Short-term prediction of traffic flow conditions in a multilane multiclass network

The research described in this contribution constitutes dedicated traffic flow theory development, aimed at establishing operational models supporting the analysis of multiclass (MUC) network traffic operations (i.e. estimation and prediction) in the context of ATCS (Advanced Traffic Control Systems) and ATIS (Advanced Traffic Information Systems) under a variety of circumstances. To this end, with a concise outline of the theory of multilane multiclass (MLMC) traffic flow, we present a new model-based approach to forecasting class-specific multilane network traffic conditions. The research is aimed at an operational traffic flow model for the off-line and on-line analysis of MLMC flow operations in networks, given a specific set of dynamic traffic management measures. The developed theory provides a generalization of traditional macroscopic traffic flow theory, while providing new insights into both the interactions between the distinct user-classes and the nature of the control problem at hand, given complex traffic network configurations and traffic control instruments, yielding a better understanding of the response of heterogeneous traffic flows to multiple control measures.