This study presents a model to characterize changes in network traffic flows as a result of implementing connected and autonomous vehicle (CAV) technology based on traffic network and built-environment characteristics. To develop such a model, first, the POLARIS agent-based modeling platform is used to predict changes in average daily traffic (ADT) under CAV scenario in the road network of Chicago metropolitan area as the dependent variable of the model. Second, a comprehensive set of variables and indicators representing network characteristics and urban structure patterns are generated. Finally, three machine learning techniques, namely, K-Nearest neighbors, Random Forest, and eXtreme Gradient Boosting, are used to characterize changes in ADT based on network characteristics under a CAV scenario. The estimated models are validated and are found to yield acceptable performance. In addition, SHapley Additive exPlanations (SHAP) analysis tool is employed to investigate the impact of important features on changes in ADT.

Accident detection is a vital part of traffic safety. Many road users suffer from traffic accidents, as well as their consequences such as delay, congestion, air pollution, and so on. In this study, we utilize two advanced deep learning techniques, Long Short-Term Memory (LSTM) and Gated Recurrent Units (GRUs), to detect traffic accidents in Chicago. These two techniques are selected because they are known to perform well with sequential data (i.e., time series). The full dataset consists of 241 accident and 6,038 non-accident cases selected from Chicago expressway, and it includes traffic spatiotemporal data, weather condition data, and congestion status data. Moreover, because the dataset is imbalanced (i.e., the dataset contains many more non-accident cases than accident cases), Synthetic Minority Over-sampling Technique (SMOTE) is employed. Overall, the two models perform significantly well, both with an Area Under Curve (AUC) of 0.85. Nonetheless, the GRU model is observed to perform slightly better than LSTM model with respect to detection rate. The performance of both models is similar in terms of false alarm rate.