Access as a performance indicator in a work-from-home world

David Zipper writes Post-Covid, Transit Agencies Must Look Beyond Ridership  for Bloomberg CityLab.


In the article, Zipper talks about using transit access as a performance measure. My graf below:

The idea of transit access isn’t new, but our ability to put a useful number on it is. David Levinson, a professor at the University of Sydney who has written numerous books about transportation access, says that quantitative breakthroughs now allow planners to make far more precise calculations than before. “We’ve got better data now through the General Transit Feed Specification and GPS, as well as from Census Bureau datasets. For each person, the data tells which block they live in and which block they work in. This didn’t exist at that detailed a level until the mid-2000s.”

Obviously I like access, which measures how many valued destinations people can reach in a given amount of time. But in the end, ridership is the raison d’être for transit shops. For all the access in the world, if a bus doesn’t actually serve any actual people, it has failed.

When the ridership is unknown (e.g. for planning a change in service or new construction, where at best we can make an informed guess about a future number of riders, a guess buried in a lot of modeling obscurantism), then access has to date been an excellent performance metric because access is correlated with ridership. The more places people can reach by public transport, the more places they will go.

Ridership fluctuates for many reasons, pandemic among them. Access will be more stable as an indicator. In the absence of other information, I would argue that increases in person-weighted access most per dollar spent will be the most useful for society. When we find that post-pandemic demand for offices (especially CBD offices) falls, jobs that are nominally at a site, but not really (because of 2, 3, 4, or 5-day per week work from home, or work elsewhere, e.g.) will imply more access than they really create. It will be years before the data properly accounts for this. In a perhaps idealised world in which decisions are made based on analysis, the use of access without controlling for this problem will distort our conclusions about where transit services should run. We will favour serving offices where people don’t actually work 5-days a week over sites like schools and hospitals and factories where they do. 

Where is a job, which was a crystal clear number (not really, but we imagined it was) in the days when people worked 9-5 jobs in offices and factories, no longer has the same kind of meaning, and our accessibility metrics will somehow need to account for this. We may need to fractionalize jobs in our access calculations.

The resulting designs for transit systems will have to catch up with these changes in work patterns.


Transportist: February 2021

So in personal news, my family and I were awarded Permanent Residency status in Australia (this is like the Green Card for US immigrants). So now we can buy real estate without penalty, and get some other benefits, the most important of which is the sword of Damocles is now no longer poised above our heads.

Transportist Posts



  • Aoustin, Louise, and David M Levinson. 2021. “Longing to Travel: Commute Appreciation during COVID-19.” Findings, January. [doi].




  • Wang, Haoyun, and Robert Noland. 2021. “Changes in the Pattern of Bikeshare Usage Due to the COVID-19 Pandemic.” Findings, January.
  • Aharoni, Jordan, Ron Buliung, and Raktim Mitra. 2021. “University and College Travel for Students with Mobility Impairment(s) in the Greater Toronto and Hamilton Area, Canada.” Findings, January.
  • Laverty, Anthony A, Rachel Aldred, and Anna Goodman. 2021. “The Impact of Introducing Low Traffic Neighbourhoods on Road Traffic Injuries.” Findings, January.
  • Aoustin, Louise, and David M Levinson. 2021. “Longing to Travel: Commute Appreciation during COVID-19.” Findings, January.
  • Rauws, Robin, and Dea van Lierop. 2020. “Returning to Public Transit after an Epidemic.” Findings, December.
  • Shi, Xiao, Anne Vernez Moudon, Brian H. Y. Lee, Qing Shen, and Xuegang (Jeff) Ban. 2020. “Factors Influencing Teleworking Productivity – a Natural Experiment during the COVID-19 Pandemic.” Findings, December.
  • Goodman, Anna, Anthony A Laverty, and Rachel Aldred. 2020. “The Impact of Introducing a Low Traffic Neighbourhood on Fire Service Emergency Response Times, in Waltham Forest London.” Findings, December.
  • Krall, Verena, Max F. Burg, Malte Schröder, and Marc Timme. 2020. “Number Fluctuations Induce Persistent Congestion.” Findings, December.
  • Adams, Thomas, and Rachel Aldred. 2020. “Cycling Injury Risk in London: Impacts of Road Characteristics and Infrastructure.” Findings, December.
  • Ravensbergen, Lea, and Bruce Newbold. 2020. “‘I Wouldn’t Want to Get on the Bus’: Older Adult Public Transit Use and Challenges during the COVID-19 Pandemic.” Findings, December.
  • Zhang, Yixue, Matthew Palm, Jonathan Scheff, Steven Farber, and Michael Widener. 2020. “Travel Survey Recruitment Through Facebook and Transit App: Lessons from COVID-19.” Findings, December.
  • Goodman, Anna, Scott Urban, and Rachel Aldred. 2020. “The Impact of Low Traffic Neighbourhoods and Other Active Travel Interventions on Vehicle Ownership: Findings from the Outer London Mini-Holland Programme.” Findings, December.

