The SRGM Framework of Integrated Fault Detection Process and Correction Process

Abstract

In this paper, the hypothesis that the detected faults will be immediately removed is revised. An integration of fault detection process and correction process is considered. According to the statistic of the faults, the two new frameworks, which are the SRGM framework including repeated faults (CRDW) and the SRGM framework excluding repeated faults (CNRDW), are presented. The above two frameworks, not only can predict the number of cumulative detected faults, but also can predict the number of corrected faults. In this paper, as an example, two reliability models are gained from CNRDW with different detection process and different correction process. The fitting capability and prediction capability of the two models are evaluated by an open software failure data set. The experimental results show that the presented models have a fairly accurate fitting capability and prediction capability compared with other software reliability growth models.

A Generic Framework For Capturing Reliability in Cyber-Physical Systems

“…The software reliability growth model (SRGM) is an important model to measure software reliability. The most general SRGM model is the Non-Homogeneous Poisson Process (NHPP) SRGM with mean value function [10]. The base model for SRGM is derived from G-O model [11] and modifies only hypothesis of G-O model.…”