The Research On NHPP Software Reliability Measurement

The rapid development of computer technology led to software crisis,which attracted more attention on software quality, and then there came software reliability engineering and software reliability measure of technology. Software reliability growth model is a method of technology on software reliability metrics;it is also the basis of software reliability metrics.In this paper, the method of software reliability metrics in system testing and field trial operation is studied. Nonhomogeneous Poisson Process (NonhomogeneousPoisson Process,referred as NHPP) software reliability growth model is one of the most important models in software reliability metrics,and it becomes an important tool in the practice of software reliability engineering.So far, the most suitable and simple evaluation model for software reliability is still the NHPP. The method in this paper is based on the NHPP software reliability growth model theory.There are two aspects studied in this paper:firstly, software reliability metric parameter plays an important role in the research of the reliability metrics.Fault removal efficiency, an important parameter in the software reliability metrics,is mainly studied.We analyzed the relationship among software fault exposure rate,the number of detection and the fault removal efficiency in one of the software testing.A component-based software fault exposure rate method based on triangular fuzzy number arithmetic operations is given. Secondly, based on literature,a NHPP software reliability growth model considering correlation of software detects and software fault removal efficiency is framed, then a test coverage function is introduced,at last another NHPP software reliability growth model considering test coverage function and fault removal efficiency is proposed.