To handle the complex cases that the uncertain parameters of the Powertrain Mounting Systems （PMSs） of electric vehicles are of imprecise information and correlation， a reliability analysis method of electric vehicle PMS is proposed by considering the imprecise information and correlation of PMS parameters. Firstly， evidence variables were employed to describe the parametrical imprecise information， and the correlation between evidence variables was described by the correlation coefficient matrix. Then， based on the affine transformation and standardization techniques， the relevant evidence variables were transformed into standard irrelevant evidence variables. Finally， by calculating the inherent characteristic response of the PMS as well as the belief function and plausibility function that meets design requirements， the reliable probability intervals of PMS responses were evaluated. The reliability analysis results of the numerical example of an electric vehicle PMS show that the parametrical imprecise information and correlation can be effectively handled by the correlated evidence variables； the parametrical imprecise information has a great influence on the system reliability， especially on the reliability related to the natural frequency in Roll direction； the system reliability changes obviously as the parametrical correlation increases， and ignoring the parametrical correlation may lead to large errors of reliability analysis.