2011, 37(2): 149-154,160.
Abstract:
The shape of the observed reliability growth curve depends strongly on the time distribution of the testing effort. Thus, the reliability estimation power of software reliability growth model (SRGM) can be improved by considering the testing effort function (TEF). Due to the integrated effects of software structure and learning factor on testing process, testing effort increasing rate may exhibit an S-shaped varying trend first increasing and then decreasing. To accurately describe this S-shaped varying trend, two S-shaped testing effort functions were proposed first, i.e. delayed S-shaped TEF and inflected S-shaped TEF, which were simple and flexible. Then two new SRGMs were put forward by combining two S-shaped TEFs into SRGM, i.e. DSTEF-SRGM and ISTEF-SRGM. Finally, for two real failure data-sets, the case study was done by comparing these two S-shaped TEF-SRGMs with several classical SRGMs and other representative SRGMs considering TEF in the form of estimation power. The result shows that compared with these comparison models, the proposed ISTEF-SRGM has the best fitting results on each data-set. In other words, the proposed ISTEF-SRGM provides novel applicability and a significantly better power of reliability estimation.