Agricultural Sciences: Techniques and Innovations Vol. 11

Artificial intelligence has moved from a peripheral tool to a central organising technology across the crop production cycle, reshaping how water, nutrients, yield information, and post-harvest logistics are managed. This review synthesises recent scholarship on four interlocking domains of artificial-intelligence-enabled precision agriculture: predictive irrigation scheduling, remote sensing and deep-learning-based yield monitoring, variable-rate and sensor-guided fertilization, and blockchain-artificial-intelligence convergence in agri-food supply chains. Machine learning and deep learning models, including random forests, support vector machines, convolutional neural networks, and deep reinforcement learning agents, have demonstrably improved the accuracy of evapotranspiration estimation, soil moisture prediction, and irrigation timing, while unmanned aerial vehicles and satellite-based multispectral imagery paired with convolutional and transformer architectures have advanced early yield forecasting and pest and disease detection. Variable-rate technologies coupled with artificial-intelligence-driven decision-support systems have improved nutrient-use efficiency and reduced environmental externalities associated with uniform fertilizer application, while blockchain ledgers integrated with machine learning forecasting tools have strengthened traceability, food safety assurance, and demand-driven waste reduction across agri-food value chains. Despite these advances, adoption remains markedly uneven: infrastructural, financial, and digital-literacy barriers concentrate benefits among large, well-resourced producers, particularly in high-income regions, while smallholder farmers in low- and middle-income countries risk further marginalisation. The review also identifies persistent methodological limitations, including narrow geographic representativeness of training datasets, weak external validation, and underdeveloped frameworks for data governance and algorithmic accountability. It concludes that realising the productivity, sustainability, and equity potential of artificial intelligence in agriculture depends less on further algorithmic refinement than on coordinated investment in rural digital infrastructure, interoperable data standards, and inclusive policy design.