A birth-process approach to Moranda's geometric software-reliability model

被引:19
作者
Boland, PJ [1 ]
Singh, H
机构
[1] Natl Univ Ireland Univ Coll Dublin, Dept Stat, Dublin 4, Ireland
[2] W Virginia Univ, Dept Stat, Morgantown, WV 26506 USA
[3] Panjab Univ, Chandigarh 160014, India
基金
美国国家航空航天局;
关键词
birth process; mean and intensity function of a counting process; mission time; optimal release time; software failure; software reliability;
D O I
10.1109/TR.2003.813166
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
To alleviate some of the objections to the basic Jelinski Moranda (JM) model for software failures, Moranda [14] proposed a geometric de-eutrophication model. This model assumes that the times between failures are statistically-independent exponential random variables with given failure rates. In this model the failure rates decrease geometrically with the detection of a fault. Using an intuitive approach, Musa, Iannino, Okumoto [15], see also Farr [5], derived expressions for the mean and the intensity functions of the process N(t) which counts the number of faults detected in the time interval [0, t] for the Moranda geometric de-eutrophication model. N(t) is studied as a pure birth stochastic process; its probability generating function is derived, as well as its mean, intensity and reliability functions. The expressions for the mean and intensity functions derived by MIO are only approximations and can be quite different from the true functions for certain choices of the failure rates. The exact expressions for the mean function and the intensity function of N(t) are used to find the optimum release time of software based on a cost structure for Moranda's geometric de-eutrophication model.
引用
收藏
页码:168 / 174
页数:7
相关论文
共 27 条
[1]  
Boland P. J., 2001, INT J RELIAB APPL, V2, P147
[2]   A generalised shock model for software reliability [J].
Chatterjee, S ;
Misra, RB ;
Alam, SS .
COMPUTERS & ELECTRICAL ENGINEERING, 1998, 24 (05) :363-368
[3]   Unification of software reliability models by self-exciting point processes [J].
Chen, YP ;
Singpurwalla, ND .
ADVANCES IN APPLIED PROBABILITY, 1997, 29 (02) :337-352
[4]   WHEN SHOULD ONE STOP TESTING SOFTWARE [J].
DALAL, SR ;
MALLOWS, CL .
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1988, 83 (403) :872-879
[5]  
FARR W, 1996, HDB SOFTWARE RELIABI, pCH3
[6]  
GAUDOIN O, 1992, IEEE T RELIAB, V41, P518, DOI 10.1109/24.249578
[7]  
GAUDOIN O, 1994, IEEE T RELIAB, V43, P536, DOI 10.1109/24.370229
[8]   TIME-DEPENDENT ERROR-DETECTION RATE MODEL FOR SOFTWARE RELIABILITY AND OTHER PERFORMANCE-MEASURES [J].
GOEL, AL ;
OKUMOTO, K .
IEEE TRANSACTIONS ON RELIABILITY, 1979, 28 (03) :206-211
[9]  
Jelinski Z., 1972, Statistical Computer Performance Evaluation, P465, DOI DOI 10.1016/B978-0-12-266950-7.50028-1
[10]  
Karlin S, 1975, A First Course in Stochastic Process