A systematic mapping of artificial intelligence, digital twins, and blockchain applications in nuclear plants

Abstract

The nuclear energy industry, including large-scale plants and small modular reactors (SMRs), confronts challenges that undermine efficiency, safety, and public trust. These include fragmented supply chains limiting scalability, escalating cybersecurity threats from control systems, proliferation risks due to dual-use materials, inadequate waste management practices with transparency gaps, and unreliable traceability of components and fuels. Such issues not only hinder operational reliability but perpetuate societal concerns rooted in historical accidents. In response, digital technologies such as artificial intelligence (AI), digital twins (DTs), and blockchain (BC) are emerging as tools for transformation. This systematic review, employing Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 methodology, synthesizes 161 studies published from 2011 to the present, sourced from Web of Science and Scopus. This analysis consolidates fragmented research and provides actionable insights for industry stakeholders, including guidance for developing systems that enhance transparency, strengthen compliance, and rebuild public trust in nuclear energy. It also employs an independent screening reliability assessment and a structured quality appraisal rubric to enhance transparency, reproducibility, and the credibility of evidence. Findings reveal the prominence of AI in reactor autonomy and waste classification, the applications of DTs in lifecycle management and training, and the role of BC in supply chain visibility and safeguards. This systematic review highlights a key gap: the absence of a unified approach that combines these technologies to address systemic vulnerabilities comprehensively. This review offers a tridomain evidence map that systematically charts how AI, DTs, and BC have been investigated across distinct nuclear engineering subdomains.