Bayesian Business Continuity Analytics for Cyber-Disrupted Industrial Process Workflows
Journal of Business and Data Analytics · 30.06.2026 · 4 (8) · 133-152
DOI: 10.63646/tjrk5925 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Business and Data Analytics, online erschienen am 30.06.2026. Verfasst von Yisong Chen und 2 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Industrial process workflows depend on interconnected operational technology, information systems, people, and external services. When a cyber incident interrupts or undermines confidence in one dependency, continuity decisions must be made before the state of the whole workflow is known. This article develops a Bayesian decision framework that connects uncertain cyber and physical states to production, quality release, contractual obligations, and recovery choices. A dynamic Bayesian network represents evolving asset and workflow states; evidence from security alerts, process observations, and verified recovery checks updates the posterior distribution. A decision layer evaluates feasible actions in terms of expected business loss, service shortfall, and, where appropriate, downside risk, subject to independently assessed safety and operational constraints. The framework makes explicit the difference between inferring a state, predicting an outcome under an action, and authorizing an intervention. A fully specified hypothetical example illustrates how an imperfect diagnostic signal changes a choice between controlled shutdown and a conditionally available isolated operating mode. With an initial compromise probability of 0.20 and the stated loss assumptions, the minimum expected loss falls from 28 to 20 thousand US dollars before a diagnostic cost; the expected value of the signal is therefore 8 thousand US dollars. This is an analytical illustration, not an empirical estimate or a claim of tested plant performance. The article discusses dependence, sparse data, calibration, model governance, and a prospective evaluation design. Its contribution is a transparent bridge from evolving evidence to safety-constrained continuity decisions, with a reproducible example and clearly stated limits on operational use.
Zitieren
APA
Chen, Y., Nie, C., & Tian, Y. (2026). Bayesian Business Continuity Analytics for Cyber-Disrupted Industrial Process Workflows. Journal of Business and Data Analytics, 4(8), 133-152. https://doi.org/10.63646/tjrk5925
Vancouver
Chen Y, Nie C, Tian Y. Bayesian Business Continuity Analytics for Cyber-Disrupted Industrial Process Workflows. Journal of Business and Data Analytics. 2026;4(8):133-152. doi:10.63646/tjrk5925
BibTeX
@article{Chen2026,
author = {Chen, Yisong and Nie, Chuanhao and Tian, Yexin},
title = {Bayesian Business Continuity Analytics for Cyber-Disrupted Industrial Process Workflows},
journal = {Journal of Business and Data Analytics},
year = {2026},
volume = {4},
number = {8},
pages = {133--152},
doi = {10.63646/tjrk5925}
}
EUVIMED