The aim of the paper is to propose analytical coupled thermomechanical model of the crankshaft system, which includes the mutual interaction between thermodynamic and mechanical phenomena occurring in engines. The most relevant dynamic effects observable in the crank system are connected with its kinematics. When the mechanism operates there are also additional effects corresponding with stress, strain and thermal fields. Elastic properties of the system parts and changeable stiffness of the fuel-air mixture cause different dynamics of the entire device. The authors assumed that rigid motion of the crank mechanism, parts deformation and thermodynamic effects and their mutual dependencies will be included in the modelling process. Elasticity of the crankshaft system components is the reason for the difference between a rigid ’ideal’ motion and the real movement of crankshaft elements. In most cases, it is enough to assume linear elastic material features based on the relatively high stiffness of the system preventing big deformations. This ensures small displacements and the correctness of the applied model. The performed investigations have shown an influence of the crank system flexibility on the overall device response. Moreover, the parameters that change due to thermodynamic and mechanical properties of the working medium were taken into account. The authors have applied simple engine cycles (Otto, Diesel or combined model) for determining engine load including the connection between mechanical and thermodynamic state variables. This caused another decrease of the total system stiffness. Further numerical testing proved a visible effect of the applied approach in the global system response. The main discrepancies are observable in natural frequencies and vibration modes. It can also be stated that the utilization of different engine cycles results in different engine features. The paper is concluded with an analysis of the existing systems and mutual reactions from the assumed phenomena. The authors have shown the necessity to take a transdisciplinary approach into account in order to model complex systems.