A new approach for evaluation of risk priorities of failure modes in FMEA

Abstract

This paper presents a method for carrying out the calculus of the risk priority of failures in Failure Mode and EŒect Analysis (FMEA). The novelty of the method consists of new management of data provided by the design team, normally given on qualitative scales, without necessitating an arbitrary and arti ® cial numerical conversion. The practical eŒects of these issues are shown in an application example. 1. Preliminary considerations Since its introduction as a support tool for designers, FMEA (Failure Mode and EŒect Analysis) has been extensively used in a wide range of contexts (Stamatis 1995, Hatty and Owens 1995, Bowles 1998). The great number of papers published in various technological and service areas bear witness to this interest (Hatty and Owens 1995, Wirth et al. 1996). Designers ’ interest in FMEA is due to its capacity to perceive two very important aspects:. the capability of stimulating the application of the continuous improvement concept in design (Franceschini and Rossetto 1995b);. the possibility of methodical documenting of the design evolution. FMEA is a reliability tool, which requires identifying failure modes of a speci ® c product or system, their frequency and potential causes. It is normally applied by an interfunctional work team, with the right know-how to analyse the whole product life cycle. As a result of its application, it allows `quantifying ’ how `dangerous ’ a failure mode is, and also provides a rank of risk priorities of failure modes and a list of corrective actions to remove them. A typical form used for FMEA development is illustrated in table 1. It shows a list of items that identify:. the system or component part. the potential failure mode. the potential eŒect of failure. the severity index (S)