INTELIGÊNCIA ARTIFICIAL E MACHINE LEARNING NA OTIMIZAÇÃO DA GESTÃO DA AGRICULTURA 4.0: UMA REVISÃO SISTEMÁTICA DA LITERATURA

Abstract

The Brazilian agribusiness sector, responsible for 23.2% of the national GDP in 2024, faces the challenge of increasing food production sustainably in light of population growth and global demand, with UN projections indicating a 70% rise in production needs by 2050. In this context, Agriculture 4.0, supported by digital technologies such as Artificial Intelligence (AI) and Machine Learning (ML), emerges as a strategy to optimize agricultural management by integrating physical and digital environments, monitoring crops, climate, and soil in real time, and supporting strategic decisions. The general objective of this research is to evaluate how AI and ML tools transform management processes in Agriculture 4.0, creating value in operational efficiency, strategic decision-making, and sustainability. This study, characterized as qualitative, exploratory, and bibliographic, was based on a systematic literature review (SLR). The guiding question was: How do AI and ML solutions transform management processes in Agriculture 4.0, creating value in operational efficiency, strategic decisionmaking, and sustainability? To address this question, the following search string was defined: (“Artificial Intelligence (AI)” OR “Machine Learning (ML)”) AND (“Agriculture 4.0” OR “Smart Agriculture”), applied to the Scopus and Web of Science (WoS) databases in July 2025, restricted to “articles” and “review articles,” in English, open access, and published between 2020–2025. After applying filters and inclusion and exclusion criteria, from 649 records in Scopus and 260 in WoS, a total of 13 studies were considered sufficient for the SLR. The results indicate that AI and ML, integrated with related digital technologies, enable real-time monitoring and control of variables such as climate, soil, crops, and livestock, as well as yield prediction, consultancy support, automation of irrigation, smart spraying, and harvesting, optimization of resource use, and improved traceability. Common benefits identified in the literature include increased productivity, reduced operational costs, efficient use of inputs, minimization of environmental impact, improved product quality, and support for data-driven strategic decision-making. However, substantial challenges remain, including high initial costs, the need for adequate digital infrastructure, shortage of skilled labor, lack of field connectivity, technological integration difficulties, and resistance to adoption. The findings reinforce that AI and ML, when applied in an integrated and strategic manner, hold the potential to redefine agriculture and promote more efficient, sustainable, and competitive management, provided they are accompanied by investments in training, infrastructure, and data governance.