Talking Headways Podcast: Complicated Measures and Public Policy

I was interviewed by Jeff Wood while at TRB for the Talking Headways Podcast. It is now available for download.

  • Talking Headways Podcast: Complicated Measures and Public Policy (Part 1)
  • Talking Headways Podcast: Unnecessary Literature Reviews, Part II (Part 2) [Transcript]
Talking Headways:
Talking Headways:

Should Australia ban petrol cars?

I have some comments on the University of Sydney release: Should Australia ban petrol cars?

A necessary step towards a zero-carbon economy”

Britain has announced it will ban petrol and hybrid cars by 2035. University of Sydney experts comment on whether Australia can and should adopt a similar policy.

Move towards electric vehicles achievable in Australia

Professor David Levinson is an academic from the University of Sydney’s Faculty of Engineering who specialises in transport engineering and believes a similar move would be achievable if implemented in Australia.

“It is great to see Britain taking the lead on banning new internal combustion engine vehicles, which is essential to reduce pollution and carbon emissions. Electrification of the vehicle fleet is coming, as the technology for electric vehicles (EVs) and batteries steadily improves and costs drop with scale, and policy can accelerate the change,” said Professor Levinson.

“It is expected that EVs will be less expensive to buy and operate in the next few years. Unfortunately, Australia has been unwilling to move aggressively on EVs, despite the almost unlimited sunshine providing the opportunity for truly inexpensive renewable power,” he said.

“Cleaner vehicles will make cities smell nicer, less noisy, and overall more pleasant to be in.

“One issue that needs more attention is the charging infrastructure. While people with garages at home can install chargers, those living in multi-family housing and parking on-street will need convenient charging locations. As EVs get more widely deployed, cities will need to build more charging facilities, and petrol stations will need to adapt.”

There are some comments from other University of Sydney academics in the full article.

Toyota iRoad one-passenger concept cars, image courtesy Toyota.
Toyota iRoad one-passenger concept cars, image courtesy Toyota.


To get our cities moving again, we need a new kind of urban professional

In this extract from my new book The 30-Minute City,  I argue that in designing our cites, we need ‘Urban Operations’ experts who can straddle the realms of both strategy and tactics. Reprinted from Foreground

Access is the driving force behind how cities were built – which is to say, cities developed with the goal of making it as easy as possible for people to reach the opportunities and activities contained within them. In the contemporary city, though, the professionals tasked with designing and developing our cities for access can often seem to be working at cross-purposes.

Our engineers are trained in engineering school to ‘do it right.’ They are trained intensively in calculations to make sure the math works out. This is very important: structural engineers do not want to misplace a negative sign or they would build the bridge upside down. In contrast, our planners retort to the engineers ‘do the right thing.’ What are the right values? And that’s really important, too. Meanwhile, our public citizens say: ‘do the right thing right’, synthesising this apparent conflict.

The 30-minute City: Designing for Access, David Levinson

In designing and managing our cities for access, we need to think about both strategy and tactics. We need to think about ideas and implementation. For instance, at train stations with entrances on only one end of the platform, the objective of enabling people to leave the station is supported, but not the broader objective of enabling them to reach their destinations in the least amount of time. Traffic signals presently are timed to minimise delay for vehicles, but not for people, and fail to count vehicle occupancy (buses wait in the same traffic as cars) or pedestrians.

“Traffic signals presently are timed to minimise delay for vehicles, but not for people” – David Levinson @trnsprtst


A Nihilistic Theory

I’m going to introduce a ‘nihilistic’ theory of transport and land use: everything is ‘pointless.’

Transit facilities are pointless. A station is not a point, it is a place.

Junctions are pointless. A junction, or intersection, is not a point, it’s a space. It has conflict points, which are also spaces, but it takes time to traverse, and those traversing it take up space.

Cities, too, are pointless, and yet planners often abstract away important details – as in the Greater Sydney Commission’s Metropolis of Three Cities plan, which, like so many regional plans, has dots on maps to represent whole communities.

