AI-Driven Cold Storage Optimization using Charcoal-Perlite Insulation for Extended Produce Shelf-Life

The conventional methods of storing agricultural produce are unreliable and necessitate prompt transportation in order to preserve their nutritional value over time. Therefore, a system that provides an innovative, cost-effective, and insightful cold storage solution designed particularly for rural agriculture supply networks has been proposed. This method significantly reduces the dependence on traditional storage methods and frequent transportation by addressing problems with workforce management, machinery, product wastage, and logistics. The system incorporates charcoal-based evaporative cooling devices augmented with radial fans for smooth airflow, perlite insulation for temperature maintenance, and energy-efficient Peltier plates for effective cooling. In addition, real-time tracking, predictive anomaly detection, predictive maintenance, and intelligent market analytics are enabled by the combination of advanced Generative AI algorithms, including Time Series Forecasting models (LSTM and ARIMA), GPT-4- based conversational AI, and anomaly detection techniques (Isolation Forest and Autoencoder Neural Networks). Experimental deployment in rural Coimbatore, Tamil Nadu, reflected a substantial improvement in shelf-life produce and storage efficiency (approximately 65%), directly impacting approximately $\mathbf{7 0 \%}$ of local farmers. With its smart, real-time storage management and market-oriented suggestions, this AIbased cold storage solution presents a groundbreaking approach for supply chain optimization and green agriculture.