Quality Estimation of Perishables in Cold Chain Network using Machine Learning: A New Approach

Perishable food preservation is an important aspect of cold chain logistics operations. Changes in humidity and temperature in cold reefers typically cause perishable food to deteriorate. The current cold chain design in India is ineffective at maintaining these values constant during transport. Therefore, a smart cold chain system is required to avoid food wastage in the cold chain network. In this study, an IoT-based system is developed to monitor the temperature and humidity of perishables during transit considering various vehicle characteristics. The study suggests measuring the Mean Kinetic Temperature (MKT) that takes into account the biochemical changes in food caused by temperature fluctuations. Machine learning algorithms are being used to estimate the quality of perishables, which is a significant advancement in cold chain technology. Machine learning algorithms improve the accuracy of time-temperature data prediction, thereby preserving food quality during transportation. The cloud and a mobile app are used to send an early warning message about temperature abuse to the concerned person. In addition, a comparative analysis of algorithms is carried out to recommend the best algorithm for prediction. The outcomes are compared to those of real-time applications