• The Transactions of the Korea Information Processing Society
• /
• v.7 no.7
• /
• pp.2076-2085
• /
• 2000

Almost all existing software reliability models are based on the assumptions of he software usage and software failure process. There, therefore, is no universally applicable software reliability model. To develop a universal software reliability model this paper suggests the predictive filter as a general software reliability prediction model for time domain failure data. Its usefulness is empirically verified by analyzing the failure datasets obtained from 14 different software projects. Based on the average relative prediction error, the suggested predictive filter is compared with other well-known neural network models and statistical software reliability growth models. Experimental results show that the predictive filter generally results in a simple model and adapts well across different software projects.

• Communications for Statistical Applications and Methods
• /
• v.17 no.6
• /
• pp.909-916
• /
• 2010

A trend in software reliability engineering is to take into account the coverage growth behavior during testing. A coverage growth function that represents the coverage growth behavior is an essential factor in software reliability models. When multiple competitive coverage growth functions are available, there is a need for a criterion to select the best coverage growth functions. This paper proposes a selection criterion based on the prediction error. The conditional coverage growth function is introduced for predicting future coverage growth. Then the sum of the squares of the prediction error is defined and used for selecting the best coverage growth function.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.10 no.3
• /
• pp.1-9
• /
• 2014

The inverse Rayleigh model distribution and Rayleigh distribution model were widely used in the field of reliability station. In this paper applied using the finite failure NHPP models in order to growth model. In other words, a large change in the course of the software is modified, and the occurrence of defects is almost inevitable reality. Finite failure NHPP software reliability models can have, in the literature, exhibit either constant, monotonic increasing or monotonic decreasing failure occurrence rates per fault. In this paper, proposes the inverse Rayleigh and Rayleigh software reliability growth model, which made out efficiency application for software reliability. Algorithm to estimate the parameters used to maximum likelihood estimator and bisection method, model selection based on mean square error (MSE) and coefficient of determination($R^2$), for the sake of efficient model, were employed. In order to insurance for the reliability of data, Laplace trend test was employed. In many aspects, Rayleigh distribution model is more efficient than the reverse-Rayleigh distribution model was proved. From this paper, software developers have to consider the growth model by prior knowledge of the software to identify failure modes which can helped.

• Communications for Statistical Applications and Methods
• /
• v.18 no.5
• /
• pp.667-674
• /
• 2011

A recent trend in software reliability engineering accounts for the coverage growth behavior during testing. The coverage growth function (representing the coverage growth behavior) has become an essential component of software reliability models. Application of a coverage growth function requires the estimation of the coverage growth function. This paper considers the problem of estimating the coverage growth function. The existing maximum likelihood method is reviewed and corrected. A method of minimizing the sum of squares of the standardized prediction error is proposed for situations where the maximum likelihood method is not applicable.

• International Journal of Reliability and Applications
• /
• v.6 no.2
• /
• pp.87-99
• /
• 2005

Coverage-based software reliability growth models (SRGMs) have been developed and successfully applied in practice. Performance of a coverage-based SRG M depends on the coverage function employed by the SRGM. When the coverage function represents the coverage growth behavior well irrespective of type of the testing profile the corresponding coverage-based SRGM is expected to be widely applicable. This paper first conducts a study of selecting the most representative coverage functions among the available coverage functions. Then their performances are empirically evaluated and compared. The result provides a foundation for developing widely applicable coverage-based SRGMs and monitoring the progress of a testing process.

• Journal of Applied Reliability
• /
• v.14 no.2
• /
• pp.129-136
• /
• 2014

Reliability test is focusing to detect the unexpected reliability failure and solve them for the high quality of products. The test data should be used to assess and project the current level of interesting product reliability and so it is very important to have the accurately assessing methodology with test data. There are two type of trend for test data as constant and changing one during testing and this paper shows the difference in the assessing results of these two cases. There is less information how to define the existence of reliability growth rate changing and calculate the parameters of the reliability growth models to make an accurate assessment with such condition, so i established the process and mathematical model to calculate the parameters at such condition to make reliability growth curve with high Goodness of Fit. I validated the new method with the data made from Monte Carlo Simulation and case from Demko (1993). Even the assessed result with the new methodology may be different with the case by case because of very diversity in test condition and testing product quality, but the process and method founded in this research can be applied to any case using Duane and AMSAA model for their test data assessment. I also present the evaluation method to see the effectiveness with new one which is a conventional knowledge and not popular to use, so it is possible to compare the results with the newly presented and conventional method for better business decision.

• The Transactions of the Korea Information Processing Society
• /
• v.7 no.3
• /
• pp.852-861
• /
• 2000

Software reliability growth models are used in testing stages of software development to model the error content and time intervals betwewn software failures. In this paper, using priors for the number of fault with the negative binomial distribution nd the error rate with gamma distribution, Bayesian inference and model selection method for Jelinski-Moranda and Goel-Okumoto and Schick-Wolverton models in software reliability. For model selection, we explored the sum of the relative error, Braun statistic and median variation. In Bayesian computation process, we could avoid the multiple integration by the use of Gibbs sampling, which is a kind of Markov Chain Monte Carolo method to compute the posterior distribution. Using simulated data, Bayesian inference and model selection is studied.

• Journal of The Korean Association of Information Education
• /
• v.2 no.1
• /
• pp.35-51
• /
• 1998

Software reliability growth model is one of the evaluation methods, software quality which quantitatively calculates the software reliability based on the number of errors detected. For correct and precise evaluation of reliability of certain software, the reliability model, which is considered to fit dose to real data should be selected as well. In this paper, the optimal model for specific test data was selected one of among five software reliability growth models based on NHPP(Non Homogeneous Poission Process), and in result reliability estimating scales(total expected number of errors, error detection rate, expected number of errors remaining in the software, reliability etc) could obtained. According to reliability estimating scales obtained, Software development and predicting optimal release point and finally in conducting systematic project management.

• Journal of KIISE
• /
• v.42 no.12
• /
• pp.1551-1560
• /
• 2015

Software Reliability Growth Models (SRGMs) are used to quantitatively evaluate software reliability and to determine the software release date or additional testing efforts using software failure data. Because a single SRGM is not universally applicable to all kinds of software, the selection of an optimal SRGM suitable to a specific case has been an important issue. The existing methods for SRGM selection assess the goodness-of-fit of the SRGM in terms of the collected failure data but do not consider the accuracy of future failure predictions. In this paper, we propose a method for selecting SRGMs using the trend of failure data and failure prediction ability. To justify our approach, we identify problems associated with the existing SRGM selection methods through experiments and show that our method for selecting SRGMs is superior to the existing methods with respect to the accuracy of future failure prediction.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.2
• /
• pp.329-335
• /
• 2019

Reliability growth analysis was conducted for a guided weapons system. In the development phase, reliability management activities were continuously carried out by identifying failure modes and causes and analyzing faults found during the testing. The missile system consists of an all-up-round missile and a launcher, and the analysis was carried out according to the test results of each system. The test results for the all-up-round missile were obtained with discrete data, which were success and failure as a one-shot-device. The test results for the launcher were obtained with continuous data by operating the equipment continuously in the test. For each test result, the reliability growth model was applied to the Standard Gompertz model and the Crow-Extended model. The models were used to identify the growth analysis results of the test so far. It was also possible to predict the reliability growth results by assuming the future test results. The study results could be useful in achieving the desired reliability goal and in determining the number of tests. Then, the planned test will be confirmed and the growth analysis of the missile system will continuously be conducted.