Other Research by Others

News & Opinion

  • Elon Musk announces a $100M prize for “best carbon capture technology”. This seems useful compared to many things he does.
    • My view: The obvious answer is trees. But that is not marketable to Elon, so I suggest we re-brand trees as ArborLoop.(TM)

Road traffic almost back to pre-COVID levels as commuters shun public transport | Sydney Morning Herald

Andrew Taylor wrote a comprehensive review of the data on Traffic and Transit use in Sydney for the Sydney Morning Herald in Road traffic almost back to pre-COVID levels as commuters shun public transport. My quotes below:

Professor of transport engineering at the University of Sydney David Levinson said the nature of public transport usage meant it was affected by lockdown more than other modes.

“There is a lot less travel to work for office workers – especially CBD-based office workers – which hits trains pretty hard, since the rail network radiates from the Sydney CBD,” he said. …

The lack of tourists and students had also reduced travel compared to previous years, Professor Levinson said. “I don’t think congestion will go all the way back to normal as restrictions ease, the rise in working from home and deliveries in lieu of shopping and eating out will be at least somewhat irreversible.” ….

Professor Levinson said the COVID-19 crisis would have long-term effects on peak-hour traffic as well as public transport usage.

He also said road and rail projects not already under construction should be reconsidered: “The extent to which travel demand changes (like working from home) are permanent should change where the most important investments are.”

Professor Levinson said a shift towards working from home would encourage more people to live further from the inner-city “and probably favour the auto over public transport, as autos provide greater flexibility, and can reach places that frequent public transport cannot”.

Commuters continue to shun public transport, with Opal data showing patronage down more than 40 per cent on pre-COVID levels. CREDIT:SAM MOOY [Sydney Morning Herald]

What if it’s only once a week?

I attended a real full-on meeting in person yesterday for only the third time in the last year. It was at the CBD offices of a large organization. Aside from us and reception, no one was there. I assume everyone was working offsite.

This has devastating consequences for commercial real estate and public transport. Though there is some elasticity (if some firms retrench on space, others might relocate, if some passengers don’t travel to work, other trips that are no longer squeezed out might come back) obviously it’s going to hurt. Also commute travel will be less peaked. This has follow on consequences for residential preferences and will lower relative house prices near the CBD.

At the meeting we talked about the what the after COVID period looks like. For CBD office workers, the consensus expectation was people going back to the office about 3 days per week. (Recognizing about half the workforce isn’t office workers and that most office workers are not CBD based in most cities).

But it occurs to me that we may only see people coming in once a week for the group meeting, or client / vendor meetings. While in person coordination won’t drop to zero, and people are social, they may not need to be that social with those people. As digital tools get better, we will face less and less need to coordinate work in person. And while today’s software and networks were sufficient for most organizations to muddle through the pandemic, these tools are only getting better over time.

I’ve described the Futurist Fallacy before. In the future everyone will live and behave as the futurist does now. (Btw, for the record I still go into work except when I am in mandated annual leave) This fallacy is a form of projection. So we need to be careful.

But the pandemic is the largest shock to the social and economic system of most developed countries since World War II. What comes out on the other side of this will be different. Preexisting arrangements about requiring in person supervision to achieve productivity have been falsified.

So, suppose we do only have office workers meeting weekly instead of daily or 3 days a week?

For those markets (more than half of CBD oriented transit). Transit demand may be off more than 80% since now parking is easy, and there isn’t enough demand from non work trips to compensate.

