Software reliability modeling: an approach to early reliabilityprediction

摘要

Models for predicting software reliability in the early phases ofndevelopment are of paramount importance since they provide earlynidentification of cost overruns, software development process issues,noptimal development strategies, etc. A few models geared towards earlynreliability prediction, applicable to well defined domains, have beenndeveloped during the 1990s. However, many questions related to earlynprediction are still open, and more research in this area is needed,nparticularly for developing a generic approach to early reliabilitynprediction. This paper presents an approach to predicting softwarenreliability based on a systematic identification of software processnfailure modes and their likelihoods. A direct consequence of thenapproach and its supporting data collection efforts is thenidentification of weak areas in the software development process. AnBayes framework for the quantification of software process failure modenprobabilities can be useful since it allows use of historical data thatnare only partially relevant to the software at hand. The keyncharacteristics of the approach should apply to othernsoftware-development life-cycles and phases. However, it is unclear howndifficult the implementation of the approach would be, and how accuratenthe predictions would be. Further research will help answer thesenquestions