A Design of IoT-based Monitoring System for Intelligent Meat Quality in Vyas Municipality, Nepal

Abstract

The integration of Internet of Things (IoT) technologies and machine learning (ML) has introduced transformative solutions in food quality monitoring, particularly in developing nations like Nepal. This research presents an IoT-based smart monitoring system designed to assess meat quality through environmental and gas-sensing techniques. Methane and ammonia gases primary indicators of meat spoilage were measured using gas sensors, alongside temperature and humidity sensors. Laboratory experiments were conducted to validate sensor reliability by producing controlled gases and comparing outputs with standard indicators. The validated system was deployed across meat shops in Vyas Municipality for six months, monitoring buff, mutton, chicken, and fish. Findings indicated that methane and ammonia levels began to increase after 3 hours at room temperature and peaked significantly between 48 to 60 hours. Fish exhibited the fastest spoilage rate, with high gas levels detected within 5 hours of exposure. Machine learning models including random forest, decision tree, and logistic regression were trained on the collected dataset. The random forest model achieved the highest accuracy and was used for predicting meat freshness. Analysis revealed that 95% of fish samples were sold in spoiled condition, while 60% of buff meat and 10% of mutton were found deteriorated. In contrast, chicken meat was sold 100% fresh.This study, conducted with support from the Vyas Municipality Office, Nepal Food Institute Tanahun, and the Meat Traders Association, highlights the alarming state of meat quality in local markets. It also addresses the absence of standardized methods for evaluating meat freshness in Nepal, where consumers largely rely on sensory judgment. The proposed system offers a scalable, cost-effective, and data-driven approach for real-time meat quality monitoring. By leveraging IoT and ML technologies, this research contributes a practical solution to improve food safety and public health awareness, particularly in resource-constrained settings. It also provides a replicable framework for similar applications in other developing countries facing parallel challenges in food quality assurance.