| In the age of Internet, our life, work and study are already can not leave without software, with the growing of dependence on software, the requirements of software reliability are also increasing.. Software failures may make a lot of damage economically,even to personal safety. It is very important that using the objective condition and more closer to the software running environment when accurately evaluate and predict the reliability of the software system. During the testing process the test engineer’s working efficiency was kept growing too. In order to get much closer to the real test process, we proposed a software reliability growth model framework based on working efficiency measurement, then we proposed software reliability growth model with work efficiency measurement. The main contribution of this article could be summarized as follow:(1)This article proposed the definitions and measurements of working efficiency, real working time and effective working time. Then give the mathematical expressions of the conceptions. Based on that, a working efficiency improvement model has been derived from some basic assumptions of real scenario.(2)This article imported the derived working efficiency improvement model into Non-Homogeneous Possion Process(NHPP) software reliability growth model(SRGM) and proposed a new Logistic-WE model and its two variants Logistic-WE-ID1 and Logistic-WE-ID2 under imperfect debugging circumstance.(3)This article designed and implemented an on-line reliability assessment tool integrating the proposed SRGMs.This article also validated the proposed methods by experiments under both perfect debugging circumstance imperfect debugging circumstance. The experiments ware performed on three classic software failure data sets, Ohba, Wood and Okumoto. The results showd that the proposed methods outperformed than other SRGMs, which implied the proposed could predict software reliability more precisely. |
PDF Full Text Request