In view of the problems that the existing testability verification schemes usually select the principle of minimum sample size, which causes that the same sample size and fault criteria are obtained under the same risk and requirements without considering different structural and functional complexity of different equipment systems in engineering, a restricted fault sample size determination method based on posterior distribution is proposed. According to the structure of equipment systems, the hierarchical Bayesian network testability verification model is constructed, and the priori information contained in each structure level is used to derive the posterior distribution of fault detection rate. Based on the posterior distribution sample set, firstly the sample size determination method under the principle of minimum sample size is given, and then the proportional stratified sample size allocation method is combined with the functional characteristics of equipment in order to give a restricted fault sample size determination method, finally the method is validated by an example. The results show that on the one hand the method can fully utilize the equipment structure information, and can effectively reduce the sample size under the same index constraint compared with the classical verification scheme and the traditional Bayes scheme, on the other hand the functional characteristics of the system can be considered to ensure the accuracy and rationality of the testability verification conclusion.