원문정보
A Study on the Characteristics of Software Reliability Model Using Exponential-Exponential Life Distribution
초록
영어
In this paper, we applied the shape parameters of the exponentialized exponential life distribution widely used in the field of software reliability, and compared the reliability properties of the software using the non-homogeneous Poisson process in finite failure. In addition, the average value function is also a non-decreasing form. In the case of the larger the shape parameter, the smaller the estimated error in predicting the predicted value in comparison with the true value, so it can be regarded as an efficient model in terms of relative accuracy. Also, in the larger the shape parameter, the larger the estimated value of the coefficient of determination, which can be regarded as an efficient model in terms of suitability. So. the larger the shape parameter model can be regarded as an efficient model in terms of goodness-of-fit. In the form of the reliability function, it gradually appears as a non-increasing pattern and the higher the shape parameter, the lower it is as the mission time elapses. Through this study, software operators can use the pattern of mean square error, mean value, and hazard function as a basic guideline for exploring software failures.
목차
1. 서론
2. 유한고장 NHPP 소프트웨어 신뢰성 모형
2.1 지수화-지수분포
2.2 수명분포가 지수화-지수분포를 따르는 NHPP 소프트웨어 신뢰모형
3. 수명분포가 지수화-지수분포를 따르는 소프트웨어 고장시간 신뢰도 분석
4. 결론
References