Risk Analysis Under Uncertainty, Subjectivity, and Incomplete Knowledge: With a Use Case of Energy System Failures

Abstract

The reliability of gas turbines is crucial due to their critical applications in energy systems and the increasing complexity of their design and operation. Traditional failure mode and effects analysis (FMEA) methods face significant limitations in handling combined uncertainty under conditions of ambiguity and partial information. Although widely used, fuzzy variations of FMEA naturally fall short in simultaneously addressing both sources of uncertainty: Ambiguity and incomplete knowledge. This study investigates the application of fuzzy-rough FMEA (FR-FMEA) to bridge this gap. By integrating fuzzy logic with rough set theory, FR-FMEA effectively manages uncertainties arising from incomplete knowledge and vagueness in expert judgment, providing a more reliable framework for risk prioritization. A case study on a gas turbine demonstrates the application of the proposed method. The results show that FR-FMEA provides distinct and reliable rankings, reducing clustering while aligning more closely with conventional RPN rankings. Key components such as the combustion chamber, fuel nozzle, and turbine rotor were consistently identified as high-risk across methods, emphasizing their criticality for maintenance and design optimization. Moreover, the results are also compared with conventional FMEA and Fuzzy-FMEA to highlight the differences.