TY  - JOUR
T1  - The Analysis and Predict of Software Failure Time Based on Nonlinear Regression
AU - Yang, Tae-Jin 
JO  - Journal of Engineering and Applied Sciences
VL  - 13
IS  - 12
SP  - 4376
EP  - 4380
PY  - 2018
DA  - 2001/08/19
SN  - 1816-949x
DO  - jeasci.2018.4376.4380
UR  - https://makhillpublications.co/view-article.php?doi=jeasci.2018.4376.4380
KW  - time censoring
KW  -Software reliability
KW  -
KW  -nonlinear regression
KW  -model selection
KW  -determination
KW  -growth regression model
AB  - Software reliability is an important issue in the software development process. The software
development process with considerations of cost and failure time are essential. Software failure time have been
proposed in the literature exhibit either constant, monotonic increasing or monotonic decreasing. For data
analysis of software reliability model, trend analysis already been developed. The methods of trend analysis
are arithmetic mean test and Laplace trend test. Trend analysis only offers information of outline content. In
this study, we were discussed censoring failure time and predicted failure time using nonlinear regression
models that is growth, quadratic and S-curve type which error terms, each other are different model. Model
selection using the coefficient of determination and the mean square error were presented for effective
comparison. In result of analysis, relatively, growth regression model than any models in terms of goodness
of fit is effective model.
ER  -