Reliability analysis of software with three types of errors and imperfect debugging using Markov model

Abstract

In this analysis, we consider a Markovian software reliability model (SRM) with imperfect debugging (ID) wherein software may fail due to three types of errors called error generation (EG). For developing the governing differential equations of SRM with three types of errors, an irreducible Markov process is used. Further, we have applied the method namely 'Runge-Kutta' (R-K) under transient condition for obtaining reliability of concerned system in different configurations. Moreover, we suggest various software reliability indices like reliability, failure frequency and mean time to failure (MTTF). By taking a numerical illustration, the sensitivity analysis is done to demonstrate the validity of analytical outcomes and to explore the effects of various parameters. Moreover, the analytical results are also compared with adaptive neuro-fuzzy inference system (ANFIS). At last, conclusions are given.