The presentation focused on integrating Machine Learning (ML) methods into Simulation Data Management (SDM) systems to enable effective big data analysis. The motivation behind this integration stemmed from the large datasets available in SDM systems, particularly with some Original Equipment Manufacturers performing over 1.5 million simulations annually in the crash domain alone. A key issue highlighted was the application of advanced ML/AI methods, which are often custom or proprietary and require domain-specific methodologies. To address this, the SCALE.sdm system provides an add-on concept, enabling the distribution of custom code to both backend and frontend sides within the server environment. This concept facilitates full structured access to data within the server environment. The presentation showcased the "SIDACT Event Detection Add-on" as an example application of this add-on concept. This particular add-on focuses on ML-based anomaly detection. It defines an "event" as unknown or unwanted behavior, such as anomalies in field variables, and identifies event properties like location/parts, outlier score, and event type. The basic workflow of the add-on involves data aggregation, dimension reduction, outlier detection, and event extraction. This process is integrated into SCALE.sdm, allowing for scanning of all incoming or selected simulations to detect anomalies and integrating the process and evaluation within the SDM system. A case study was presented using the Porsche 911 model. The presentation concluded with a demonstration of visualization capabilities in the Porsche 911 example, highlighting the comparison of simulation and physical tests, data analysis through machine learning and data mining, assessment of results with respect to project targets, comprehensive reporting, and evaluation of key results.