Modeling and Analysis of Software Fault Detection and Correction Process by Considering Time Dependency

Singapore Nat. Univ., Singapore
IEEE Transactions on Reliability (Impact Factor: 1.66). 01/2008; DOI: 10.1109/TR.2007.909760
Source: IEEE Xplore

ABSTRACT Software reliability modeling & estimation plays a critical role in software development, particularly during the software testing stage. Although there are many research papers on this subject, few of them address the realistic time delays between fault detection and fault correction processes. This paper investigates an approach to incorporate the time dependencies between the fault detection, and fault correction processes, focusing on the parameter estimations of the combined model. Maximum likelihood estimates of combined models are derived from an explicit likelihood formula under various time delay assumptions. Various characteristics of the combined model, like the predictive capability, are also analyzed, and compared with the traditional least squares estimation method. Furthermore, we study a direct, useful application of the proposed model & estimation method to the classical optimal release time problem faced by software decision makers. The results illustrate the effect of time delay on the optimal release policy, and the overall software development cost.

  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: A dynamic system such as a mobile telecommunication system often still have bugs when used in the field. Test engineers hence need to have a good amount of confidence that its software components have attained a certain level of reliability before such a system is released to a customer for operational uses. Software Reliability Growth Models (SRGMs) are useful for estimating the reliability of a software component for quality control and testing purposes. However, due to the availability of a huge number of SRGMs based on the Non-Homogeneous Poisson Process (NHPP), it is very difficult to know which one is the most suitable for a certain software system. The traditional model selection methods conspicuously omit a well defined and verified mechanism to identify suitable NHPP-based SRGMs which are capable of addressing the requirements necessary for testing and debugging a software component of a dynamic system. In this paper, we present a method for selecting the most suitable software reliability model to estimate the reliability of a software component in a dynamic system. Our method is based on model selection criteria and the use of a unification scheme. A software component in the Wireless Network Switching Centre (WNSC) is used as a case study to illustrate the usefulness of our method in a dynamic system.
    Software Engineering Conference (ASWEC), 2013 22nd Australian; 01/2013
  • [Show abstract] [Hide abstract]
    ABSTRACT: In the modern society, software plays a very important part in many security-critical or mission-critical systems. Consequently, the main goal of project managers and software engineers is to develop and deliver reliable software within very limited resource, time, and budget. In the past, some research reports showed that the Weibull distribution (WD) and the Pareto distribution (PD) models can be used to describe the distribution of software faults. In this paper, based on our previous study, we further propose and show how the two-parameter generalized Pareto distribution (2-GPD) can be used to model the distribution of software faults. Some mathematical properties of proposed model are analyzed and presented. Experiments based on open source software (OSS) are performed and discussed in detail. Evaluation results show that the proposed 2-GPD model eliminates some issues in modeling that arise in the PD model and has a fairly accurate prediction capability of fault distributions of OSS and depicts the real-life situation more faithfully.
    Software Security and Reliability (SERE), 2012 IEEE Sixth International Conference on; 01/2012
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The paper focuses on creating of a software reliability model based on phase type distribution. Usually, the length of intervals between the moments of fault detection and correction have unknown distributions. In this paper, a new approach how to approximate any distribution of positive random variable by mixture and convolution of exponential phases, known as the general type of phase-type distribution, is proposed. The optimization algorithm of Local Unimodal Sampling (LUS) is applied to estimate parameters of phase-type distribution. After such procedure, the dynamics of a software reliability model can be described by a continuous time absorbing Markov chain. The probabilities of the resulting absorbing Markov chain are used to compute performance measures of the software reliability model.
    ICSEA 2014 : The Ninth International Conference on Software Engineering Advances; 10/2014

Min Xie