On Change-Point Detection and Its Application to Software Reliability Measurement Based on Hazard Rate Models  
Author Shiho Hayashida

 

Co-Author(s) Shinji Inoue; Shigeru Yamada

 

Abstract A software hazard rate model is known as one of the important and useful mathematical models for describing the software failure-occurrence phenomenon and conducting quantitative software reliability assessment. This paper discusses more plausible software hazard rate modeling for software reliability assessment in an actual testing-phase. That is, we discuss software hazard rate modeling with effect of changing the testing-environment at change-point, which influences the accuracy of the software reliability assessment and is often observed in an actual testing-phase. Especially in this paper, we discuss change-point detection methods by using Laplace and arithmetic mean tests for more practical application method of our change-point model. Finally, we show numerical examples for our model and results of model comparisons by using actual data.

 

Keywords Hazard rate model, Software reliability growth model, Change-point detection, Testing-environmental function, Laplace trend test, Arithmetic mean test, Software reliability assessment
   
    Article #:  1983
 
Proceedings of the 19th ISSAT International Conference on Reliability and Quality in Design
August 5-7, 2013 - Honolulu, Hawaii, U.S.A.