Everyone working in the urban sphere should recognise this ‘pointlessness.’

Just as small spatial relations matters, so too does time. Do small amounts of time savings matter? Yes! Absolutely!

A traffic engineer proposes a change that will save somebody five seconds, and someone inevitably retorts that nobody cares about five seconds. But we can never get to larger time savings (or accessibility gains) when we’re always talking about how unimportant the small changes are. There is no way to save 15 seconds if you don’t save five seconds. There is no way to save 30 seconds unless you save 15, or one minute unless you save 30 seconds, or five minutes unless you save one minute.

Trips comprise many time elements, and use many bits of the transport network, and we are not going to save time all at the same place or with the same project or process. So the better practice is to take the gains that are possible, as they will accumulate over time. Saving time, or increasing speed, increases the area that can be covered in the same amount of time, and since accessible area increases with the square of the radius, time savings have disproportional effects on accessibility.

This argument applies to all modes. The traffic signal engineers use it to justify their signal timings for automobiles. The potential flaw here is not in saving time, but in doing so at the expense of pedestrians and the neighbourhood at large.

There is the argument that time, unlike money, cannot be ‘saved’, as there is no way to store it. And of course there is an element of truth there. But I would argue that time can be used for things that are valued more highly than standing at an intersection waiting to cross – which is to say, anything else. The time not spent waiting at the intersection might be spent in a more pleasant environment, or walking or riding farther to a slightly better or higher paying job, or a shop with somewhat better goods, or from a slightly better or less expensive home. These are the trade-offs people make all the time, and by increasing the area that can be traversed in a given amount of time, we increase opportunity and choice.

A profession that is interdisciplinary in real time – or, doing the right thing right

To do the right thing right, we want to forge a new profession that is interdisciplinary in real time. Planners create long-term plans covering large areas – they, at least in theory, aim to optimise for all of society. Analysts develop policies over large areas, which have a shorter-term time horizon, and also should at least consider all of society. But the local-looking professions – engineers, architects, urban designers, and technicians of various kinds – whether they are involved in building for the long-term or managing and operating the system in the short-term, by definition optimise locally, for the site, rather than the city. How the site interacts with the city is neglected.

We need a profession not of more urban planners, nor of more transport engineers, but urban operators – people engaged in today’s city, not tomorrow’s, but who can optimise for the system as a whole (that is, by thinking about accessibility) and not just their small piece of it.

The world is changing ever-faster. Yet strangely, today’s professionals undertake and celebrate very long-term plans where they acknowledge the existence of a problem (i.e. congestion), and technology (i.e. autonomous vehicles), but don’t acknowledge that anything changes.

Instead, we should forge new urban operators as a strong alloy of planning, engineering, economics, and design. Urban operators take ideas in real time and solve today’s problems with resources on-hand, rather than solving imagined problems that bring distant dangers near. We have enough problems today. We also have solutions available to us today, and we don’t implement them. And yet people are employed to work on 40-year plans.

“We need a profession not of more urban planners, nor of more transport engineers, but urban operators – people engaged in today’s city, not tomorrow’s, but who can optimise for the city as a whole” – David Levinson @trnsprtst

Today’s disciplines are excellent for admiring and nurturing today’s problems, but not nearly so adept at solving them. Engineers and planners are so focused on the long term, their jobs effectively require them to build it and then abandon it. Operating and maintaining the system is someone else’s responsibility. Once they have made their design they hand it over to a contractor for construction, who then hands it over to the client.

And then we have people who are making microscopic decisions without thinking about the big picture. Where do you put the bus stop relative to the train station? This affects accessibility, but the decision is made based on what is convenient for the bus operator rather than passengers, or worse, to minimise delay for cars.

As Bill Garrison argued, we want people who can bridge the hard and the soft – the hardware engineering of infrastructure and vehicles and the software of management, control, and financial systems.

