Once a semi-submersible is operating at an inconvenient draft (A shallow draft with limited water column above the pontoons), the passing waves over the pontoons can not keep their linear motion and energy will be transferred to higher harmonic wave frequencies. This means the hydrodynamic response will also contain energy at these higher frequencies. Conventional linear diffraction solvers are not able to solve the combined response of the wave frequency and its higher harmonics, which then result in a non-physical solution for the wave loads, motions and internal loads. This research aims to obtain a better insight into the hydrodynamic response at the inconvenient draft and ultimately on the internal loads of the semi-submersible. In the first part of this research, model test solutions, linear potential solutions obtained with WAMIT and CFD solutions obtained with ComFLOW are compared. the comparison shows that ComFLOW is able to provide more accurate wave loads compared to the linear potential solver. The higher harmon- ics observed in the measured wave loads during model tests are correctly predicted by ComFLOW, although maximum deviations of 30% are still observed between measured and predicted wave-load amplitudes. However, ComFLOW is not able to solve the motions of a free-floating semi-submersible correctly. Due to pressure peaks in the wave-exciting forces, solving the equation of motion results in an incorrect motion response and ultimately results in incorrect internal loads. Although a significant effort was made - in close collaboration with the ComFLOW developer - to improve this functionality, the results remained unsatisfactory. Even so, in order to obtain an insight into the higher harmonic response contributions to the internal loads, a parameter study has been conducted, using tests in which the semi-submersible was held captive. This parameter study was conducted by systematically varying wave amplitudes and draft, and resulted in situational limits at which the higher harmonic response contribution becomes significant and the linear relation between the incoming wave and the hydrodynamic response is lost. This limit is shown to be dependent on the Ursell number. Furthermore, it is demonstrated that the significance of the higher harmonic response contribution increases from 10% to 40% throughout the inconvenient draft, while the most severe situations resulted in a higher harmonic response contribution of 50% of the total response amplitude. In the final part of this study, an attempt is made to couple the wave loads from ComFLOW to internal loads. A quantitative analysis is made on the effect of the higher harmonic responses of the wave loads on the internal loads. The time domain simulator aNySim is used combined with the wave- exciting forces on a captive semi-submersible calculated with ComFLOW. A multi-body analysis is used to obtain the internal loads on the aft and front part of the semi-submersible. This did not provide the correct answers because the stiffness of the spring damping between the two sections affects the higher harmonic response contribution. This method overestimated the higher harmonic response contribution. A better understanding of the joint and the joint stiffness/damping of a dual-body simulation may solve the encountered problems.