AI in Agriculture: Predictive Irrigation, Crop Yield Monitoring, Smart Fertilization and Agricultural Supply Chain
José Antonio Valles Romero
*
Department of Engineering and Technology, University of the Incarnate Word, Texas, USA.
Gerardo Manuel Alonzo Medina
Department of Engineering and Technology, National Technological Institute of Mexico, Mexico.
Emilio Raymundo Morales Maldonado
Sustainable Agricultural Innovation, National Technological Institute of Mexico, Mexico.
*Author to whom correspondence should be addressed.
Abstract
Background and Aims: This article introduces an AI-driven Smart Agriculture framework that integrates artificial intelligence, IoT, GIS, remote sensing, predictive analytics and autonomous environmental monitoring systems to support irrigation management, crop productivity monitoring, smart fertilisation and farm-level agricultural supply-chain operations. It also presents the PlaSmA Agricultural Intelligence Index (PAII) as a composite KPI for assessing agricultural performance.
Study Design: A multidisciplinary Agriculture 5.0 software architecture was developed and assessed by combining computational intelligence, geospatial technologies, environmental monitoring systems and machine-learning analytics.
Place and Duration of Study: The research was conducted within the Platform for Smart Agriculture (PlaSmA), using the agricultural laboratory of the Huichapan Higher Technological Institute for the Sustainable Agricultural Innovation Engineering degree in Mexico, for smart farming applications from 2 January 2025 to 23 March 2026.
Methodology: The framework integrated IoT sensors, satellite and UAV remote sensing, GIS databases, smart soil analytics laboratories and AI-powered predictive models within a data-driven decision-support and management platform. NDVI-based remote sensing was applied for vegetation monitoring, biomass and production estimation, and drought analysis. Machine-learning and deep-learning algorithms were used for irrigation scheduling, crop-yield prediction and smart nitrogen management. A Digital Twin framework was adapted to represent the agricultural environment contextually. Results: The integrated framework supported remote crop monitoring, predictive optimisation and smart automation of agricultural processes. The PAII combined NDVI, biomass, productivity and water-stress measures into a consolidated KPI for assessing agricultural operations in Agriculture 5.0.
Conclusion: The PlaSmA framework demonstrates the convergence of AI, GIS, IoT, remote sensing, predictive analytics and autonomous systems as a scalable Agriculture 5.0 model. The PAII provides a practical metric for representing crop health, productivity, sustainability and operational performance within a single decision-support indicator.
Keywords: Smart agriculture, Agriculture 5.0, Artificial Intelligence, precision agriculture, predictive irrigation, crop yield monitoring, smart fertilisation, remote sensing, NDVI, agricultural supply chain, digital twin, PAII KPI