A COMPARATIVE EVALUATION ON THE PERFORMANCE OF FINITE FAILURE NHPP SOFTWARE RELIABILITY MODEL BASED ON GAMMA FAMILY DISTRIBUTION

In this study, the performance of software reliability is evaluated by applying the Gamma family distributions (Erlang, Log-Logistic) to the finite-fault NHPP reliability model. Also, the Gamma family distribution models were compared with the GoelOkumoto basic model to verify reliability performance. For this, software failure time data was used, the parameters were estimated by the maximum likelihood estimation method (MLE), and nonlinear equations were calculated using the bisection method. As a result, in the analysis of the intensity function, the Log-Logistic model was efficient because the failure occurring rate decreases significantly as the failure time passed and the mean square error (MSE) is also small. In the analysis of the mean value function, all the proposed models showed a slightly underestimated value compared to the true value, but the Log-Logistic model showed the smallest error to the true value together with the Goel-Okumoto model. As a result of evaluating the software reliability after putting the mission time in the future, the Erlang model showed a higher reliability trend than the other models in which the reliability decreases with the mission time. In conclusion, it can be seen that the Log-Logistic model has the best performance among the proposed models. In this study, the reliability performance of the Gamma family distribution model without the existing research case was newly analyzed, and it is expected that software developers will be able to use it as basic information to search for the optimal reliability model.