Investigating innovative technique for managing Failure Modes and Effects Analysis

The investigation of failures and their prevention throughout a design process forms an effective approach that can reduce such occurrences during manufacturing. To address this issue, Failure Modes and Effects Analysis (FMEA) was developed as a methodology to proactively evaluate the potential failures in a product, process or service. In the context of the product development process, FMEA is employed in three distinct areas, labelled: (1) Functional FMEA – looking at the potential failures associated with product and system functions, (2) Design FMEA – considering the potential failures identified in the product sub-system functions and component designs and (3) Process FMEA – evaluating the potential failures found in manufacturing and assembly processes. In its current form, FMEA implementation is marred by high resource requirements, lack of available data and difficulty of deployment. Consequently, attempts to integrate FMEA with other systematic methodologies have, at best, been sporadic, yielding inconclusive outcomes. To address these problems, this research used a score-based metric to empirically identify the challenges associated with FMEA deployment and the impact of combining other methodologies with it. The influence of other factors, such as industry membership, amount and level of engineering experience held and organisational size were also analysed by executing an ordered probit model, which sought to examine the effect of such factors on the score-based metric developed to measure the level of challenge associated with FMEA deployment. The findings reveal that Fault Tree Analysis and Axiomatic Design methodologies significantly reduce the perceived level of challenge for FMEA deployment in the sample of firms that were investigated. These findings represent a promising development as it has helped to improve the understanding of the impact associated with integrating these methods and the variation in other managerial factors on FMEA implementation. In terms of measuring the performance of FMEA deployment, this is also an area that has not previously been clearly understood in the literature. In order to investigate this part of FMEA deployment, this research applied graphical and descriptive analysis methods in order to understand the current systems/targets used for measuring FMEA-related performance. The results identify diverse approaches/targets that were recognised as being the most frequently used based on the sample investigated, in the following order: mitigating failure modes, the capability of FMEA practice to recognise failure modes and document the relevant knowledge and actions. It was also noted that these approaches were widely managed subjectively for assessing the themes outlined above. These results suggest that these are the most important goals for deploying FMEAs. Also, these results indicate the demand for a computable metric, which would help to certify the outputs of FMEA execution. Initial investigation was then conducted into determining such a cost model. However, the host company was unable to disclose its records to validate the proposed cost model and so the validation process will be conducted at a later date. Building on the above findings of the ordered probit model and the targets required by deploying FMEAs in industry, this research also sought to integrate Functional, Design and Process FMEAs into a working model in an attempt to achieve its targets outlined in above; an area that has been hitherto under-researched. This research developed the framework of combining FMEA applications to offer a hybrid risks assessment model that should be able to synchronise FMEA applications and catch the cascaded failures. In this regard, it could be said that the framework developed in this research is novel, as it aims to improve FMEA deployment in and reduce associated costs. The developed framework was then evaluated based on feedback provided by FMEA subject-matter experts, whose responses were analysed using the t-test model. The findings suggest that this framework is easy to use, useful, applicable and able to cascade and catch failures over Functional, Design and Process FMEAs