Fault Prediction Model by Fuzzy Profile Development of Reliability Relevant Software Metrics

Abstract

This paper presents a fault prediction model using reliability relevant software metrics and fuzzy inference system. For this a new approach is discussed to develop fuzzy profile of software metrics which are more relevant for software fault prediction. The proposed model predicts the fault density at the end of each phase of software development using relevant software metrics. On the basis of fault density at the end of testing phase, total number of faults in the software is predicted. The model seems to useful for both software engineer as well as project manager to optimally allocate resources and achieve more reliable software within the time and cost constraints. To validate the prediction accuracy, the model results are validated using PROMISE Software Engineering Repository Data set.

A new fuzzy rule based algorithm for estimating software faults in early phase of development

“…Incidentally very few works are available in this area (Jiang et al (2007); Kim et al (2013); Mohan et al (2011);Yadav et al 2012Yadav et al , 2014Mohanta et al 2010;Cheung et al 2008;Pandey andGoyal 2009, 2010;Kumar and Misra 2008;Smidts et al 1998;Tripathi and Mall 2005;Fenton et al 2008;Xie et al 1999) compared to the SRGMs based on failure data gathered during testing phase. Among these models some are based on fuzzy logic (Yadav et al 2012(Yadav et al , 2014Pandey andGoyal 2009, 2010;Kumar and Misra 2008).…”

“…Yadav et al (2014) have proposed a fuzzy logic based model for predicting software defect density indicator at requirement, design and coding phase using reliability relevant software metrics. Pandey and Goyal (2010), have mentioned about fuzzy profiles of the RRSMs. They have considered two types of profiles: linear and logarithmic.…”

“…Among these models some are based on fuzzy logic (Yadav et al 2012(Yadav et al , 2014Pandey andGoyal 2009, 2010;Kumar and Misra 2008). Yadav et al (2012) have proposed a model to predict the number of residual defects using fuzzy logic on RRSMs.…”

“…Kumar and Misra (2008) have identified the phases in which corrective actions are needed to be performed to achieve the required target reliability. In all above mentioned fuzzy logic based model (Yadav et al 2012(Yadav et al , 2014Pandey andGoyal 2009, 2010;Kumar and Misra 2008), the fuzzy rule base has been developed by the domain experts. Generally, total number of fuzzy rules becomes very large when number of software metrics increases in each phase of software development.…”

“…As a consequence, there is a need of methods for assessing, achieving and prediction of safety and reliability of software in different sectors like industries, research and development, academics, etc. Software reliability is defined by IEEE as "the probability of a software system or component to perform its intended function under the specified operating conditions over the specified period of time" (Pandey and Goyal 2010). Software reliability study mainly concentrates on prediction of residual faults, estimation of failure intensity, reliability, development cost, release time, etc.…”