agricultural operations.