Safety Critical Software Reliability Model Considering Multiple Influencing Factors

Abstract

Safety-critical software accounts for a large proportion of computer applications in high-tech fields such as aerospace, aerospace, and nuclear power. Once a failure occurs, it can seriously endanger human life, cause large-scale environmental damage, or cause huge economic losses. How to accurately evaluate the reliability of safety-critical software has become one of the problems that software reliability engineering needs to solve. Therefore, this paper takes safety-critical software as the research object. Based on the in-depth analysis of the characteristics of safety-critical software, combined with the characteristics of safety-critical software and the common model optimization and improvement ideas, the total number of software failures, the failure detection rate changes, and the troubleshooting rate is less than the correlation between faults and errors is analyzed. The fault detection and elimination process are analyzed. The influencing factors that should be considered in the reliability modeling process of safety critical software are summarized and mathematically described. A multi-influence factor is considered. Safety Critical Software Reliability Model (SCSRM). Based on the safety critical software failure data set, the comparison experiment between SCSRM and the other four software reliability models was completed. The experimental results show that SCSRM can obtain better data fitting and prediction effects in the reliability assessment scenarios of safety-critical software, and proves the validity and stability of the proposed model.