.png){kind=icon}
.png){kind=icon}
Return
Performance Evaluation Method for Complex Systems Based on Outcome-Oriented Correlated Evidence Reasoning
.png){kind=icon}
PRE
AI
DOI：10.1109/TIE.2025.3629342.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Performance evaluation is one of the most important methods to guarantee the safe and reliable operation of complex systems. Currently, there are three main problems in the performance evaluation of complex systems: the evaluation model is difficult to establish; the correlation between indicators is difficult to measure; and expert knowledge is not effectively incorporated. To solve these problems, this article proposes a performance evaluation model considering indicator correlation on the basis of considering the cohesion ratio factor evidence reasoning rule (ERr-CRC). In the process of calculating indicator correlations, expert knowledge is first used to determine whether there is a correlation between two indicators, and then an outcome-oriented correlation calculation method is proposed, which transforms the problem of calculating the correlation between indicators into the problem of calculating the correlation between the confidence distributions of each outcome level. On this basis, the order of evidence fusion is determined according to the size of the correlation factor, and it is integrated into the process of assigning the basic probability quality of evidence in the form of a global discount factor for the final performance evaluation. To prove the effectiveness of the correlation factor, this article carries out a sensitivity analysis and a perturbation analysis. Finally, a case study of the ERr-CRC model is conducted using the rocket structure state-aware simulation platform.
Keywords:
Correlation
evidential reasoning rule
outcome-oriented
performance evaluation
Journal
IF:
7.2
Papers:
1.8W
Citations:
9.8W