Industrial Internet of Things-Enabled Fault Detection and Diagnosis in Measurement and Control Systems: A Bibliometric Analysis and Systematic Mapping Review
DOI:
https://doi.org/10.31224/8128Keywords:
industrial Internet of Things, fault detection and diagnosis, condition monitoring, measurement uncertainty, edge intelligenceAbstract
Industrial Internet of Things (IIoT) architectures are shifting fault detection and diagnosis (FDD) from centralized supervisory computers toward smart instruments, edge controllers and distributed cyber-physical systems, yet evidence remains fragmented across metrology, networking, embedded computing and machine learning. This bibliometric analysis and systematic mapping review maps the field's structure and examines the measurement-to-action chain. A PRISMA-ScR informed protocol retrieved 447 OpenAlex records published in 2010-2025. After normalized-title deduplication, deterministic engineering-relevance screening, retraction checks, Crossref DOI verification and direct source/year auditing against the July 2026 official Scopus source list, 192 journal articles and reviews formed the frozen analytic corpus. The source list was used for source/year parity, not document-level Scopus coverage. All 192 DOI-title pairs matched Crossref. Output rose from five papers in 2018 to 55 in 2025 (40.9% compound annual growth). The corpus spans 97 sources, 742 disambiguated author identifiers and 54 countries. Bradford zones showed a compact core but not a strict geometric source sequence; a discrete Lotka model yielded an exponent of 4.16 with a bootstrap goodness-of-fit p value of 0.306. A three-community co-citation network had moderate modularity (Q = 0.267). Explicit edge-cloud architecture occurred in 86.3% of 2023-2025 title/abstract records, whereas explicit decision or closed-loop control language occurred in 8.5%. Because 34 records lacked abstracts, reporting frequencies are lower-bound indicators rather than study-quality estimates. The review concludes that progress depends on traceable measurement uncertainty, grouped cross-machine validation, calibrated abstention, bounded end-to-end latency, semantic interoperability and safely governed supervisory action.
Downloads
Downloads
Posted
License
Copyright (c) 2026 Bui Thanh Lam

This work is licensed under a Creative Commons Attribution 4.0 International License.