基于马尔科夫链的软件故障分类预测模型

Software fault classification prediction model based on Markov chain

传统的软件可靠性模型一般未考虑故障的危害严重程度对软件失效的影响.然而在很多研究(如软件可信性研究)中,不仅要考虑软件的失效率、故障总数与失效间隔时间等,同时也要考虑故障的危害严重程度对软件可信性的影响.为解决上述问题,提出一种基于马尔科夫链的预测方法,用以预测软件未来发生故障的种类,即预测软件未来发生故障的危害严重程度.这种分类预测方法可以更加全面地描述由软件故障引起的失效问题.

Traditional software reliability models did not consider the severity of the fault affected the reliability of software.However,in many occasions,such as software trustworthy study,we must consider not only the software failure rate,the total number of faults,and failure interval time,but also the severity of the impact on software reliability.For solving the above problem,we propose a prediction method based on Markov chain,which can be used to predict the software failure types,that is,to predict software failure severity occurring in the future.This classification prediction method can fully describe failures caused by software failure.