No Detail Found!

Comparative analysis on the reliability attributes of finite failure nhpp software reliability model with exponential distribution characteristics

Author(s)
SEUNG KYU PARK Department of Electronic Engineering, Namseoul University, 91 Daehak-ro, Seonghwan-eup, 31020, Seobuk-gu, Cheonan-si, Chungnam, Korea
Abstract
In this study, the reliability attributes of the finite failure NHPP software reliability model with exponential distribution (Exponential Basic, Inverse Exponential, Lindley, Rayleigh) characteristics were comparatively analyzed, and based on this, the optimal reliability model was also presented. To analyze the software failure phenomenon, the failure time data collected during system operation was used, and the parameter estimation was solved by applying the maximum likelihood estimation method (MLE). As a result, first, in the analysis of mean square error (MSE), the Lindley model was effective because it had the smallest error value. Second, in the analysis of the true value estimation power of the mean value function, all of the proposed models showed an overestimated pattern, but it was found that the Lindley model was excellent because the width of the error was the smallest. Third, in the evaluation of the strength function, the Lindley model and the Rayleigh model were effective in terms of fit as the failure rate increased and then decreased significantly as the failure time passed. Fourth, as a result of evaluating the reliability by applying the mission time, the Rayleigh model appeared to be the highest and most stable, but the Exponential Basic model showed the largest decreasing trend and was found to be inefficient. In conclusion, it was found that the Lindley model is an efficient model with the best performance. Through this study, the reliability attributes of the distribution with the characteristic of the exponential form, which has no existing research case, were newly identified, and through this, basic design data that software developers could use in the initial stage could be presented.
Publication Details
Page(s) 5220-5231
DOI DOI not available
Published Journal: Journal of Theoretical and Applied Information Technology, Volume: 100, Issue: 18, Year: 2022

Request the full text from PASTIC

Keywords
Lindley, Reliability Performance, Inverse exponential, Rayleigh, NHPP, Exponential Basic
0

Citations

0

Downloads

38

Views