To cope up with the ever growing amount of simulation runs being performed, tools and techniques are needed to make use of the huge amount of simulation data being stored. While current Simulation Data Management systems and the IT infrastructure already allow storing and accessing huge datasets and would facilitate putting this into action for analysis, the users usually only have tools and the time to make rather straight forward model to model comparisons, between current model versions and their immediate predecessors. To take analysis capabilities and model development a leap forward, it is necessary to also make use of whole model development branches to learn from the gathered simulation information. Making use of the Principal Component Analysis, a dimension reduction algorithm out of the unsupervised learning techniques, a new database was developed. Continuously being fed with new simulation runs, this database now enables us to automatically detect unknown behaviour in the most recent simulation runs compared to all predecessors at a time. To achieve this, the database does not only need to store and detect every new deformation pattern, but in addition several obstacles like a mapping between different simulation models, a performance efficient database format and a technique to also detect local effects had to be addressed. All these developments now make it possible to automatically detect anomalies/outliers within the crash deformation behaviour, pinpointing exactly to the location in space and time where the model is showing unknown or unwanted deformation patterns. While in daily work the engineer often only has time to compare single simulations with each other, this approach shows how to compare the current simulation with hundreds of predecessors at a time. In addition to this automatic detection of new crash patterns a second application becomes possible: The search for deformation patterns. Thus the engineer can interactively ask the database, whether or not the behaviour of certain parts has already shown up in any of the simulations in the database before. As an outcome he will also see which of the runs have a rather similar behaviour and which are different. Also time-shifted events can be seen, as the search does not only compare deformation patterns of a single time step, but instead does a comparison of all states among all simulation runs to give the engineer the best insights possible. Thus we are able to incorporate complete development branches into the daily engineering process and provide a tool to the engineer, to exploit all related simulations.