No open-access location is known for this record yet.
No open-access location is known for this record yet.
This paper integrates real-time traveler information on prevailing traffic conditions (adaptive driving) into time-dependent origin-destination demand estimation in a simulation-based dynamic traffic assignment model. The objective is to explore the impact of adaptive driving on demand estimation, and to recognize a growing component of the traveling public in the methods used to infer network flow patterns given partial observations. Time-dependent origin-destination demand estimation can be formulated as a bi-level optimization problem in which the deviation from observations and historical demand are subject to a minimization. In the classical problem, it is assumed that users will follow paths that they have been assigned based on the user equilibrium model. Growing penetration of a wide array of devices enable users to receive real-time information on prevailing (and possibly predicted) traffic conditions, which enables them to update their path choice en-route. However, not all travelers receive or follow such information. This paper proposes a modified optimization algorithm to estimate the time-dependent origin-destination trip tables for large-scale networks considering varying population of adaptive drivers. The modified algorithm is solved with a tri-level algorithm. This algorithm is applied to a highly congested urban network of Chicago.
Loading…
Imported from OpenAlex (CC0) — nothing here is generated.