Towards data-driven smart composting techniques and control systems

Composting is a widely adopted method for recycling organic waste into stable organic matter, thereby supporting sustainable agriculture and waste management. However, conventional practices rely heavily on manual regulation of aeration, turning, and moisture, often resulting in inefficiencies, nutrient losses, and inconsistent product quality. Data-driven smart composting offers a promising alternative by integrating environmental sensing, Internet of Things (IoT) connectivity, and machine learning-based modeling for real-time, adaptive control. Continuous monitoring of temperature, moisture, oxygen, and odor-related gases, combined with predictive algorithms, enables dynamic adjustment of process conditions to improve stability, conserve nutrients, and reduce emissions. This review synthesizes recent progress across four domains: sensor technologies, data-driven modeling, intelligent control, and deployment at household, community, and industrial scales. Despite rapid advances, critical barriers remain, including sensor durability in corrosive environments, limited standardized datasets, and the computational scalability of control systems. Future research should focus on developing robust and corrosion-resistant sensors, establishing standardized indicator systems linking sensor signals to compost quality, and creating tiered AI-based decision-support tools tailored to different operational scales. Addressing these challenges will accelerate the transition of smart composting from pilot demonstrations to scalable, low-carbon systems that enhance nutrient recycling, reduce nitrogen and greenhouse gas losses, and position composting as a cornerstone technology for circular agriculture.