The dynamic reliability evaluation method is applied to four model systems to study the following three problems: are the computing time and computation cost reasonable; is it possible to reduce the computing time; and how much error is caused by reducing the number of power system states. The results show that the computing time of the proposed method is reasonable in practice, and can be reduced appreciably by employing the methods discussed, and the calculation error is tolerable. To apply the proposed methods to actual power systems more effectively, the following problems should be studied further: establishment of a method to calculate the cascaded failure probability more efficiently; and use of the proposed method to actual power systems to gain experience.