Proportional Intensity-Based Software Reliability Modeling with Time-Dependent Metrics

Rinsaka, Koichiro; Shibata, Kazuya

doi:10.1109/compsac.2006.68

Cited by 16 publications

(20 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The proportional intensity-based SRMs in [33] are direct applications of the NHPP-based SRMs, by incorporating the multiple testing-effort parameters. Suppose that l (≥ 1) kinds of software metrics data x k = (x k1 , · · · , x kl ) T (k = 1, 2, · · · ) are available at each testing time t k (= 0, 1, 2, · · · ).…”

Section: Proportional Intensity-based Software Reliability Modelingmentioning

confidence: 99%

“…Since our PIM-based SRMs can involve the usual NHPP-based SRMs as special cases, the software reliability modeling and measurement based on PIM is considered as a generalized approach. In fact, it has been shown numerically that PIM-based SRMs could outperform the existing NHPP-based SRMs in almost all cases [33].…”

Section: Introductionmentioning

confidence: 98%

“…Recently, Rinsaka et al [33] developed a proportional intensity-based SRM with both software fault data and testing metrics data. The proportional intensity model (PIM) was proposed by Lawless [18] as a natural extension of the usual PHM in a different context from software reliability assessment, where the proportional hazard function in the Cox regression is directly used for the intensity function of an NHPP.…”

Section: Introductionmentioning

confidence: 99%

“…The most different point from the existing PHM is that PIM is a stochastic counting process to describe the time-series discrete events. That is, we introduced the Cox regression in [33] to incorporate the timedependent development/testing metrics instead of the linear and non-linear regressions with white noise or the logistic regression, and used the stochastic counting process to describe the cumulative number of faults detected in the software testing. It is worth mentioning that our PIM-based SRMs are quite different from the static regression models in [15,16,17,35,36].…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

PISRAT: Proportional Intensity-Based Software Reliability Assessment Tool

Shibata¹,

Rinsaka²,

Dohi³

2007

13th Pacific Rim International Symposium on Dependable Computing (PRDC 2007)

Self Cite

View full text Add to dashboard Cite

In this paper we develop a software reliability assessment tool, called PISRAT: Proportional Intensity-based Software Reliability Assessment Tool, by using several testing metrics data as well as software fault data observed in the testing phase. The fundamental idea is to use the proportional intensity-based software reliability models proposed by the same authors. PISRAT is written in Java language with 54 classes and 8.0 KLOC, where JDK1.5.0 9 and JFreeChart are used as the development kit and the chart library, respectively. This tool can support (i) the parameter estimation of software reliability models via the method of maximum likelihood, (ii) the goodness-of-fit test under several optimization criteria, (iii) the assessment of quantitative software reliability and prediction performance. To our best knowledge, PISRAT is the first freeware for dynamic software reliability modeling and measurement with timedependent testing metrics.

show abstract

Section: Proportional Intensity-based Software Reliability Modelingmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 98%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

PISRAT: Proportional Intensity-Based Software Reliability Assessment Tool

Shibata¹,

Rinsaka²,

Dohi³

2007

13th Pacific Rim International Symposium on Dependable Computing (PRDC 2007)

Self Cite

View full text Add to dashboard Cite

show abstract

“…According to Rinsaka et al (2006), the classical and perhaps the most important software reliability model is the non-homogeneous Poisson process (NHPP) model, which characterizes the number of faults detected in the software testing phase and can be used to predict software reliability. The traditional approach in this venue of software reliability has been focused on studying the stochastic process itself.…”

Section: Software Reliabilitymentioning

confidence: 99%

Introduction of first passage time (FPT) analysis for software reliability and network security

Krings

Millar

2009

Proceedings of the 5th Annual Workshop on Cyber Security and Information Intelligence Research: Cyber Security and Information

View full text Add to dashboard Cite

The study of the First Passage Time (FPT) problem (also known as first passage problem, FPP) started more than a century ago, but its diverse applications in science and engineering mostly emerged in the last two to three decades. Assuming that X(t) is a one-dimensional stochastic process, the First Passage Time is defined as the time (T) when X(t) first crosses a threshold. Engineering reliability is obviously a suitable application domain, and indeed applications such as optimal dam design in hydrology and analysis of structural failure in civil and mechanical engineering are typical examples. Although we envision that the FPT problem has great potential in network and software reliability, it should be more useful for network security and survivability because the approaches developed for the FPT problem are mostly analytical. The assumption for this inference is that in reliability analysis, experimental or historical data are often more readily available, which makes statistical approaches such as survival analysis more convenient and likely more realistic. In contrast, data is generally more difficult to obtain in security and survivability analyses, and analytical approaches can be leveraged to play more important roles. Furthermore, security and survivability often have to deal with malicious actions that may be driven by sophisticated cognition and behavioral processes, which are highly variable over time and very difficult to detect with short term data. If the behavior of an intruder can be characterized with some stochastic process such as Brownian motion, then the FPT approach may be applied to find the closedform solution of the probability density function (PDF) of the first passage time, which can be the time when the system breaks down or when the hacker is successful in compromising a network. In addition, the solutions to FPT depend on boundary and initial conditions of the corresponding partial differential equations, and they also describe the evolution of PDF over time. This may suggest that it is possible to model the behavior changes of an intruder over time and circumstances. Another advantage of FPT analysis is that it may help solve some non-Markov stochastic process problems in reliability analysis and survival analysis. In this article, we first briefly introduce the FPT problem with Brownian motion as an example, and then suggest its potential applications in software reliability and network security.

show abstract