TRB 2016 Blue Ribbon Committee
16th National Transportation Planning Applications Conference

Simulating Detailed Parking Location Choice in an Activity-Based Model Framework


Corresponding Author: Mark Bradley, RSG

Presented By: Mark Bradley, RSG

Abstract

Parking supply and pricing are key policy considerations in many US cities, including San Francisco. While activity-based models have provided great improvements in simulating both geographic and socio-demographic details influencing travel choices, the treatment of parking has, in most cases, not been improved considerably over what is done in trip-based models.

This paper describes the incorporation of parking choice at the destination end for all trips to destinations in San Francisco County in the DaySim AB model system. The model uses a fairly complete inventory of parking supply, treating each off-street public parking lot or garage as a separate spatial alternative, and also treating metered and free on-street parking as alternatives at the spatial level of Census blocks (used as "microzones" in the DaySim demand models). For each tour and trip to San Francisco, the choice of parking type and location is modeled, taking into account driving time to the parking location, walking time from the location to the destination, the price of parking, and capacity constraints. Parking prices are varied by both time of day and duration (e.g. "early bird" parking prices). Capacity constraints are also enforced by time of day and duration.

The auto ownership model was also re-estimated to take into account the availability of off-street and on-street residential parking spaces within each Census block in the City of San Francisco.

The paper describes the data that is input to the model and how the model uses the data to simulate parking choice, and how the attractiveness and availability of parking feeds up into other travel choices that are simulated, including mode choice, destination choice, and time of day choice. It then describes the results of policy sensitivity tests run using the enhanced AB model.

No presentation file uploaded. Contact Corresponding Author for Presentation

Discuss This Abstract