Advances and Applications in Statistics
Volume 50, Issue 4, Pages 315 - 327
(April 2017) http://dx.doi.org/10.17654/AS050040315 |
|
SOFTWARE RELIABILITY MODEL WHEN A RANDOM NUMBER OF FAULTS OCCURRING AT A TIME WITH CORRECTION PROCESS
K. Harishchandra and Balakrishna
|
Abstract: Most of the software reliability models (SRMs) in the literature focus on fault detection process and assume that only a single fault occurs at a time and the fault correction process is ignored assuming that the detected fault is removed immediately. In this paper, we propose a software reliability model when a random number of faults occur at a given instant of time point taking into consideration a fault correction process. The study emphasizes on the development of a model based on the two state Markov process (namely, detection state (0) and correction state (1)). In the model development, both fault detection time and fault correction time are assumed to be exponential, while the size of the faults is random having a geometric distribution. Maximum likelihood estimators and Bayesian estimators of the model parameters are obtained. Numerical studies are shown. |
Keywords and phrases: software reliability, fault detection process, fault correction process, random size faults, maximum likelihood estimation, Bayesian estimation, two states Markov process. |
|
Number of Downloads: 436 | Number of Views: 1332 |
|