Modular-spoke-type permanent-magnet (MSTPM) machines are novel in-wheel traction machines for electric vehicles (EVs). Considering the strict constraints including the mass and volume of electric machines in EVs, which are heavily dependent on the dissipation condition, an accurate thermal model and resultant temperature distribution analysis are key challenges. Therefore, in this paper, a lumped parameter thermal network (LPTN) model for thermal analysis of MSTPM machines is proposed, where the heat transfer coefficients are determined theoretically instead of empirically, which means that the proposed LPTN model can be applicable for the MSTPM machines with different specifications and parameters. Both the steady-state and transient-state temperature rises are analyzed by the LPTN model, and the results are verified by both 3-D finite element analysis (FEA) predictions and experiments, indicating that the proposed method has advantages in both computational efficiency and accuracy.