Integrated pest management (IPM) in the AI era

Le résumé fourni par la source

Pest infestations pose a formidable threat to global agriculture, causing significant yield loss and environmental challenges. Conventional Integrated Pest Management (IPM) frameworks often struggle with timely detection, accurate monitoring, continuous field surveillance, and efficient decision-making. Artificial Intelligence (AI) offers a transformative framework that unifies Machine Learning (ML), Deep Learning (DL), the Internet of Things (IoT), remote sensing, robotics, and Decision Support Systems (DSS) to modernize pest management. This review highlights how AI-driven applications are reshaping IPM by enabling automated pest detection, dynamic population forecasting, spatial risk mapping, precision interventions, and post-management evaluation. Integrated robotics and automation facilitate targeted pesticide application and autonomous field monitoring, and unmanned aerial vehicles (UAVs) support large-scale and real-time surveillance. AI-enabled DSS, coupled with mobile and web platforms, translates complex analytics into actionable recommendations, promoting farmer adoption through accessibility, adaptability, and timely advisories. The synergistic integration of these technologies reduces chemical reliance, optimizes resource utilization, and enhances operational efficiency, thereby advancing sustainable pest management practices. Such advancements would enhance predictive accuracy by coupling AI with ecological and climatic models, leveraging participatory and citizen science platforms for real-time data acquisition, employing generative AI for knowledge dissemination, and developing robust policies and educational frameworks to accelerate adoption. By addressing these avenues, AI-driven IPM can evolve into an intelligent, adaptive, and sustainable paradigm, fostering next-generation precision agriculture while safeguarding both productivity and environmental resilience.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.