Office demand collapses. Maybe some suburban firms relocate to the city to get a taste of those local agglomeration economies, but those will largely disappear … because … there will be no spontaneous in person interaction or serendipity anymore. Everything will have been scheduled.

People will regain 4 or more hours per week. I say “4” because the average commute is an hour round trip, now one day instead of 5, I say “or more” because congestion will dissolve with so many fewer commuters, and that’s another third or so of travel time depending on where you are. Commuters may use some of that on relocating regionally or to the exurbs, increasing distances.

Investment strategies have yet to acknowledge this possibility, and everything is being built as if the before times will be restored. That’s certainly possible, but it’s also possible we are far off on the wrong track with new infrastructure construction and real estate development.

Thanks to technology (rail, elevators, air conditioning, etc) we now can support cities that would have been unthinkably large two centuries ago. But thanks to other technology (broadband internet, wireless, computers, software) we don’t have to.

The deployment phase of this transition from the twentieth century city to twenty-first century new spatial pattern will take decades to achieve, and in the meantime new technologies will emerge (vehicle automation, drones, aerial vehicles, things we don’t yet take seriously) which will have either centralizing or decentralizing effect (I’d bet the latter, but not all my money). The office based CBD has peaked in significance. The consequences will be felt for the rest of our careers.

Longing to Travel: Commute Appreciation during COVID-19

Recently published:

  • Aoustin, Louise, and David M Levinson. 2021. “Longing to Travel: Commute Appreciation during COVID-19.” Findings, January. [doi].


Based on a survey of 197 Sydneysiders undertaken during the COVID-19 Lockdown, this study shows time spent in transport was missed the most by public transport users, followed by push bike users, e-bike users, pedestrians, and finally drivers. Men missed time spent in transport more than women. It also finds that for public transport users, the more transfers, the less they miss time spent commuting.

You might also be interested in this recent article in The Washington Post about a similar question.


Transport Engineering and Planning at the University of Sydney

Emily MoylanJan 8

Welcome to the start-of-year update from TransportLab at the University of Sydney.

We’re looking forward to meeting people face-to-face again, but in the meantime, you can follow us on Twitter or LinkedIn.


The TransportLab Seminar started up again in the second half of 2020. In addition to student presentations, the virtual format allowed us to host several academic and industry speakers, including:

  • Meead Saberi (UNSW) on Strategic Modelling for Walking Infrastructure
  • Jonathan Busch (SCT Consulting) on a practitioner’s perspective on transport innovation
  • Emilie Gunaratnam and Matthew Jones (TfNSW) on Cost-Benefit Analysis at TfNSW
  • Sue McNeil (University of Delaware) on using real-time time data to provide situational awareness to first responders in emergencies. 
  • Gabriel Metcalfe (Committee for Sydney) on Advocacy for Change in Mobility Systems

TransportLab participated in the annual Transport Research Association for NSW (TraNSW) Symposium on 17-19 November. Six TransportLab students presented their work, and Linji Chen (fourth from left) received an award for Best Research Demonstration.

Together with the UC Berkeley Institute for Data Science, University of Sydney organised a virtual symposium on Big Data and Neighborhood Change. Somwrita Sarkar was a project lead on the event. 


Teck Kean Chin has started his PhD on Smart City Applications in Land Use and Transport

Yang Gao will start a PhD in early 2021.

Mengyuan (Derek) Zhu has started his MPhil on Optimising Space-time Matching in Ridesharing through Predictive Modelling

Former post-doc Mengying Cui started as an Associate Professor at Chang-An University

Jing Chen and Louise Aoustin finished as a visiting scholars with TransportLab at Sydney. 

  • Louise graduated from EPFL and is currently working at Keolis Downer in Sydney. 
  • Jing Chen will finish her PhD at Southeast University in Nanjing, China

Awards, Accolades and Memberships

Jennifer Kent’s book “Planning Australia’s Healthy Built Environments” was awarded a Commendation Award for Planning Excellence in the category of Cutting Edge Research and Teaching by the Planning Institute of Australia, NSW.

The University of Sydney has the most influential academics of any university in Australia. This includes David Levinson who is in the top 2% of citations. 

