Accessibility and the Evaluation of Investments on the Beijing Subway

Recently Published

Efficiency of Beijing Subway Lines: Change in Person Weighted Accessibility per km.

This study measures the job and population accessibility via transit for Beijing using the cumulative opportunity metric. It is shown that transit accessibility varies widely across Beijing, but is highly focused on subway stations. Early lines added far more accessibility than more recently planned lines.

Transit Riders’ Perception of Waiting Time and Stops’ Surrounding Environments

Recently published

Reported vs. Observed Waiting Times
Reported vs. Observed Waiting Times

Reducing the burden of waiting in transit travel is critical to increasing the attractiveness of public transportation. Waiting time perceptions are highly subjective and vary according to mode, availability of schedule information, and stop amenities. The research on pedestrian design finds that high-quality and natural environments reduce stress and encourage walking and bicycling. It seems reasonable that similar effects would apply for transit users on the basis of the environments around transit stops, but little research directly explores the issue. This paper responds to this knowledge gap by examining how perceptions of waiting time vary in relation to stop environments. The research compared transit users’ actual and estimated wait time at 36 stops and stations in a mix of environmental situations in the region of the Twin Cities, Minnesota. Regression analysis explained the variation in riders’ self-reported wait- ing time as a function of their externally observed waiting times as well as characteristics of the environment surrounding the stop and station. For waits longer than 5 min, perceptible pollution and exposure to traffic led to significant overestimates of waiting time. Riders waiting at stops with dense, mature tree cover, however, significantly underestimated their waiting times. The effect of dense, mature tree cover is strong enough to compensate for the effects of both air pollution and traffic awareness. Policy implications and further research needs are discussed.

Access Across America: UMN ranks accessibility to jobs by transit

New research from the Accessibility Observatory at the University of Minnesota ranks 49 of the 50 largest (by population) metropolitan areas in the United States for connecting workers with jobs via transit.

The new rankings, part of the Access Across America study that began in 2013, focus on accessibility, a measure that examines both land use and transportation systems. Accessibility measures how many destinations, such as jobs, can be reached in a given time.

“This project updates our detailed evaluation of access to jobs by transit,” said Andrew Owen, director of the Observatory. “Transit is an essential transportation service for many Americans, and we directly compare the accessibility performance of America’s largest metropolitan areas.”

The findings have a range of uses and implications. State departments of transportation, metropolitan planning organizations, and transit agencies can apply the evaluations to performance goals related to congestion, reliability, and sustainability. In addition, detailed accessibility evaluation can help in selecting between project alternatives and prioritizing investments.

Top 10 metro areas: job accessibility by transit (January 2015)

  1. New York
  2. San Francisco
  3. Chicago
  4. Washington
  5. Los Angeles
  6. Boston
  7. Philadelphia
  8. Seattle
  9. San Jose
  10. Denver

The report—Access Across America: Transit 2015—presents detailed accessibility values for each of the 49 metropolitan areas, as well as detailed block-level color maps that illustrate the spatial patterns of accessibility within each area. New analysis of the data from the accessibility to jobs by transit rankings will continue to be published periodically.

The accessibility metrics presented in this report are designed to be comparable to those presented in the Accessibility Observatory’s earlier Access Across America: Auto 2015 report. “Taken together, these reports provide a comprehensive view of the relative accessibility impact of auto and transit systems across different cities,” Owen said.

For example, the Phoenix and Minneapolis–St. Paul metropolitan areas have effectively the same total number of jobs (1.7 million; ranked 13th and 14th respectively), and their auto accessibility rankings are also very close—13th and 12th. “However, they differ significantly in their transit accessibility rankings: Minneapolis–St. Paul ranks 12th in transit access to jobs, while Phoenix ranks 22nd,” Owen said.

In the study, rankings are determined by a weighted average of accessibility, giving a higher weight to closer, easier-to-access jobs. Jobs reachable within 10 minutes are weighted most heavily; jobs were given decreasing weight as travel time increases up to 60 minutes. Travel times were calculated using detailed pedestrian networks and full transit schedules for the 7:00 to 9:00 a.m. period. The calculations include all components of a transit journey, including “last mile” access and egress walking segments and transfers, and account for minute-by-minute variations in service frequency.

Future comparison reports will track the way that accessibility in these metropolitan areas evolves in response to transportation investments and land-use decisions, Owen explained.

The research was sponsored by the National Accessibility Evaluation Pooled-Fund Study, a multi-year effort led by the Minnesota Department of Transportation and supported by partners including the Federal Highway Administration and 10 state DOTs.

The Accessibility Observatory at the University of Minnesota is the nation’s leading resource for the research and application of accessibility-based transportation system evaluation. The Observatory is a program of the Center for Transportation Studies and the Department of Civil, Environmental, and Geo- Engineering. CTS is nationally renown for developing, fostering, and spreading innovation in transportation.

