Research And Application Of Software Reliability Model Based On NHPP

As software and computers are widely utilized in various systems including civil and military equipments, software faults have become the major factor that causes critical problems of computer systems. Hence, there exists an increasing demand of developing reliable software products. Software testing is an important means to ensure the software reliability in the software development process. However, when to stop the testing, and minimize the cost of development and maximize the reliability of software is one of the key researches of the software reliability, which is called the optimal software release time of decision-making problem.The traditional research on optimal software release time problem focuses on the software reliability model and testing cost model and make sure the model fits the practical more accurately. So far, the software reliability growth model exist more than one hundred. However, many proposed models are based on some unrealistic or over-simplified assumptions (e.g. perfect debugging, independent software faults, etc.),which to a large extent limits their applicability to real-1ife situations as well as their reliability estimation or prediction accuracy. In this paper, by studying the exist models, we do research on the factors which is affect the software reliability, and propose improved models and software cost models based on the improved assumption conditions. The main research contents of this thesis are summarized as follows:(1)We analyze the relationship between the software testing and the software reliability. Based on the exist models which is incorporate the testing coverage, considering the learn phenomena in the testing process and the software fault detection rate inherent, we proposed an improved model based on the testing coverage. And also, we have a fault detected function considering the testing coverage.(2)There may be imperfect debugging and repair delay existing in the software debugging process. In this case, we propose a model that incorporates a varying fault removed fate and a varying fault introduced fate, and then improve the model further considering the delay factor. Experimental results show that the proposed model fits the failure data quite well and has a fairly accurate prediction capability which is compared with other models. For the improved model, we attempt to extend the proposed model for two-dimensional model.(3)Based on the proposed models, we proposed software cost models which considering the testing cost, software maintenance cost and the risk cost caused by the software failure. We also discuss the optimal software release time.