Estimating the Impacts of Transit Service on Roadway Congestion

Principal Investigator:

Steven E. Polzin
Director, Mobility Policy Research
National Center for Transit Research (NCTR) at the
Center for Urban Transportation Research
University of South Florida
4202 E. Fowler Avenue, CUT 100
Tampa, Florida 33620-5375


Background Statement

For the past several years the Urban Mobility Report, the preeminent research initiative to quantify trends in urban travel congestion, has included an analysis of the impact that the presence of public transportation has been reducing congestion relative to the levels that might exist in the absence of public transit service. This hypothetical scenario is intended to provide one measure of one of the benefits of public transportation service, i.e., reducing congestion in urban areas. The current methodology can be improved by using a more refined methodology and more specific urban area data in order to come up with a more robust method. This research will provide a refined methodology for consideration in subsequent years of assessing the impact public transportation has on congestion. This research will be conducted in coordination with Texas Transportation Institute (TTI) faculty.

The issue of assessing the congestion impact of public transportation is a conceptual or hypothetical question to some extent as there are no serious proposals under discussion that would suggest public transportation services would no longer exist in congested urban areas. However, such an assessment provides some insight on the value of transit and perhaps weighs in on future transit investments. The congestion consequence of public transit services is only one aspect of the impacts of public transportation. Congestion reduction is among the goals cited for investment in public transportation and in select markets the collective nature of public transit travel minimizes the land area or travel way area required to accommodate a given number of people, thus, everything else held constant could result in less congestion. The sequence of photos below characterizes the differences in travel way space requirements in traveling by auto versus loaded public transit vehicles.

Research Objectives

One could frame a discussion of the issue of the impact of transit on urban congestion in a number of different ways. One might frame the issue in a number of significantly different ways each of which could represent a legitimate question and each of which would require a different type of methodology to more fully explore. Three possible different questions are presented below. The third question represents the one being addressed in this research initiative.

1. In the absence of transit how would land-use patterns be different and what is the congestion consequence?
2. If public resources directed to public transit were instead directed to roadway capacity how would congestion be different?
3. If transit were discontinued and land use patterns remained the same, what would be the impact on roadway congestion?

Transportation in general including public transportation are known to influence land-use location decisions for businesses and individuals. Thus, in the long run, the presence or absence of transportation capacity would influence development patterns and the subsequent level and geographic distribution of travel demand. Therefore, the full consequence of not having transportation capacity would not realistically be known in the short-term as land-use responses are long range impacts. Scenarios that would be looking at near-term impacts of removing public transit are interesting impact assessment exercises but are not realistic in the long term as travel behaviors and land-use development patterns would be markedly different in the absence of public transportation – particularly in situations where public transportation transports a significant share of travel.

Similarly, this research does not purport to look at the net impact that might result if resources currently invested in public transportation were instead deployed on alternative modes. While this is a legitimate policy question, there is no such serious proposal on the table and such an analysis is beyond the scope of this work.

What this analysis will do is support an impact assessment analysis carried out by the Texas Transportation Institute in an effort to determine a conceptual estimate of the congestion consequence should public transportation no longer be available. Implicit in that process has been the assumption that the demand for travel remains constant and that, in the absence of public transportation, the vast majority of the demand would be accommodated a personal auto travel in the same geography.

This research will explore several assumptions regarding reasonable conditions and behaviors that might exist in the short term were public transit no longer available. These assumptions are implicitly hypothetical however, they will be based on the best available data and understanding of travel behavior.

Task 1:  Post-Transit Travel Behavior Assumptions

Task 1 will define estimates for the post transit travel behavior of transit travelers.  This will include estimates with respect to the number of forgone trips, temporally shifted trips, altered trip destinations (trip length), and post transit mode choices including walk, bike, shared ride, and drive.   The social demographic characteristics of travelers as well as urban area characteristics would collectively be expected to influence the probable alternative means of travel should transit no longer be available. Various data sources will be used to provide estimates of how persons might travel in the absence of transit. Factors to consider would include whether or not they were from a household with autos available, trip purpose, trip length, and other social demographic characteristics. Some onboard surveys have included specific questions regarding what the traveler might do in the absence of the available transit service. Options include numerous responses such as:  not take the trip, take the trip at a different time when an auto is available, walk, bike, ride with a friend, choose a different destination for the activity that’s accessible by an available mode, etc.  Some data sources also have measures of whether or not transit travelers are licensed drivers, and have physical or mental disabilities, etc.