Accessibility Observatory reports, including the new analysis of job accessibility by transit (Access Across America: Transit 2015), are available at:

Journal of Transport and Land Use, Volume 10

With the New Year, the Journal of Transport and Land Use is moving to a continuous publication model, out with the old periodic batch mode, in with the ‘it’s done, it goes online’ model. So we cleared the backlog and hereby post Volume 10, with 25 papers.


Vol 10, No 1 (2017)

Table of Contents

Petrus van der waerden, Harry Timmermans, Marloes de Bruin-Verhoeven
Sarah Louise Brooke, Stephen Ison, Mohammed Quddus
Michael Manville
Wei-Shiuen Ng
Jean-François Doulet, Aurélien Delpirou, Teddy Delaunay
Carey Curtis, Jan Scheurer
Chi-Hong (Patrick) Tsai, Corinne Mulley, Matthew Burke, Barbara Yen
Ren Thomas, Luca Bertolini
James Robert McIntosh, Peter Newman, Roman Trubka, Jeff Kenworthy
Erik Elldér
Rolf Moeckel
Jinhyun Hong
Yuntao Guo, Srinivas Peeta, Sekhar Somenahalli
Sara Ishaq Mohammad, Daniel J. Graham, Patricia C. Melo
Liang Ma, Jennifer Dill
Louis A Merlin
Lei Zhang, David M Levinson
Reza Banai
Michael Manville
Haibing Jiang, David M Levinson
Niels Heeres, Terry Van Dijk, Jos Arts, Taede Tillema
Carole Turley Voulgaris, Brian D. Taylor, Evelyn Blumenberg, Anne Brown, Kelcie Ralph
Lewis Lehe
Na Chen, Gulsah Akar
Robin Lovelace, Anna Goodman, Rachel Aldred, Nikolai Berkoff, Ali Abbas, James Woodcock

On transit subsidies, first and second best

A reader writes:

Hi David, you postulate that transit that can’t be profitable or break even will shut down, which is certainly plausible and could even be welcome, but unless roads (serving human or automated drivers) and other transportation infrastrucutre are also subject to the same market discipline – if annual revenue doesn’t cover annualized full life cycle costs, including externalities, they are shutdown and the land is put to other uses – I would suggest we have a double standard about which is being disguised as a concern over subsidy.

Futhermore, this sidesteps questions of how non market forces such as zoning force the landscape into something for which transit is unsuitable and only vehicles will do.  Again, pointing to concerns about transit’s viability in suburbs or low density areas as a market concern ignores the very large non-market forces which mandate the low density in which transit does poorly.
I’m not trying to argue for transit subidies, I’d be happy to see a “pay what you use” model for all transport, even walking, provided we subject land used for transportation to the same market forces as we do the land for coffee shops and laundromats and provided people can make real market choices about how to use the land they own.  We are very far from such a free market and automated cars aren’t going to suddenly level the playing field.
In the absence of serious reform of these non-market forces, subsidizing transit should be seen as a way of managing the harm inflicted by those non-market forces.  Getting rid of non-market forces is often justified, but I’m always puzzled why proposals to jettison non-market forces always start with transit rather than free parking, free roads, free traffic control, free collision response, etc etc.
Yeah, life is unfair. Transit started out private and serves a far small share of the population in the US and so has a different political hurdle to overcome. I think I have written enough about road pricing that readers understand the position I hold, and not every sentence can say “but, but but.”
In the first best world, everything is priced properly. In our second best world, nothing is. The question is, should we accept the second best or do something to move to the first best? How distortionary is moving from second best to first best in steps (since it will never be done all at once)?

TRB 2017

I won’t be there in person for the first time in 20 years, but my research group’s research will be at the 96th Annual Meeting of the Transportation Research Board in Washington, DC, January 8-12, 2017.

Sunday, January 8, 2017

Workshop 107
Harnessing the Power of Transportation Data for Understanding Travel Behavior: Hands-on Tutorials on Traditional, Niche, and Novel Data Sets
9:00 a.m. – noon
Building a National Accessibility Data Set
Andrew Owen, University of Minnesota, Twin Cities

Workshop 131
Doctoral Student Research in Transportation Operations and Control
9:00 a.m. – noon
Network Econometrics and Traffic Flow Analysis
Alireza Ermagun, University of Minnesota, Twin Cities

Monday, January 9, 2017

Session 235
Freeway Operations
Poster Session
8:00 a.m. – 9:45 a.m.
Using Temporal Detrending to Observe the Spatial Correlation of Traffic
Alireza Ermagun, University of Minnesota, Twin Cities
Snigdhansu Chatterjee, University of Minnesota, Twin Cities
David Levinson, University of Minnesota, Twin Cities

Session 285
Exploring Innovative Data Sources and Approaches for Bicycle and Pedestrian Data Collection
Poster Session
10:15 a.m. – noon

Urban Trails and Demand Response to Weather Variations
Alireza Ermagun, University of Minnesota, Twin Cities
Greg Lindsey, University of Minnesota, Twin Cities

