Dynamic and Stochastic Transition of Traffic Conditions and Its Application in Urban Traffic Mobility

Analytical models developed using field data can provide useful information with acceptable confidence to evaluate and predict the operational characteristics of a highway. As such, this study presents statistical models that can be used to estimate the travel time or speed distribution, cluster different traffic conditions, to model the dynamic transition of traffic regimes (DTR), and quantify the disparity-effects on the DTR associated with different lateral lane positions (i.e., lane near shoulder, middle lane(s) and lane near a median) as well as different days of the week. In the analysis, this study uses Bayesian frameworks to estimate the model parameters. These frameworks reduce the impact of model over-fitting and also incorporate uncertainty in the estimates. Data from a freeway corridor along I-295 located in Jacksonville, Florida were selected for analysis. It includes data from individual microwave vehicle sensors, segment level aggregated traffic data and data aggregated at a corridor level. The proposed probabilistic frameworks developed by this study can be a useful resource in detecting and evaluating different traffic conditions, which can facilitate the planning action to implement congestion-related countermeasures in urban areas. In addition, findings from the hierarchical regression model presented by the current study can be used in the application of intelligent transportation systems, mainly in the dynamic lane-management strategy.