Toward adaptive and high‑precision integrated pest management in the big data era

Integrated pest management (IPM) has long reduced pesticide use while improving economic and ecological sustainability through monitoring and systems analysis. However, traditional IPM models face limitations in predictability, cost, and system specificity. Recent advances in machine learning (ML) provide flexible predictive frameworks that enable reliable short‑term pest forecasting when sufficient data are available. At the same time, Internet of Things (IoT) technologies enable continuous acquisition of pest monitoring data for ML-based predictions. They also collect outcome metrics, such as yield, pest resistance, and environmental impacts, supporting feedback-driven IPM optimization. Emerging multimodal modeling approaches now offer new opportunities to integrate diverse data sources, including textual information, and guide more targeted, integrated, and minimally chemical-dependent intervention strategies. Combining IoT monitoring, ML-based pest prediction, and adaptive optimization supports diverse ways of delivering actionable IPM insights to stakeholders, from large-scale enterprises to smallholder farming communities. This convergence of technologies marks the emergence of truly adaptive, high‑precision IPM in the big‑data era.