An Efficient Bayes Classifiers Algorithm for Traceability of Food Supply Chain Management using Internet of Things

Abstract

The conventional Food Supply Chain Management (FSCM) faces a variety of provocation such as ambiguity, security, cost, complication and quality concerns. To resolve these issues, supply chain must be precise. A challenging assignment in today’s food industry is distributing the high quality of foods throughout the supply chain management. In this paper, proposes an efficient Bayes Classifiers Algorithm which integrated with FSCM using Internet of Things (IoT) to allow tracking, tracing and managing the entire process of food supply chain such as supplier, exporter and customers. The objective of this paper is to determine the food safety and to optimize chronological data produced to analyze the effective possibility of future assump- tions. It also aims to foods carrying from the manufacturers to the customers with help of IoT technologies to bond the producer to the customer with delivery of high class of food products. IoT based Food Supply Chain Traceability is utilized t o data transaction effectively with indeterminate, uncertain and insufficient information. So the proposed efficient Bayes Classifiers Algorithm will be capable to overcome all provocation of conventional supply chain and afford secure background and food safety for FSCM process using IoT technology.

Application of machine learning to the monitoring and prediction of food safety: A review

“…Balamurugan et al. (2019) report on the development of a Bayes classifier for monitoring any irregular food condition (e.g., contaminated food needed to be recalled) throughout the entire food supply chain—from the producer to the consumer—based on internet‐of‐things (IoT) data. Their results can be used to prioritize food monitoring in order to ensure high food quality.…”

“…This can result in an incorrect and/or unrealistic estimate of model performance (Kuhn & Johnson, 2019). Most of the selected articles apply internal validation, that is, cross‐validation (Balamurugan et al., 2019) or random subsampling validation (Geng et al., 2017; Sun et al., 2013; Wang et al., 2017; Zhang et al., 2018). In these studies, unused data from the whole dataset are used to validate the model performance.…”

“…This can result in an incorrect and/or unrealistic estimate of model performance (Kuhn & Johnson, 2019). Most of the selected articles apply internal validation, that is, crossvalidation (Balamurugan et al, 2019) or random subsampling validation (Geng et al, 2017;Sun et al, 2013;Wang et al, 2017;Zhang et al, 2018) In addition to evaluating the generalization capacity of the ML model, the validation process also distinguishes ML models from statistical models. Although statistical models serve primarily to make inferences, ML models provide predictions and classifications.…”

“…Information related to the food/feed and food/feed business operators can be gathered from a traceability system and analyzed to facilitate the monitoring of the supply chain (Balamurugan et al, 2019). Studies mainly focus on traceability and monitoring general food safety hazards in food supply chains (Alfian, Syafrudin, Farooq, et al, 2020;Balamurugan et al, 2019;Wang & Liu, 2019), as well as traceability based on animal features, which can trace the original source, uphold food safety standards, and ensure consumer confidence (Bennion et al, 2019;Ibáñez, 2015;Song et al, 2019;Tharwat et al, 2014). Balamurugan et al (2019) report on the development of a Bayes classifier for monitoring any irregular food condition (e.g., contaminated food needed to be recalled) throughout the entire food supply chain-from the producer to the consumer-based on internet-of-things (IoT) data.…”

“…These devices and technologies (including radiofrequency identification technology, laser scanners, infrared sensors, information sensors, and global positioning systems) are able to collect various types of information (e.g., with regard to chemistry, biology, light, sound, heat, electricity, location, or mechanics) for food products in the supply chain. Balamurugan et al (2019) use data from IoT systems for Bayes classifier modeling to build a traceability system for the food supply chain, which facilitates decision-making concerning the quality of food products. This system provides tracing and monitoring functions across various stages in the food supply chain (e.g., production, processing, warehousing, distribution, retail, and the end customer).…”