A lower approximation based integrated decision analysis framework for a blockchain-based supply chain

Organizations and their supply chains generate vast amounts of structured and unstructured data, popularly known as Big Data. This paper proposes a framework in the prescriptive analytics category of big data analytics (BDA) and is especially relevant for organizations on a permissioned blockchain network. The framework in-tegrates the concepts of Artificial intelligence (Case-based reasoning, Interval-valued rough fuzzy set theory, and Value closeness relation algorithm) and blockchain to help decision-makers make accountable and conformable decisions. The framework comprises five stages -the first three stages generate the decisions, while the last two stages enable validation and communication of decisions. Its application is demonstrated on an organizational dataset with 20 supply chain KPIs. The numerical example shows how the organization's business strategy can translate to framework parameters resulting in varying outputs. The proposed framework is compared with the rough set approach through sensitivity analysis, highlighting the differences. The concept of L-Graphs is pro-posed to facilitate the visual display of prescribing decisions, which can help decision-makers make optimal decision choices. The framework has several unique features: (i) it is one of the first decision-making frameworks proposed for a blockchain-based supply chain that can be a part of the BDA toolbox of an organization; (ii) it uses concepts of artificial intelligence; (iii) the optimal decisions are proposed based on analysis of multicriteria; (iv) it uses multiple approaches to validate the decision rules on blockchain.

Automated summary

This paper proposes a decision-making framework based on blockchain and artificial intelligence for organizations and their supply chains to analyze and validate big data. It has unique features and advantages in application.