John B. Bowles

Fuzzy logic prioritization of failures in a system failure mode, effects and criticality analysis

Abstract This paper describes a new technique, based on fuzzy logic, for prioritizing failures for corrective actions in a Failure Mode, Effects and Criticality Analysis (FMECA). As in a traditional criticality analysis, the assessment is based on the severity, frequency of occurrence, and detectability of an item failure. However, these parameters are here represented as members of a fuzzy set, combined by matching them against rules in a rule base, evaluated with min-max inferencing, and then defuzzified to assess the riskiness of the failure. This approach resolves some of the problems in traditional methods of evaluation and it has several advantages compared to strictly numerical methods: 1) it allows the analyst to evaluate the risk associated with item failure modes directly using the linguistic terms that are employed in making the criticality assessment; 2) ambiguous, qualitative, or imprecise information, as well as quantitative data, can be used in the assessment and they are handled in a consistent manner; and 3) it gives a more flexible structure for combining the severity, occurrence, and detectability parameters. Two fuzzy logic based approaches for assessing criticality are presented. The first is based on the numerical rankings used in a conventional Risk Priority Number (RPN) calculation and uses crisp inputs gathered from the user or extracted from a reliability analysis. The second, which can be used early in the design process when less detailed information is available, allows fuzzy inputs and also illustrates the direct use of the linguistic rankings defined for the RPN calculations.

Using fuzzy cognitive maps as a system model for failure modes and effects analysis

This paper explores the application of Fuzzy Cognitive Maps (FCM) to Failure Modes and Effects Analysis (FMEA). FMEAs are used in reliability and safety evaluations of complex systems to determine the effects of component failures on the system operation. FCMs use a digraph to show cause and effect relationships between concepts; thus, they can represent the causal relationships needed for the FMEA and provide a new strategy for predicting failure effects in a complex system.

Application of fuzzy logic to reliability engineering

The analysis of system reliability often requires the use of subjective-judgments, uncertain data, and approximate system models. By allowing imprecision and approximate analysis fuzzy logic provides an effective tool for characterizing system reliability in these circumstances; it does not force precision where it is not possible. Here we apply the main concepts of fuzzy logic, fuzzy arithmetic and linguistic variables to the analysis of system structures, fault trees, event trees, the reliability of degradable systems, and the assessment of system criticality based on the severity of a failure and its probability of occurrence. >

An assessment of RPN prioritization in a failure modes effects and criticality analysis

The risk priority number (RPN) methodology for prioritizing failure modes is an integral part of the automobile FMECA technique. The technique consists of ranking the potential failures from 1 to 10 with respect to their severity, probability of occurrence, and likelihood of detection in later tests, and multiplying the numbers together. The result is a numerical ranking, called the RPN, on a scale from 1 to 1000. Potential failure modes having higher RPNs are assumed to have a higher design risk than those having lower numbers. Although it is well documented and easy to apply, the method is seriously flawed from a technical perspective. This makes the interpretation of the analysis results problematic. The problems with the methodology include the use of the ordinal ranking numbers as numeric quantities, the presence of holes making up a large part of the RPN measurement scale, duplicate RPN values with very different characteristics, and varying sensitivity to small changes. Recommendations for an improved methodology are also given.

A survey of reliability-prediction procedures for microelectronic devices

The author reviews six current reliability prediction procedures for microelectronic devices. The device models are described and the parameters and parameter values used to calculate device failure rates are examined. The procedures are illustrated by using them to calculate the predicted failure rate for a 64 K DRAM; the resulting failure rates are compared under a variety of assumptions. The models used in the procedures are similar in form, but they give very different predicted failure rates under similar operating and environmental conditions, and they show different sensitivities to changes in conditions affecting the failure rates. >

Applying fuzzy cognitive-maps knowledge-representation to failure modes effects analysis

A failure mode and effects analysis (FMEA) seeks to determine how a system will behave in the event of a device failure. It involves the integration of several expert tasks to select components for analysis, determine failure modes, predict failure effects, propose corrective actions, etc. During an FMEA, numerical values are often not available or applicable and qualitative thresholds and linguistic terms such as high, slightly high, low, etc., are usually more relevant to the design than numerical expressions. Fuzzy set theory and fuzzy cognitive maps provide a basis for automating much of the reasoning required to carry out an FMEA on a system. They offer a suitable technique to allow symbolic reasoning in the FMEA instead of numerical methods, thus providing human like interpretations of the system model under analysis, and they allow for the integration of multiple expert opinions. This paper describes how fuzzy cognitive maps can be used to describe a system, its missions, failure modes, their causes and effects. The maps can then be evaluated using both numerical and graphical methods to determine the effects of a failure and the consistency of design decisions.

Combining sneak circuit analysis and failure modes and effects analysis

The feasibility of integrating failure modes and effects analysis (FMEA) and sneak circuit analysis (SCA) into a comprehensive reliability analysis technique is examined especially from the perspective of automation. FMEA looks at a systems strengths and weaknesses; SCA looks for latent circuit conditions which may lead to unplanned or unexpected modes of operation. The goals of the two techniques complement each other and their combination results in a more comprehensive analysis than either technique alone can achieve. The rich collection of heuristics used in SCA can be applied to design validation and also used as guidelines at various stages of system design. At both the functional level and the component level, the combined analysis is done using the same circuit representation as for the SCA and for the FMEA, and draws on the same database.<<ETX>>

Using fuzzy logic for system criticality analysis

Fuzzy logic provides a tool for directly manipulating the linguistic terms that an analyst employs in making a criticality assessment for a failure modes, effects and criticality analysis (FMECA). This allows an analyst to evaluate the risk associated with item failure modes in a natural way. Appropriate actions to correct or mitigate the effects of the failure can be prioritized even though the information available is vague, ambiguous, qualitative, or imprecise.<<ETX>>

A comparison of commercial reliability prediction programs

Discussed are demonstration versions of six reliability prediction software packages: FRATE, MilStress, PC Predictor, REAPmate, ReCalc 2 and RELEX. The programs are compared on the basis of price, data-input process, output-reporting options, error handling, user friendliness, and other options. No attempt is made to rank the programs in any way since any such ranking is highly dependent on the users perspective and needs. Overall, the programs have many advanced features and provide powerful tools for assessing the reliability of a design.<<ETX>>

Functional reasoning in a failure modes and effects analysis (FMEA) expert system

The issue of representing the knowledge of how systems work is approached from a functional perspective. A knowledge base organized around a functional representation provides the inference procedure with a focus of attention directed toward expected goals, and guides the reasoning process in determining the effects of a systems failure modes. The functional representation described includes relationships to more detailed schemes, including numerical techniques and qualitative simulations of the causal behavior of systems. A functional representation is domain-general in that functional primitives provide a language more general than any one system being modeled. The blackboard framework is proposed as a comprehensive problem-solving architecture for integrating the functional approach with other simulation and representation techniques.<<ETX>>