Design and Experimental Validation of a Low-Cost Cold Storage System for Mixed Vegetables Using Internet of Things

Small cold rooms can significantly reduce postharvest losses, yet many facilities still operate without continuous environmental supervision or reliable feedback control. This gap becomes particularly important when mixed vegetables with different physiological requirements are stored together. In this study, we develop and evaluate a low-cost Internet of Things monitoring and control system for small cold storage environments, with emphasis on reproducibility, validation, and operational reliability. The prototype integrates an ATmega328 microcontroller, ESP8266 wireless connectivity, DHT22 temperature and humidity sensing, local SD-card logging, and mobile visualization. We tested the system in a 0.25 m³ insulated chamber storing tomatoes, cucumbers, lettuce, and carrots at approximately 5 °C and relative humidity above 90 percent for seven days. The controller maintained temperature within approximately ±1 to 1.5 °C of the target range and generated alerts during induced disturbances, while relative humidity remained high despite the absence of active humidification. Agreement with a calibrated reference logger was strong, with a temperature bias of-0.05 °C and RMSE of 0.06 °C, and a relative humidity bias of-0.30 percent with RMSE of 0.34. Abnormal-event testing demonstrated timely alerts and recovery to the control band within 40-45 minutes following a simulated power outage. Energy consumption was not measured directly and produce quality was assessed qualitatively; therefore, the findings reflect a single chamber, short storage duration, and limited instrumentation. The results show that affordable sensing combined with simple feedback control can maintain stable storage microclimates for short-term mixed-vegetable holding and provide a validated baseline for future work involving multi-site field trials, direct energy measurement, and quantitative quality assessment.