A New Homogeneous Pure Birth Process based Software Reliability Model
Software Reliability models has been developed for decades. The majority of them are based on non homogeneous Poisson processes, where the failure rate is a non linear function of time. They are also well described by pure birth processes what leads to non homogeneous continuous time Markov chains (...
Guardado en:
| Autor principal: | |
|---|---|
| Formato: | Objeto de conferencia Resumen |
| Lenguaje: | Inglés |
| Publicado: |
2016
|
| Materias: | |
| Acceso en línea: | http://sedici.unlp.edu.ar/handle/10915/57254 http://45jaiio.sadio.org.ar/sites/default/files/asse-16.pdf |
| Aporte de: |
| Sumario: | Software Reliability models has been developed for decades. The majority of them are based on non homogeneous Poisson processes, where the failure rate is a non linear function of time. They are also well described by pure birth processes what leads to non homogeneous continuous time Markov chains (NHCTMC), as it is usually used in the simulation of the stochastic software failure process. We propose in this work a different and novel approach. We consider a failure rate that does not depend on time but depends non linearly on the number of failures λr(t) = λr. We use the parametric Empirical Bayes framework in order to estimate λr. |
|---|