Emily Moylan received the 2020 Dean’s (Faculty) Award for Teaching Innovation in 2019 for her work incorporating informatics into the transport curriculum.

Mohsen Ramezani was promoted to Senior Lecturer.

Mohsen Ramezani has received an ARC DECRA on Market Design of Next Generation of Shared and Automated Transport Services

PhD student Linji Chen won the Best Research Demonstration Award at the TraNSW 2020 Symposium for his presentation on Decentralised Cooperative Cruising of Autonomous Fleet.

Honours student April Alcock is the 2020 University of Sydney winner of the ITE-ANZ Trafficworks Student Award.

Articles, books and chapters

Davis, Blake, Ji, Ang,  Liu, Bichen, and Levinson, D. (2020) Moving Array Traffic ProbesFrontiers in Future Transportation. doi: 10.3389/ffutr.2020.602356 [doi]

Ji, Ang and Levinson, D. (2020) Injury severity prediction from two-vehicle crash mechanisms with machine learning and ensemble models. IEEE Open Journal of Intelligent Transportation Systems. [doi]

Ji, Ang and Levinson, D. (2020) An energy loss-based vehicular injury severity model. Accident Analysis and Prevention. 146 October 2020, 105730. [doi]

Kamal, M. A. S., Ramezani, M., Wu, G., Roncoli, C., Rios-Torres, J., & Orfila, O. (2020). Partially Connected and Automated Traffic Operations in Road Transportation. Journal of Advanced Transportation. [doi]

Kent, J.L. (2020). The role of car-sharing in sustainable transport. In Curtis, C. (Ed.) Handbook for Sustainable Transport. Edward Elgar, Cheltenham

Kent, J.L. and Thompson, S. (2020) Healthy Cities. In Rogers, D., Keane, A., Nelson, J. and Alizadeh, T. (Eds.) Introduction to Urbanism: An Interdisciplinary Approach. Palgrave McMillan, Camden

Kent, J.L. (2020) Transport, access and health. In Mulley, C. (Ed.) Urban Form and Accessibility. Elsevier, London

Levinson, D. (2020) Logistic Curve Models of CO2 Accumulation. Transport Findings. [doi]

Levinson, D. (2020) A Timeline of Future Transport in Sydney as Revealed in Tablet Form. In Derrible, S. & Chester, M. (Ed.) Urban Infrastructure: Reflections For 2100: An Edited Volume Imagining Infrastructure Transitions And Goals At End-Of-Century. Independently published.

Paine, G., Thompson, S., Prior, J., Connon, I., & Kent, J. L. (2020). Bringing History Forward: Learning from Historical Context when Translating Contemporary Health Evidence into Planning Practice. Journal of Planning History. [doi]

Sarkar S., Farid R. (2020) Data, Science and Cities. In: Rogers D., Keane A., Alizadeh T., Nelson J. (eds) Understanding Urbanism. Palgrave Macmillan, Singapore. [doi]

Reports and Projects

Liverpool Sustainable Urban Mobility Study. iMOVE CRC/Liverpool City Council

The Transport Access Manual was published by the Committee for the Transport Access Manual, chaired by David Levinson 

Opportunities to build capability in traffic management for Austroads. Dissemination webinar is on the 21st January

New housing supply, population growth, and access to social infrastructurefor Australian Housing and Urban Research Institute (AHURI)


Transport Findings has been renamed Findings, and a new section Urban Findings, will launch in 2021

Mohsen Ramezani is editing a special issue of Transportmetrica B: Transport Dynamics on Advanced Modeling and Control for AI-enabled Multimodal and Automated Transport Systems.

Mohsen Ramezani is editing a special issue of Frontiers in Future Transportation on Integration of Real-Time Information in Transport Planning and Operations.


Jennifer Kent presented at the Sydney Environment Institute’s Critical Companion Series on Sustainable Urban Mobility

David Levinson keynoted at IARAI  Traffic4cast 2020 Special Session: The End of Traffic and Future of Access

David Levinson guest lectured at UBC: The New, New Normal

David Levinson presented at Australia Build Week: 30-Minute City

David Levinson presented at Festival of Urbanism: The New New Normal: Mobility and Activity in the After Times. The article was originally published in The Fifth Estate, November 2, 2020.

