Fault samples simulation based on Monte Carlo method in testability virtual test

Fault injection is a major method applied in the testability verification, so the fault samples selection is a key factor to the confidence level of evaluation conclusion. There are some deficiencies in the exiting fault samples selection methods because of not considering the fault effect and fault rate varying. Aiming at the shortage of the existing fault samples selection method, a fault samples generation method based on Monte Carlo method is proposed respectively to the corrective maintenance and regular maintenance under the background of testability virtual test in this paper. The Multisim 10 is utilized to build the fault models and simulate the process of fault injection. The results show that this method can simulate the fault samples effectively that can be injected in the testability virtual test.