A Statistical Testing Approach for Quantifying Software Reliability; Application to an Example System

[摘要] The U.S. Nuclear Regulatory Commission (NRC) encourages the use of probabilistic risk assessment (PRA) technology in all regulatory matters, to the extent supported by the state-of-the-art in PRA methods and data. Although much has been accomplished in the area of risk-informed regulation, risk assessment for digital systems has not been fully developed. The NRC established a plan1 for research on digital systems to identify and develop methods, analytical tools, and regulatory guidance for (1) including models of digital systems in the PRA?s of nuclear power plants (NPPs), and, (2) incorporating digital systems in the NRC?s risk-informed licensing and oversight activities. Under NRC?s sponsorship, Brookhaven National Laboratory (BNL) explored approaches for addressing the failures of digital instrumentation and control (I and C) systems in the current NPP PRA framework. Specific areas investigated included PRA modeling digital hardware2, development of a philosophical basis for defining software failure3, and identification of desirable attributes of quantitative software reliability methods4 7044. Based on the earlier research, statistical testing is considered a promising method for quantifying software reliability.

[发布日期] 2016-12-30

[学科分类] 工程和技术（综合）

[关键词] Statistical Testing Approach