The Exploratory Advanced Research Program

The Impact of Automated Transit, Pedestrian, and Bicycling Facilities on Urban Travel Patterns Summary Report

Introduction

This report summarizes the results of a Federal Highway Administration (FHWA) Exploratory Advanced Research (EAR) Program research project that explored the potential of a hypothetical driverless vehicle to improve access to and use of available rapid-transit rail service. The distance between a traveler’s origin or destination and the nearest public transit station is often time-consuming, inconvenient, or unsafe to travel and may discourage potential transit riders from using the system. This distance, known as the “last-mile problem,” impedes full usage of existing transit systems, particularly in outlying suburban and exurban areas.

Walking, cycling, driving or being driven, or riding a station feeder bus line are the usual options for reaching the nearest station. In some public transit systems, community shuttle buses or vans supplement existing feeder bus lines to improve residents’ access to commuter rail stations, but the challenge in providing community shuttles is the high cost of labor, fuel, and equipment, particularly in neighborhoods with low population density. In these areas, the cost per rider for shuttle service can be prohibitively high, whether in the form of rider fares or tax-supported subsidies. Automated vehicles would eliminate the cost of operator labor from this equation, possibly lowering costs sufficiently to make high-frequency service feasible and offering a solution to the last-mile problem. Recent rapid advances in automated vehicle technology have brought the driverless vehicle within reach, potentially within the next decade for low-speed neighborhood travel.

Whether the availability of an innovative transit shuttle service would be incentive enough to attract large numbers of people to public transit is an open question, however. In this project, the University of Michigan and the University of Illinois at Chicago conducted an initial inquiry into the potential impact of three types of neighborhood improvements on residents’ use of transit:

• Automated, high-frequency community shuttle service to and from commuter rail stations.
• Cycling facilities.
• Urban design improvements in the vicinity of the transit station to facilitate walking and cycling.

The researchers also explored the effect of changes in the costs of parking and driving on public transit use.

This project was not a feasibility study; rather, the research design was a “best-case” approach to help assess whether the concept of using automated vehicles as transit shuttles holds promise for further research and development. The project’s purpose was to inform discussions on policy choices regarding this option. Of specific interest was the relative capacity of the potential improvements to generate changes in travel behavior, that is, to shift trips from automobile to other modes of transportation and to increase use of regional public transit based on improved station access.

The best-case assumptions for the research included no congestion on local streets, no limitations due to weather or nighttime conditions, and unrestricted ability to alter rights-of-way to make transit and environmental improvements. Because the researchers examined whether the frequency and quality of service deemed possible with automated neighborhood shuttles might increase transit use and did not evaluate the feasibility, practicability, or acceptability of deploying the shuttles, the driverless aspect of the shuttles was not discussed with the research participants.

The study had several related elements:

• A literature review to identify urban design improvements shown to have a positive effect in promoting multimodal travel.
• An inventory and selection of sites and station approaches in four metropolitan Chicago neighborhoods, including mapping a 1.5-mi (2.4-km) buffer zone around the selected station by using geographical
  information system data (rights of way, land use, and census data).
• Preparation of photographs and illustrations of station approaches and potential improvements.
• A mail and telephone survey of households located within about 1.5 mi (2.4 km) of the station.
• Simulations using activity-based and agent-based modeling.

The research focused on four Chicago-area neighborhoods—Evanston, Skokie, Pilsen, and Cicero, Illinois. All are first-ring suburbs or, in the case of Pilsen, within Chicago city limits, and all are served by Chicago Transit Authority (CTA) rapid transit lines with capacity for increased use. The researchers selected the neighborhoods to represent different combinations of housing density and affluence, as shown in table 1.

Table 1. Neighborhood Selection Scheme.

The following sections discuss the characteristics of the four neighborhoods and transit stations and the neighborhood improvements that were the basis for the research.