The volatile flavor substances of 5 representative Mongolian cheese samples collected from different regions of Inner Mongolia were determined by headspace purge and trap (PT) coupled with triple quadrupole gas chromatography (QqQ-GC), and a flavor intensity evaluation model was constructed based on principal component analysis (PCA). The results showed that a total of 86 volatile substances were identified in Mongolian cheese, among which 24 were the main volatile substances. The flavor intensity evaluation model (F = 0.35351F₁ + 0.28046F₂ + 0.19846F₃ + 0.16756F₄) showed that the cumulative contribution rate of the first four principal components was 100.000%, which included ethyl caprylate, n-hexanol, ethyl decanoate, n-decanoic acid, acetic acid, and acetaldehyde, ect. They could reflect the original data accurately. In addition, the sensory evaluation results of the 5 cheese samples were consistent with the scores from the flavor intensity evaluation model, which indicated that the constructed model could effectively evaluate the flavor of Mongolian cheese, providing a new method for the evaluation of the flavor of Mongolian cheese.