Macroscopic Modeling of Freeway Networks with Managed Lanes

Alex Kurzhanskiy
University of California, Berkeley (UC Berkeley)

A managed lane is a type of a highway lane that is operated using some special policy, such as admitting only vehicles of a certain class. Examples of managed lanes are: High-Occupancy Vehicle (HOV) lanes; toll lanes; High-Occupancy or Tolled (HOT) lanes; bus lanes; truck lanes; dynamic lanes, which may open or switch direction based on time of day or traffic condition.
We will describe the macroscopic model components needed to adequately represent traffic network with managed lanes: multi-commodity traffic flow, policy for junctions with multiple input and multiple output links, and local traffic assignment, where vehicles of certain types may choose between multiple downstream links.
We will then discuss how this theory applies to modeling of freeways with HOV and HOT lanes, and show techniques that reproduce phenomena inherent to HOV/T traffic: the inertia effect and the friction effect. The inertia effect reflects drivers’ inclination to stay in their lane as long as possible and switch only if this would obviously improve their travel condition. The friction effect reflects the empirically observed drivers’ fear of moving fast in the HOV lane while traffic in the adjacent GP links moves slowly due to congestion.
Some of the theoretical results that will be presented can be found in the paper by Matt Wright, Gabriel Gomes, Roberto Horowitz and Alex Kurzhanskiy: “A new model for multi-commodity macroscopic modeling of complex traffic networks”, available online:

Back to Workshop III: Traffic Control