Abstract: The rental service of electric vehicles (EVs) has attracted considerable attention due to its unique advantages such as environmental protection and convenience. To better provide service, companies must maintain long-term customer loyalty (CL) . Existing research mainly clarifies the relationship between customer perceived value (CPV) and CL, and fails to optimize the rental service system. First, multi-dimensional measurement models and path hypothetical models of customer perceived value (CPV), customer trust (CT), planned behavior (PB) and CL are established. Then, factor analysis is used to verify that the questionnaire data has good reliability and validity. Finally, the path analysis of structural equation modeling (SEM) is used to derive the main factors and value-added paths that drive CL in EV rental service. The results show that CPV in rental service of EVs effectively improves CT, actively drives PB, and ultimately plays a positive role in CL. This study can improve CL of EV rental service, promote the urban governance and provide companies with decision support and service optimization direction.