TrustFish: A decentralized AI framework for trustworthy end-to-end seafood quality and safety monitoring

This paper introduces a novel framework called ”TrustFish” to monitor seafood quality and safety, ensuring effective and efficient traceability in the seafood supply chain from fishers to customers. The challenges of seafood traceability include (i) high verification costs, (ii) decentralized structure of the supply chain, (iii)volume of heterogeneous data, and (iv) the lack of fail-safe detection techniques. The TrustFish framework combines decentralized technologies, i.e., federated learning and blockchain, to develop a decentralized, secure, and privacy-preserving system for seafood monitoring in the supply chain. Additionally, data from the supply chain is gathered by Internet of Things devices for monitoring the safety and quality of seafood. In TrustFish, dynamic sharding and directed acyclic graph are used to improve fault tolerance and scalability in diverse supply chain network environments. TrustFish demonstrates how FL and blockchain combine to produce a cooperative, effective, and reliable seafood monitoring. By giving stakeholders access to thorough product histories and environmental circumstances, TrustFish increases consumer trust, decreases contamination, and lowers fraud by allowing stakeholders to access seafood histories and environmental circumstances. The proposed TrustFish solution improves operational effectiveness and public health results while laying the groundwork for the seafood supply chain’s advancement in Industry 5.0.