Open AccessAn Empirical Software Reliability Growth Model for Identification of True Failures
Author(s) -
Jagadeesh Medapati,
Anand Chandulal Jasti,
Ranajikanth TV
Publication year - 2019
Publication title -
international journal of innovative technology and exploring engineering
Language(s) - English
Resource type - Journals
ISSN - 2278-3075
DOI - 10.35940/ijitee.i8756.0881019
Subject(s) - computer science , software development , software sizing , software construction , reliability engineering , software quality , verification and validation , software reliability testing , software quality analyst , software engineering , avionics software , package development process , software , software metric , software development process , engineering , operating system , operations management
Software Reliability is a special topic of software engineering that deals with the finding of glitches during the software development. Effective analysis of the reliability helps to understand the quality of the software. It also helps to reveal the number of failures occurred in development phase which facilitates refinement of the failures in the developed software’s. If the failures are not minimized the number of reviews in the software development process increases which in turn increase the expenditure to develop the software. Every software organization aims at releasing the software in time and also it becomes a mandate to manage the software such that the time to release the software is optimized. It becomes a mandate for any organization to release software patches so as to minimize the errors after software release and thereby if the number of patches increases, the credibility of the software together with the storage area will be at stake. This article presents a novel case study wherein a procedural layout is presented such that the number of failures can be reduced instantaneously and the failures are identified at the early stage. The development procedure laid in this article helps to formulate a basis for the distinction between true failures and non-failures. The work is presented using benchmark datasets and the results showcase a better recognition rate and failure deduction rate.