Title :
Design of safety-critical software reliability demonstration test based on Bayesian theory
Author :
Yumei Wu ; Haifeng Li ; Yongli Yu
Author_Institution :
Sch. of Reliability & Syst. Eng., Beihang Univ., Beijing, China
Abstract :
The original software reliability demonstration test (SRDT) takes no consideration of prior knowledge and priori distribution adequately, which costs a lot of time and resource. A new improved Bayesian based SRDT method was proposed. First, a framework for SRDT scheme was constructed. According to the framework the decreasing function was employed to construct the priori distribution density functions for discrete and continuous safety-critical software respectively, and then discrete Bayesian software demonstration function (DBSDF) scheme and continuous Bayesian software demonstration function (CBSDF) scheme were presented. A set of comparative experiments have been carried out with the classic demonstration testing scheme on several published data sets. The experimental results reveal that both DBSDF scheme and CBSDF scheme are more efficient and applicable especially for the safety-critical software with high reliability requirements.
Keywords :
Bayes methods; safety-critical software; software reliability; Bayesian based SRDT method; Bayesian theory; CBSDF scheme; DBSDF scheme; continuous Bayesian software demonstration function; discrete Bayesian software demonstration function; distribution density functions; published data sets; safety-critical software reliability demonstration test; Distribution functions; Measurement; Probability density function; Safety; Software; Software reliability; Bayesian Theory; decreasing function; hierarchical priori distribution function; safety critical software; software reliability demonstration test (SRDT);
Conference_Titel :
Quality, Reliability, Risk, Maintenance, and Safety Engineering (QR2MSE), 2013 International Conference on
Conference_Location :
Chengdu
Print_ISBN :
978-1-4799-1014-4
DOI :
10.1109/QR2MSE.2013.6625734