The size of vessels has increased significantly over the last decades and will increase even further in the decades to come. This trend is mainly visible in the size of container ships. A decade ago the largest container ship had a capacity of 13.500 TEU, nowadays the largest container ship has a capacity of almost 24.000 TEU. The capacity of the largest container ship has almost doubled in only one decade. Due to their increasing size, vessels tend to become more flexible and their structural natural frequencies move towards the range of typical wave frequencies. Excitation of these structural modes will lead to resonant hull vibrations. The traditional method to determine wave loads and the response of vessels due to these wave loads is based on the assumption that the vessel acts as a rigid body. For large vessels, which tend to be more flexible, this is not a reliable method. A method is required that takes into account the fluid-structure interaction, which is in this case called hydroelasticity. The most used method to approach hydroelasticity is based on modal decomposition and was first established by Bishop \& Price in 1979. Since then, this method has been developed into commercial software and is nowadays accessible to the industry. However, the currently available commercial software is limited to first order hydroelasticity. Larger vessels become more flexible, but even the largest vessels still have their first structural natural frequency above the range of typical wave frequencies. It is thus very unlikely that first order theory can explain the excitation of these structural modes. However, several evidences of resonance around the wet natural frequencies have been reported. The structural modes could be excited by second order wave loads. Second order wave loads contain sum-frequency terms that occur at higher frequencies than the encountered wave frequencies. These second order wave loads have been considered for elastic bodies by including these loads in the method based on modal decomposition. It is expected that the second order hydroelastic response has a significant contribution to the total response for very large vessels. Pioneering Spirit is currently the largest construction vessel in the world. The vessel is designed for the single-lift installation and removal of large oil and gas platforms. A correct prediction of the response of this vessel requires a hydroelastic analysis. The vessel has its structural natural frequencies far above the range of expected wave frequencies, but evidences of resonance around the wet natural frequencies have been reported for this vessel. This research focuses on explaining the observed response by considering second order wave loads. The second order hydroelastic response is obtained with the method based on modal decomposition with a non-commercial version of Hydrostar. It was found that the second order wave loads cause the observed high-frequent response of the vessel. The main disadvantage of this method is that the second order hydroelastic response can not be computed with the use of commercial software. This makes it impossible to predict the hydroelastic response of the vessel in the future for different drafts and loading conditions. An alternative method is proposed to overcome this problem. This method is based on multi-body dynamics. The advantage of this method is that there is commercial software that can obtain second order wave loads for multiple interacting bodies. A multi-body model is created for Pioneering Spirit by dividing the vessel into a number of rigid bodies. These bodies are connected to each other by beam elements. The linear transfer functions obtained with this method are compared to the ones obtained with modal decomposition and show good agreement. This multi-body model is then used to obtain the second order hydroelastic response of the vessel by calculating the second order wave loads and substituting these loads in the equation of motion. The second order hydroelastic response shows very little resemblance with the method based on modal decomposition and with the measured response. It is found that in the calculated second order wave loads, a part of the second order wave loads is missing. This part has a significant contribution to the sum-frequency terms of the second order wave loads according to a number of studies. Taking into account this missing part should lead to a more accurate approximation of the second order wave loads and eventually to a more accurate approximation of the second order hydroelastic response obtained with multi-body dynamics.