Selection of a viable blockchain service provider for data management within the internet of medical things: An MCDM approach to Indian healthcare

Raghunathan Krishankumar,Sundararajan Dhruva,Kattur S Ravichandran,Samarjit Kar

DOI: https://doi.org/10.1016/j.ins.2023.119890

IF: 8.1

2024-02-01

Information Sciences

Abstract:Health 4.0 is gaining significant attention globally to support better treatment and care for people. Digital technologies such as the IoMT “(”Internet of Medical Things“”) and blockchain substantially promote quality health services. Literature shows that embedding blockchain in IoMT is an effective way to attain secure and quality healthcare. Selecting a viable blockchain service provider (BSP) becomes a complex task and can be considered an MCDM “(”Multi-Criteria Decision-Making“”) problem. Earlier studies on BSP selection indicate that complex expressions cannot be well-modelled and methodical estimation of decision parameters by capturing hesitation and interaction of entities is not adequately explored. Driven by the issues, authors put forward a novel MCDM framework with (i) a double hierarchy linguistic structure for data collection in natural expressions; (ii) regret measure for experts' reliability calculation; (iii) a weighted CRITIC approach for criteria weight determination, and (iv) CRADIS-Copeland algorithm for ranking BSPs. Finally, a case example from Recent Indian healthcare is exemplified to demonstrate the framework's applicability. Recently, the Indian healthcare sector launched initiatives/plans as part of Health 4.0 to promote data management for a better quality of treatment by seeking support from digital technologies such as IoMT and blockchains to ensure data privacy and improved customer experience. Sensitivity analysis and comparison with extant methods infer that (i) BC 1 , BC 5 , and BC 3 are the top three BSPs for the considered problem; (ii) criteria such as privacy aspects, intercommunication capability, global accessibility, and total cost constitute 58% importance in the supposed decision problem; (iii) developed model reduces human intervention/biases; (iv) developed model is robust to weight alteration and yields unique rank orders with both personalised and cumulative ordering of BSPs; and (v) proposed model produces broader rank values that aid in effective backup planning.

computer science, information systems