An efficient IoT-based perspective view of food traceability supply chain using optimized classifier algorithm

Abstract

The food industry is composed of a complex network of processing food and supply to the market. Supplying the food product to the market requires logistics with its operations. Nowadays the management of the food supply chain plays a vital role. In the food industry, supplying food with high quality and minimizing cost is required throughout the implementation of supply chain management. The quality of food is an essential thing for tracing the issues in food safety. Due to the increase in demand for quality food products by consumers IoT-based Food Supply Chain Management (FCM) is required for satisfying the consumer's needs. The IoT-based FCM helps the suppliers in the aspects of managing food safety procedures and tackling the issues in food safety, monitoring the lifespan of food on the manufacturing side, analyzing the process of reproduction or farming of raw material, warehousing, transporting the food product, sales in wholesaler etc. The main aim of using IoT is to receive sensor signals for monitoring the proper growth of crops in good quality soil, weather status, and so on. Also, in the transporting of food products, the sensors, and Radio-Frequency Identification (RFID) tags transmit the information to the server for monitoring the food safety parameters like temperature level, frozen level, etc. The issues are unsafe, ambiguous, the cost is high, low quality, and inaccurate. To overcome these issues, this paper proposed Internet of Things (IoT) based Food Supply Chain management is implemented. This paper proposes Tree-Augmented Naïve Bayes (TANB) with particle swarm optimization algorithm (TANB-PSO). This proposed work uses IoT which track, trace and manages the operations of food supply chain management. TANB-PSO handles optimized chronological data with IoT. The accuracy rate of three models in the FSCM of our proposed work TANB-PSO produces a higher accuracy rate of 95.02%. TANB got 92.44%, and PSO got 93.68%. The error rate of root mean square error (RMSE) and mean error (MAE) for the proposed work TANB-PSO is 0.017 and 0.031.