Abstract: Case-based reasoning method is presented after analyzing the complexities of vehicle fault and characteristic of diagnosis system. A key issue in case-based reasoning is how to retrieve similar cases. A kind of case retrieval algorithm—multi-layer synthetic similarity measure is preseuted that integrates fuzzy set concepts into the case retrieval process, which focuses fuzzy set center distances influencing on similarity. In the end, an example of abnormal tire wear validates that the algorithm is brief, rational and effective.

Key words: Case retrieval, Case-based reasoning, Diagnosis, Synthetic similarity measure