Understanding pedestrian flow is important to keep public spaces safe. This can be done by using pedestrian models to simulate individual movement and crowd behaviour. Most pedestrian models are based on individuals, while pedestrians tend to walk in groups. The M4MA model is a discrete choice model now being created combining individual and social factors. The goal of this research is to make the M4MA model behave conform to real life group behaviour.
To do this, first, data from real life experiments will be analysed. Secondly, this data will be used to tune two parameters in the model involving group behaviour. The data had problems with missing data points on the left and right side of the experimental area, but could be used for the tuning of the parameters in the next phase. The model was able to perform similar to real life behaviour, but was not perfect. For better results, it might be necessary to add different parameters to the model.