Macroscopic discontinuity modeling for multiclass multilane traffic flow operations

Abstract

Congestion in traffic networks causes severe problems in and around large cities. It is the source of important economic inefficiencies, both on the level of individual persons and of the society as a whole. However, societal and environmental constraints prohibit large-scale extensions of the currently available network infrastructure. Consequently, solutions need to be sought in using the existing traffic networks more efficiently. Dynamic Traffic Management (DTM) plays an important role in this, providing the possibilities to aim for a sustainable transportation system. Empirical studies have also shown that solving traffic problems locally frequently only amounts to relocating bottlenecks, either within the freeway network, or from a freeway to the underlying urban network. Improving traffic conditions on a ring road around a city may also increase the use of this ring road by urban traffic. This is why the net profit of isolated DTM measures if often small, and integrated approaches are needed. At the same time, currently synergetic effects between different DTM measures are not fully utilized for improving network-wide traffic conditions effectively. We envisage that to effectively deploy DTM to resolve congestion on freeways and in urban networks, an integrated and coordinated approach to DTM is required. To develop such a control method, an analytical model-based approach is required. To this purpose, an elaborate model describing the dynamics of traffic flow on freeways and in urban networks is necessary. In this thesis, we establish a new approach to model the (infrastructure) discontinuities of heterogeneous traffic flow operations on multilane freeways and in multilane urban networks. The model provides insight into the interactions of vehicles between lanes near to a bottleneck (that is, on--ramps, off--ramps, weaving areas, and intersections). Research conducted in this thesis involves three parts: the development of a generalized gas-kinetic and macroscopic model for interrupted traffic flow, the development of a dedicated numerical scheme for general high(er) order macroscopic traffic models (of the hyperbolic type), and the development and application of an automated calibration procedure for general macroscopic traffic models, used to cross-compare the performance of the proposed model with other models.