Due to a change in what leading clients demand from the architecture, engineering and construction (AEC) industry, the AEC industry has to work more efficient. Knowledge management (KM) has a great potential to make the design process of the AEC industry more efficient. Nonetheless the traditional knowledge management systems (KMS) do not meet the requirements necessary to be effective. Therefore new types of KMS are needed. A possible solution is to combine the potential of BIM and data mining into a KMS. The main scope of this research is to develop a KMS based on BIM and data mining and test the effectiveness of this KMS. One of the most important aspects of the data mining process is the dataset that is analysed. To create such a dataset from BIM models a tool is created to automate the extraction of data from the BIM models into a dataset useful for data mining. This has led to insights about the current state of both the quality and quantity of BIM models. After the application of the data mining algorithms on the dataset it is suggested that the amount of available BIM models has to increase for the KMS to be successful. To test this statement, additional data has been collected from publically available data. These data have been added to the dataset and the data mining process has been applied on the newly created dataset. It could be concluded that if the quality of the BIM models meets the necessary standard, the KMS system can be successful as long as enough BIM data is available and this BIM data is about buildings of a similar type.