An Automated Software FMEA

Research output: Chapter in Book/Report/Conference proceedingConference Proceeding (Non-Journal item)

Abstract

The concept of software failure mode and effects analysis (FMEA) has grown in attractiveness over recent years as a way of assessing the reliability of software. Like its hardware counterpart, software FMEA is immensely tedious for an engineer to perform, as well as being error-prone. This paper presents the implementation of a novel method for automating code-level software FMEA based on treating the implemented software as a model of the desired system and propagating faults through the model to identify dependencies. The method provides results at a level where they can be understood and acted on by software engineers. A tool implementing this method has been applied to a travel expenses payment program, and some of the automatically produced results are presented. Such automation extends significantly the range of software for which software FMEA becomes a realistic proposition. The analysis is tractable, and has been shown to provide useful results for software engineers. One important use of this analysis is to focus further testing. The software FMEA can be used to improve automated or source code embedded testing since tests can exonerate many potential faults allowing the FMEA analysis to present an engineer with a reduced set of potential faults.
Original languageEnglish
Title of host publicationProceedings of the International System Safety Regional Conference
Subtitle of host publicationextending system safety frontiers
PublisherResearch Publishing Services
Number of pages10
ISBN (Print)9789810804466
Publication statusPublished - 03 Apr 2008
EventInternational System Safety Regional Conference - Singapore, China
Duration: 01 Apr 200801 Apr 2008

Conference

ConferenceInternational System Safety Regional Conference
Country/TerritoryChina
CitySingapore
Period01 Apr 200801 Apr 2008

Fingerprint

Dive into the research topics of 'An Automated Software FMEA'. Together they form a unique fingerprint.

Cite this