We expect to be able to produce a set of assumptions that would drive the amount and the temporal timing of additional vehicle travel that would result in the short term if transit were to be discontinued. These behavior responses are likely to vary based on characteristics of the urban area.  After reviewing the data the research team will determine a logical classification structure such that urban areas included in The Urban Mobility Report can be assigned a set of post-transit travel behavior assumptions. 

This analysis will also include a review of the data to determine the temporal distribution of current and an estimated post-transit travel.  This may include travel direction (peak or off-peak direction) and roadway functional classification on which trips are likely to occur.

Supporting data and methodology assumptions will be documented in the technical memorandum produced as part of this research.

Task 2:  Roadway Capacity Impacts of Removing Transit

 The current Urban Mobility Report methodology does not account for the fact that roadway capacity would be freed up if no transit were in operation.   Transit vehicles operating in mixed traffic consumed the equivalent of several private vehicles worth of roadway capacity due to their physical size and performance characteristics regarding acceleration and deceleration in addition to the impact they have associated with stopping to board and discharge passengers where the stop dwell time may be impacting roadway traffic significantly.  This task in the research will explore strategies for attempting to quantify the roadway capacity increase available due to the assumption of transit being eliminated. Obviously, this would be most significant for buses operating in mixed traffic in local service (as opposed to express) and without bus bays. In addition, guideway services might result in some impact on cross traffic roadway capacity based on the number of at grade crossings and the frequency of transit and volume of cross traffic.

This task will include a review of the literature to understand what research exist that may have quantified through simulation, measurement, or other methods, the impact of transit on roadway capacity. Based on the results of the literature review we will determine whether or not reasonable assumptions can be made or whether additional research or simulation would be required to more robustly answer this question.  We anticipate the roadway capacity consequence to also be a function of the nature of the roadway (number of lanes, functional classification, topography, etc.), near side/ far side bus stop policies, and bus dwell time as influenced by fare structure, bikes on bus policy, handicap volume, etc.  Also, the extent to which the roadway is congested would be highly relevant as removing transit service from a roadway that has adequate capacity would not impact roadway performance.

The findings from this analysis will be discussed with the TTI project team and recommendations for both near-term and perhaps longer term methodology modifications will be made.

Supporting data and methodology assumptions will be documented in the technical memorandum produced as part of this research.

Task 3:  Travel Path Length Adjustments for Mode Differences

Trip length between a given origin and destination is likely to be different between roadway and transit modes.  The density of the transit route network is less than that of the roadway network in most instances in addition, many transit networks are oriented to the central business district and require extra trip circuity to reach a destination as a person may need to travel to a central location to transfer vehicles before continuing to their destination. We are not aware of any existing research that quantifies the relationship between roadway person miles of travel and transit person miles of travel that would be required for travel between a given origin and destination. This task will explore the literature or to see if any information exists regarding that relationship. The implications of this information are that it might be appropriate to factor transit trip person miles of travel down slightly to account for the fact that the share of trips that switch to roadway might be somewhat shorter in length than the current transit trip – thus having a proportionally different impact in terms of roadway capacity consumption.

It may be possible to use existing network pathfinding software to test relationships between trip lengths on transit and roadway networks for a sample of transit origin and destination pairs. We will explore the prospect of carrying out such a test in order to determine an appropriate adjustment factor should one not be available from the literature.

Task 4: Research Documentation and Final Report

While the primary deliverable from this research will be a set of assumptions and data to support the TTI urban mobility report production process for 2013, the overall effort will be documented in a technical report.

This will include working with TTI on the documentation of their overall methodology and the in inclusion of appropriate descriptive material for inclusion in the Urban Mobility Report and supporting documents.

Project Oversight and Peer Review

Project peer review will be provided by TTI faculty in addition to NCTR faculty involved in the project. In addition, TTI will brief other Urban Mobility Report stakeholders on the changes in methodology and share that feedback with the project team.

Quarterly Progress Reports

Quarterly progress reports will be provided to NCTR. 


 The primary project deliverable will be a data set to support The Urban Mobility Report. It will be fully documented in detail to the technical memorandum that will serve as the project report.  We believe there are some innovative aspects of this work that may be suitable for publication/presentation through other venues.

Use of Graduate Students

Graduate students will be involved in the completion of this project and help with the collection and assembly of data for each of the Metropolitan areas for which The Urban Mobility Report prepares congestion estimates.


There will be no equipment acquisition required to conduct this study.


No travel is anticipated to carry out this work.

Project Schedule

November 2013 through May 2014

Project Budget

Total Project Cost     $100,000

Leave a Reply

Please complete the simple math problem prior to submitting your comment. This reduces spam. Thanks! * Time limit is exhausted. Please reload CAPTCHA.

This site uses Akismet to reduce spam. Learn how your comment data is processed.