David Levinson was a panelist at Transport Australia Society webinar: The Role of Walking in the Movement and Place Framework

David Levinson was quoted by the Daily Telegraph: Outer Sydney Orbital: M9 to be Built Next to Airport Metro, Freight Line

David Levinson was quoted by the Sydney Morning Herald: “Sydney’s bike network stuck in the slow lane“.

Emily Moylan was an invited speaker at the CSIRO Symposium Future of Meetings

Emily Moylan spoke at the Australasian Association for Engineering Education Conference about civil engineering students coding to learn.

Mohsen Ramezani was invited to be a speaker in the full-day online workshop “Emerging Control of Vehicular Traffic for Improving Sustainability and Energy Efficiency” for SICE Annual Conference 2020

Traffic Management Training

Austroads has commissioned eight online learning units from ARRB and the University of Sydney that cover the fundamentals of traffic management. (I created Units 5 and 7). The units cover 22 modules, each includes a video with in-session exercises. Some modules include additional tutorials. Paul Bannett of ARRB will be presenting the dissemination webinar on the 21st January. If you wish to tune in, you can register here: 

Unit 1: Introduction to Traffic Management
1-1 Introduction to the Learning Modules and Objectives and Principles of Traffic Management

Unit 2: Traffic Behaviour and Traffic Theory Fundamentals
2-1 The Stochastic Nature of Traffic Behaviour
2-2 Fundamental Speed-flow-density Relationships
2-3 Fundamental Microscopic Relationships

Unit 3: Transport Study, Traffic Data and Analysis Methods
3-1 Transport and Traffic Data
3-2 Traffic Analysis Concepts
3-3 Capacity Analysis

Unit 4: Transport Operations Control Strategies and Systems
4-1 Objectives and Principles of Transport Operations
4-2 Signalised Intersections – Operations and Control Strategies
4-3 Unsignalised Intersections – Stop, Give Way and Roundabouts
4-4 Overview of Traffic Management Centres

Unit 5: Network Operations Planning
5-1 Network Operations Planning Accessibility
5-2 Network Operations Planning Process
5-3 Road Space Allocation and Road Use Priority
5-4 Movement and Place

Unit 6: Network Performance Monitoring and Management
6-1 Network Performance
6-2 Traffic Congestion and Management
6-3 Traffic Incident and Event Management
6-4 Traffic Modelling

Unit 7: Safe System Approach to Traffic Management
7-1 The Safe System Approach

Unit 8: Intelligent Transport Systems
8-1 Intelligent Transport Systems for Traffic Control
8-2 Managed Motorways – Operational Principles, Managed Motorway Toolkit
Give way to Pedestrians

Traffic4cast 2020 Special Session

IARAI is organising Traffic4cast 2020 Special Session on FridayDecember 11, 2020. This is a virtual-only event to be held on Zoom. It includes 2 sessions:

  • Session 1: 9:00-13:00 CET (0:00-4:00 PST)
  • Session 2: 17:00-21:00 CET (8:00-12:00 PST)

Traffic4cast Special Session features an in-depth discussion of the Traffic4cast 2020 core competition results. The goal of the competition is to predict traffic in multiple big cities of different culture and economy based on industrial-scale real-world traffic data.

Traffic4cast has been again selected this year for the prestigious Neural Information Processing Systems (NeurIPS), the leading event in machine learning. The summary of the competition is presented at the NeurIPS competition track on Friday, December 11.

Traffic4cast Special Session will be opened by an introduction by our very own Sepp Hochreiter and two keynote lectures by David Levinson [I will be talking about The End of Traffic and the Future of Access] from  University of Sydney, Australia and Razvan Pascanu from Deep Mind, United Kingdom. Three top-ranked teams in the core competition leaderboard and four contributions selected by the Traffic4cast Scientific Committee will present their work for discussion.

Video of the full Special Session. My presentation starts at 14:57

Australia Build Week

I will be taking part at the Transport Stage at #Australiabuild Week Online, taking place between Dec 7-11th.  I am speaking on Thursday, December 10, at 14:40 AEDT on “The Thirty-Minute City”.

You can register your complimentary tickets at

And check out other presentations that will be taking place throughout the week at