Session 288
Statistical Methods In Transportation
Poster Session
10:15 a.m. – noon
An Introduction to the Network Weight Matrix
Alireza Ermagun, University of Minnesota, Twin Cities
David Levinson, University of Minnesota, Twin Cities


Session 352
Lifecycle Transport Needs
Poster Session
1:30 – 3:15 p.m.
Dawn Spanhake, University of Minnesota, Twin Cities, presiding
Intra-household bargaining for school trip accompaniment of children: A group decision approach
Alireza Ermagun, University of Minnesota, Twin Cities
David Levinson, University of Minnesota, Twin Cities

Session 397
Transportation Safety Management: From Start to Finish
Lectern Session
1:30 – 3:15 p.m.
The Safest Path: Analyzing the Effects of Crash Costs on Route Choice and Accessibility
Mengying Cui, University of Minnesota, Twin Cities
David Levinson, University of Minnesota, Twin Cities

Tuesday, January 10, 2017


Session 565
Accessibility, Agglomeration, and Economic Development
Lectern Session
10:15 a.m. – noon
Intraurban Accessibility and Employment Density
Michael Iacono, University of Minnesota, Twin Cities
David Levinson, University of Minnesota, Twin Cities

Session 581
Uninterrupted Flow
Poster Session
10:15 a.m. – noon
Finding Optimum Capacity of Freeways considering Stochastic Capacity Concept
Alireza Ermagun, University of Minnesota, Twin Cities

Session 829
Transportation Demand Forecasting Poster Mega-session, Part 1
Poster Session
10:15 a.m. – noon
Mode Choice and Travel Distance Joint Models for School Trips
Alireza Ermagun, University of Minnesota, Twin Cities

Session 830
Transportation Demand Forecasting Poster Mega-session, Part 2
Poster Session
10:15 a.m. – noon
Spatiotemporal Traffic Forecasting: Review and Proposed Directions
Alireza Ermagun, University of Minnesota, Twin Cities
David Levinson, University of Minnesota, Twin Cities


Fuel Subsidies and the Greening of Transport

Stephen Leahy interviewed me about transport subsidies and their effect on demand. He wound up writing this article for China DialogueUS fossil fuel subsidies a brake on sustainable transport (which doesn’t cite me), so I will post the interview questions and my answers:

Sources report the US oil industry gets approx $20 billion in production subsidies annually.  This keeps many high-cost operations like fracking profitable.
1. Do these subsidies have an impact on mass transit use?
Yes, but not much, the US purchases like 374 million gallons per day,  or
136,510,000,000 gallons per year, which is about $ 273,020,000,000 ($273B) per year at $2/gallon. So if your estimate of subsidy is right, that is about 10% of the price. The price of fuel varies that much on weekly basis.
2. How does such subsidies impair efforts to improve/expand mass transit systems?
The more demand for driving, the less for other modes. The lower the price of driving, the more demand for driving.  Transit use rises when the price of fuel rises.
3. How does low gas prices affect political priorities given to transit
Even if the price of fuel were two times higher (like it was 8 or 9 years ago), transit demand would only be somewhat higher.
In 2008 US ridership was  10,597,931,000
In 2015 it was  10,609,605,000. According to APTA.

Which is a 0.1% increase. Over the same period, US population increased about 5%. (320M in 2015, 304.09 million in 2008) So if fuel prices were the same as 2008, one would expect transit ridership would have at least kept pace with population (say 5% higher). Obviously that would increase the political pressure for more transit service, but that’s a more than 100% increase the price of fuel gets you a 5% increase in ridership. A 10% increase in fuel prices (the magnitude of the subsidy) might get you a tenth of that increase (in the neighborhood of 0.5%). Of course there is no guarantee that eliminating the subsidy would increase the price of fuel by the same amount, since the price of fuel at the pump is determined by global market forces, and the US is one player among many, and not the low cost producer.

4. Any impacts on greening transit systems?
If the price of fuel were higher, systems would have a greater incentive to invest in electrical or natural gas fleets, but again, this is a really small effect given the relatively small size of the subsidy.
5. What are the main barriers to greater transit use and better transit systems?
Demand is low because transit is inconvenient (especially compared to the alternative) for most people in the US making most trips. It is a positive feedback system, but this works two ways. More transit service begets more ridership begets more transit service. Less service begets less ridership begets less service. We have followed the second pattern since the late 1940s, and transit ridership has been a very small share for most of a century now. There is no reason to believe that, outside large central cities, transit is a viable transportation alternative for most people in the US. And with cars set to get cheaper and more convenient (see point 2), it will remain that way. Certainly in large cities, particularly those places developed in the transit era of before 1930, transit could have a larger market share than it does now. Most of the US is not like that though.
6. Also if the US taxed gasoline as much as most European countries would the EV market take off?
It would help accelerate things, but the EV market will take off anyway because EVs will be cheaper than ICE engines to both build and operate.