The fault diagnosis expert system for logistics vehicle could diagnose and troubleshoot logistics vehicle. In order to improve the performance of the system, the fault tree had been built based on the failure mode and failure mechanism of logistics vehicle. Then, the improved CLIPS which could carry out forward and backward reasoning was used, and the knowledge base management system was established to manage the fuzzy rules and facts. The results showed that the improved CLIPS coupled with VC++ was able to enhance the capability of the fault diagnosis expert system for diagnosing fault from the logistics vehicle fuzzily (i.e. fuzzy diagnosis) and improve the intelligence level of the system.