In the Portland, Oregon region, bicycling to transit is an important mode from a policy perspective and is common enough to have been observed 69 times in a 2011-2012 regional household-travel survey. On the Tri-Met system, transit riders are permitted to bring their bicycles on the commuter rail, light rail and streetcar vehicles, provided there is sufficient room, or to mount their bicycles on the front of buses where capacity exists for up to two bikes. Bike racks are also provided on the C-Tran buses that serves the Vancouver, Washington portion of the modeling area. Both systems also provide bike lockers at selected stations; however, nearly all of the linked trips found in the survey were persons who brought their bikes with them in order to take advantage of having it available at the destination end and, potentially, other stops along the tour.

In this presentation, we will discuss how the bike-to-transit mode is being represented in a newly developed activity-based modeling system for the Portland region. The model system includes an innovative route choice model that takes into account rider perceptions of distance, elevation gain, minimizing conflicts with motor vehicle traffic, and maximizing usage of different types of bicycle infrastructure including off-street paths. Using the bicycle path finder, along with a transit network model that includes empirically derived transit station attribute utilities in its path-finding equations, we will present a distance-calibrated model of the choice of transit access/egress stops. Although the model will not be capacity constrained due to current limitations on the representation of modeled transit vehicles, we will discuss the implications of the model’s predicted, unconstrained outcomes and the need for representation of bicycle capacity constraints on transit vehicles.

In addition, we will present an estimated tour-based mode choice models in which the bike-to-transit has been included as an explicit tour mode alternative. In doing so, we will compare the parameter estimates for the composite utility of bike-to-transit paths with those of walk-transit and drive-transit modes.