Bridging or merging the soft and the hard would vastly improve policy and policy-making processes. We should be able to simultaneously think of engineering and policy, not be restricted to engineering or policy. Those of us in the transport field should identify as transportists – not transport engineers or transport planners or transport economists. The problem must come before the mechanism of solution.

We want people who can bridge the site and the city. People who think about the position of a train platform in the greater context of the metropolitan area, so that people living on the south side of the platform can easily reach it, rather than semi-circumnavigating the train station to its only entrance on the north.

We want a fusion of planners and engineers who would focus on the ends not on the means, who can think in multiple scales and multiple time horizons.

The goal of the 30-minute city aligns with travel time budgets and human behaviour. We know that, historically, land developers and the railway builders were keen on the idea of a feasible commute, and they were keen on this idea when they deployed tram and train networks and concomitantly subdivided large tracts into lots and built homes that were within a 30-minute commute of the central city.

Lower case ‘d’ design

Architects are famous for BIG design ideas. But cities are not amenable to big designs any more. They grow (and should grow) incrementally, not comprehensively. So instead let’s talk about what I will call “lower case ‘d’ design,” the humble design decisions about where to put bus stops relative to station entrances, and how to time traffic signals. These are small urban design decisions that don’t get sufficient attention.

There are many things that we can do that involve rethinking the details – like adding train station gates to both ends of platforms to expand catchment areas, and thus patronage. Details like stop spacing and location, practices like all-door boarding, payment before boarding, optimising timetables and frequency, may just squeeze a few seconds per stop or minutes per route out of the existing configuration, but collectively they greatly expand people’s accessibility.

More strategically, this requires thinking about transport and land use balance. Offsetting today’s imbalance can give us growth without additional travel or commuting-related congestion. To achieve a 30-minute city, cities need to put new jobs in housing-rich areas and new housing in job-rich areas systematically as a way of growing. This contrasts with local government’s desire to focus employment in the central city, and developers who will tend to put more housing in the outer suburbs where there are many fewer jobs.

And we need to design for the cities we want, not ‘predict and provide’ for the city we forecast. Our future cities cannot be delivered by the same disciplinary thinking that created the cities we have.

This is an abridged extract taken from David Levinson’s book The 30-Minute City: Designing for Access, available here in PDF format and here in print.

Levinson joined the School of Civil Engineering at the University of Sydney in 2017 as Foundation Professor in Transport Engineering. He conducts research on accessibility, transport economics, transport network evolution, and transport and land use interaction.

Transportist: February 2020

Welcome to the latest issue of The Transportist, especially to our new readers.  As always you can follow along at the or on Twitter.

We skipped the January Transportist Newsletter this year (confusing future archivists), and in its place we launched the TransportLab Newsletter. Most of you should have received that, but they are separate mailing lists.

I also attended TRB along with students and colleagues from TransportLab, and then keynoted at Transforming Transportation at the World Bank/WRI, so happy to meet many of you in person again or for the first time.

Sydney has been smoky, with poor air quality, but managed to avoid the brunt of the Australian bush fires which devastated other parts of Australia. Despite reports on US media, the whole continent is not on fire, though something like 6% of the state of New South Wales did burn, destroying far fewer than 6% of houses. It does look like the end-times though.

Book: The 30-Minute City: Designing for Access.

I am pleased to report that  The 30-Minute City: Designing for Access is now available for purchase

The book reads fast, with just over 20,000 words, and contains 50 images and 6 tables.


This book describes how to implement The 30-Minute City.  The first part of the book explains accessibility. We next consider access through history (chapter 2). Access is the driving force behind how cities were built. Its use today is described when looking at access and the Greater Sydney Commission’s plan for Sydney.

We then examine short-run fixes: things that can be done instantaneously, or nearly so, at low budget to restore access for people, which include retiming traffic signals (chapter 3) and deploying bike sharing (chapter 5) supported by protected bike lane networks (chapter 4), as well public transport timetables (chapter 6).

