MARKOVIAN SOFTWARE RELIABILITY MODEL FOR TWO TYPES OF FAILURES WITH IMPERFECT DEBUGGING RATE AND GENERATION OF ERRORS

This investigation deals with a software reliability model based on Markov process. For formulating the model, we define a random variable representing the cumulative number of faultssuccessfully corrected upto a specified point of time. This model is based on the assumption that thereare two types of software failures. Further the concepts of imperfect debugging environment and error generation phenomenon are taken into consideration. Transient analysis based on Laplace transform and matrix approach has been done to find the solution of the system of differential differenceequations. Several performance indices for software reliability assessment are derived for this model. Numerical results with the help of Runge-Kutta Method show that the proposed framework incorporating both concepts of imperfect debugging phenomenon and error generation for two types of faults has a fairly accurate prediction capability.

Keywords:

Software reliability , Imperfect debugging , Error generation , Markov process , Software reliability growth