Sequential diagnostic strategy for the diagnosis of complex systems

For the hardness of diagnosing a complex system, the diagnosis often includes many cycles in order to get the enough diagnostic depth, and the sequential diagnostic strategy is used widely to enhance the efficiency of this process. In this paper, an on-line diagnostic algorithm is used in our expert system to make the sequential diagnostic decision, and the multi-signal dependency model is applied to get the relationship of faults and tests The MADM method based on the value function is introduced as the evaluation function in this algorithm for more generalization. Each test is evaluated synthetically with two factors, the information gain and cost. A method is proposed to calculate the attribute weights in MADM according to the practical constrained conditions.