Key points are not available for this paper at this time.
Tomato (Solanum lycopersicum L.) production faces increasing pressure from resource scarcity and climate change, creating demand for more precise and adaptive management. However, adoption in commercial systems remains limited because many advanced technologies are costly, poorly interoperable, or difficult for growers to interpret. This review addresses that gap by organizing recent advances into a five-stage production ecosystem framework: Measurement, Understanding, Optimization, Prediction, and Diagnosis. Unlike previous precision agriculture reviews that mainly summarize sensing, modeling, artificial intelligence, and robotics as separate topics, this framework emphasizes stage-linked integration and decision support relevance across practical tomato production. Measurement establishes the data foundation through sensor networks and imaging; Understanding converts observations into physiological insight using process-based models; Optimization applies these insights to water, nutrient, and microclimate management. Prediction uses machine learning and explainable artificial intelligence to anticipate yield, quality, and stress responses, while Diagnosis supports timely disease detection and vision-based intervention. Overall, this review shows that progress in tomato precision agriculture depends less on isolated algorithmic advances than on cost-effective, modular, interpretable, and operationally feasible systems for commercial deployment.
Lee et al. (Tue,) studied this question.