Traditional safety modeling efforts primarily focus on accurate estimation of crash frequencies or rates. The true relationship between crashes and potential causal factors are not always easily discernible from safety models. While a model consisting of multiple causal factors may produce accurate estimates of crash measures, it may not accurately explain all causal relationships. Knowing the true cause-and-effect relationships are important when choosing countermeasures to address safety problems. This Exploratory Advanced Research Program project will generate artificial realistic datasets (ARD) that will mimic the known causal relationships. Thus, the best-performing methods can be applied by practitioners with confidence in safety evaluation and countermeasure selection. Not all countermeasures can be deployed in the field due to safety concerns. Driving simulator studies offer another source of artificial realistic data for evaluating new countermeasures. This project will generate artificial realistic simulator testbeds using real-world crash data. The project team consisting of experts from safety, crash modeling, machine learning, statistics, simulation, and human factors, will develop ARD datasets and testbeds for urban interchange facilities.