Implementation of circular blockchain-based approach for food crops supply chain with bitcoin prediction using deep learning

Abstract

Blockchain technology serves as a framework for addressing the challenge of tracking and marketing materials in distributed networks. Agricultural traceability systems in food production provide farmers with food safety and smart contracts. Smart contracts establish cryptocurrency evidence of delivery with automated bitcoin payments to all parties. In this paper, we propose circular blockchain-based traceability and bitcoin prediction in agricultural food crop system to accomplish transparency and traceability. Bitcoin prediction using LSTM benefits farmers and stakeholders to buy and sell their food crops when the profit is high. Moreover, the proposed system provides safety, consensus, shared ledger, speedy payment, and decentralization. All activities are processed in a distributed shared ledger with connections to a decentralized file system, which makes the supply chain visible and traceable. All the activities that are tied up in the supply chain are transparent to consumers and stakeholders. Traceability of the agricultural product is done efficiently with QR code, which results in a trustable relationship between the farmer and the consumer. Finally, our research shows that the accuracy of LSTM outperforms by 88.67% compared to traditional machine learning algorithms such as SVM and Naive Bayes with accuracy of 62.025 and 75.32% in forecasting the bitcoin price.