![]() The results of the case study are discussed in detail in Section 6. Section 4 introduces the integrated approach, and Section 5 gives a case study. In Section 3, we explain the theoretical underpinnings of the study, such as the linguistic fuzzy method, Bayes’ theorem, and the beta-binomial distribution. Section 2 describes the main shortcomings of traditional FMEA and its improvements. The rest of this paper is organized as follows. The theoretical and practical contributions of the study are discussed in detail in the final section. The major contributions of the study are (1) to enrich the research literature on FMEA by considering both stochastic and fuzzy uncertainties simultaneously (2) a fuzzy beta-binomial distribution evaluation method is proposed to precisely describe the experts’ evaluation (3) this study presents a method for solving complex RPN models containing random uncertainty and fuzzy uncertainty with Markov Chain Monte Carlo (MCMC) method. In this way, the fuzzy uncertainty and random uncertainty in the expert scoring process can be considered simultaneously to establish an integrated evaluation method. Furthermore, an expert’s scoring result for this specific factor comes from a fuzzy linguistic evaluation. It regards the evaluation process of all experts for a specific factor (such as factor S) as a stochastic process that conforms to the beta-binomial distribution with n = 10. This study presents a novel method that incorporates fuzzy and probabilistic theories to compute RPN in fuzzy and stochastic uncertainty environments. However, few studies have examined the effect of random uncertainty and linguistic fuzzy uncertainty in expert assessment on the results of FMEA evaluation when they act together. The most important extension is the studies that consider the linguistic fuzzy uncertainty of expert evaluation in the RPN calculation. Thus, researchers have attempted to improve the traditional FMEA method from various aspects and have adopted different methods to make it more adaptable. Īlthough the FMEA has been studied for nearly 60 years, the theory and method still have many shortcomings (see Literature Review). For more information, please refer to the article. Three risk factors were assessed using a 10-point scale to obtain RPNs for potential failure modes. Finally, the continuous improvement activities are implemented to reduce the risk of failure modes. Fourth, the critical failure modes are identified based on RPN rankings. In the third step, the three factors of severity, occurrence, and detectability are multiplied together to calculate the so-called risk priority number (RPN): RPN = S × O × D. Second, three risk factors are taken into account for each potential failure mode: the occurrence/probability of the failure ( O), the severity of the consequences ( S), and the chance/probability of the failure going undetected ( D). ![]() First, a group of experts are to identify all possible potential failure modes of the product or system. The traditional FMEA analysis comprises five steps. At present, it has been widely adopted to improve the security and reliability of systems and for continuous improvements in product or process design in various fields, for instance, wind power, food, healthcare, fabrics, construction, healthcare, and mining. IntroductionįMEA was initially developed as a formal design methodology in the 1960s in the aerospace industry. The study presented a case study, which presented how to apply this model in practice and indicated the significant influence on the measurement error caused by ignoring the random uncertainty caused by expert evaluation in RPN calculations. The major contribution of the proposed model is to use the random uncertainty and fuzzy uncertainty in an integrated model and provide a Markov Chain Monte Carlo (MCMC) method to solve the complex integrated model. This model can effectively realize real-time, dynamic, and long-term evaluation of RPN under the condition of continuous knowledge accumulation. In this study, a fuzzy beta-binomial RPN evaluation method is proposed by integrating fuzzy theory, Bayesian statistical inference, and the beta-binomial distribution. Accordingly, increasing studies have ignored the important impact of the random sampling uncertainty in the FMEA assessment. Recently, RPN research under a fuzzy uncertainty environment has become a hot topic. The risk priority number (RPN) calculation method is one of the critical subjects of failure mode and effects analysis (FMEA) research.
0 Comments
Leave a Reply. |