With the increasing popularity of electric vehicles, the demand for charging stations is also growing. This not only leads to long waiting times for electric vehicle drivers, but also causes frustration and inconvenience. Accurate predictions about the availability of charging points make the charging infrastructure more efficient and user-friendly. This can help to minimise waiting times and increase the overall satisfaction of electric vehicle users. Against this background, the aim of this thesis is to predict the availability of charging stations up to two hours in advance. This is based on the occupancy history of a total of 1,833 fast charging stations in Baden-Württemberg. The thesis focuses on the evaluation of various machine learning methods, in particular an LSTM approach, which represents the current state of research.