A quantitative fuzzy Risk Graph framework successfully identifies and corrects critical safety misallocations produced by conventional risk assessment techniques, according to a study published in Scientific Reports. In industrial safety engineering, technology licensors widely rely on the traditional qualitative Risk Graph method to allocate Safety Integrity Levels. However, when applied to operational scenarios that feature multiple independent Initiating Events, this conventional method often yields ambiguous or erroneous results.
To overcome these limitations, researchers Mohammad Sadegh Haghgoo, Mokhtar Shasadeghi, and Behrouz Safarinejadian validated an uncertainty-aware fuzzy framework using an industrial case study. Their research demonstrates that integrating fuzzy logic directly addresses and rectifies the specific miscalculations inherent in traditional licensor assessments. By handling uncertainty more effectively, the framework provides an analytical alternative for complex industrial plants where traditional qualitative methods fail.
Under the proposed quantitative framework, probabilistic input intervals are fuzzified and processed through a Mamdani inference system. This computational model is specifically calibrated to reproduce conventional Safety Integrity Level assignments at category centers. At the same time, the system is designed to explicitly aggregate Demand Rates across multiple independent Initiating Events, ensuring that cumulative operational demands are rigorously accounted for during safety evaluations.
The researchers tested and benchmarked the framework against the original licensor baseline across 13 Safety Instrumented Functions within the synthesis-gas unit of a 3000-ton-per-day ammonia plant. The comparative analysis demonstrated that the conventional licensor methodology misallocated Safety Integrity Levels in four critical cases out of the 13 functions assessed in the primary reformer facility.
Notably, the conventional approach underestimated the required Safety Integrity Level from 2 to 3 for two Safety Instrumented Functions that were exposed to seven and eight Initiating Events. The study authors concluded that the validated fuzzy Risk Graph approach provides a more accurate, discriminative, and uncertainty-aware basis for determining Safety Integrity Levels in complex multi-event industrial scenarios.
Subject of Research: Safety integrity level allocation using fuzzy logic in ammonia plant systems
Article Title: Fuzzy logic-based safety integrity level determination for multi-initiating-event scenarios: a case study of an ammonia primary reformer
Article References: Haghgoo, M. S., Shasadeghi, M., & Safarinejadian, B. (2026). Fuzzy logic-based safety integrity level determination for multi-initiating-event scenarios: a case study of an ammonia primary reformer. Scientific Reports. https://doi.org/10.1038/s41598-026-80643-6
Image Credits: AI Generated
DOI: 10.1038/s41598-026-80643-6
Keywords: fuzzy logic, safety integrity level, risk graph method, initiating events, ammonia primary reformer, safety instrumented functions, Mamdani inference system, synthesis-gas unit, demand rates, industrial safety, risk allocation, probabilistic input intervals, licensor assessments, ammonia plant, scientific reports, uncertainty analysis, multi-initiating events
Cite Scienmag News
Denise Maddox. (November 22, 2026). Fuzzy Risk Graph Corrects Safety Allocation Errors in Ammonia Plant Study. Scienmag. https://scienmag.com/fuzzy-risk-graph-corrects-safety-allocation-errors-in-ammonia-plant-study/
Denise Maddox. "Fuzzy Risk Graph Corrects Safety Allocation Errors in Ammonia Plant Study." Scienmag, 22 November 2026, https://scienmag.com/fuzzy-risk-graph-corrects-safety-allocation-errors-in-ammonia-plant-study/. Accessed 22 November 2026.
Denise Maddox. "Fuzzy Risk Graph Corrects Safety Allocation Errors in Ammonia Plant Study." Scienmag. November 22, 2026. https://scienmag.com/fuzzy-risk-graph-corrects-safety-allocation-errors-in-ammonia-plant-study/








