Comparison of model-based diagnosis methods: Testability modeling rodon

This paper provides a detailed analysis and summary of diagnosis model generation rules of Rodon. These rules are different from testability modeling theory. Two cases, a binary tree and a multi-branch tree, are used to compare and illustrate the differences. Dependency matrices and diagnosis trees of these cases are generated by using testability modeling and Rodon respectively. By comparing the diagnosis results, the similar and dissimilar rules are analyzed in detailed. In the binary tree case, the diagnosis model generation rules of Rodon is same as the rules of isolation tree generation rules of testability modeling. In the multi-branch tree case, Rodon can generate a multi-branch tree, but the testability modeling method cannot generate it. Although the testability modeling method has such shortage, it can generate a dependency matrix, which is a binary matrix with higher computational effectiveness and lower storage requirements. Based on comparison, several update suggestions for testability modeling are discussed.