We explore medium-run fixes that include implementing rapid bus networks (chapter 7) and configuring how people get to train stations by foot and on bus (chapter 8).

We turn to longer-run fixes. These are as much policy changes as large investments, and include job/worker balance (chapter 10) and network restructuring (chapter 9) as well as urban restoration (chapter 11), suburban retrofit (chapter 12), and greenfield development (chapter 13).

We conclude with thoughts about the ‘pointlessness’ of cities and how to restructure practice (chapter 14).

The appendices provide detail on access measurement (Appendix A), the idea of accessibility loss (B), valuation (C), the rationale for the 30-minute threshold (D), and reliability (E). It concludes with what should we research (F).

Table of Contents

  • Preface
  • 1  Introduction 15
  • 2  The 30-Minute City: Then and Now 19
  • 3  Traffic Signals 25
  • 4  On the Four Paths 29
  • 5  Bikesharing 35
  • 6  Timetable 37
  • 7  Rapid Bus 39
  • 8  Interface 45
  • 9  Gradial: Or the Unreasonable Network 51
  • 10  Job-Worker Balance 55
  • 11  Urban Restoration 59
  • 12  Retrofit 69
  • 13  Greenfields and Brownfields 75
  • 14  A New Profession: Urban Operations 81


  • A  Theory 89
  • B  Accessibility Loss 93
  • C  Access Explains Everything 95
  • D  Why 30 Minutes? 97
  • E  Reliability 99
  • F  Research Agenda 101

Master of Transport at the University of Sydney

  • Classes start in early 2020, apply now for term 2.

Transport Accessibility Manual

  • The Committee of the Transport Accessibility Manual met at the Transportation Research Board Annual Meeting in Washington DC in January.
  • We discussed the first (preliminary) draft of the document, which was distributed to mailing list members before the meeting. Contact me directly if you would like to be added to the mailing list.

Talks and Conferences

  • I will be in Auckland, New Zealand for the IAEE – International Association of Energy Economics conference 12-15 February 2020. My talk will by on Friday Feb 14 1:40 – 3:20.

    Dual Plenary 4: Energy Transition in Transport | Chair: Professor Frank Jotzo, The Australian National University . Speakers: Professor David Levinson, University of Sydney, Dr Amela Ajanovic, Vienna University of Technology, Dr Selena Sheng, University of Auckland.
    (OGGB3 | 260-092)



Transport Findings

Transportist Blog

News & Opinion

Australian Expression of the month:


Moving Forward Framework (take 2)

The previous version of this post was eaten by WordPress.

Moving Forward Framework: For the People
Moving Forward Framework: For the People

A reader writes: “The U.S. House [Transportation and Infrastructure Committee] came out with its pre-election transportation policy: The Moving Forward Framework, and access measures made it into what is otherwise a high-level policy document (with no hint about how they plan to pay for their wishlists.)”

On Access to Jobs:

It’s infrastructure investment that is smarter, safer, and made to last – with a framework that:

  • Ensures a transportation system that is green, affordable, reliable, efficient and provides access to jobs


Modernizes Project Planning – Requires States and MPOs to prioritize transportation access and to consider during the planning process all system users, job access, connections to housing, and creation of transportation options in underserved communities.

On Fix-It-First

Revamps Existing Formula Programs

Amends core highway formula programs to prioritize investments and improve program implementation:

Fix it First – Prioritizes maintaining and improving existing infrastructure and bringing it up to a state of good repair, including roads, bridges, tunnels, and ferry systems.

On Road Pricing

Tackles Congestion Equitably – Institutes tighter standards around tolling and congestion pricing.

Tests the Viability of New Transportation User Fees

Transforms revenue collection and distribution by authorizing a multi-year national pilot program to test revenue collection to ensure the future viability and equity of surface transportation user fees, including a vehicle-miles travelled fee.


It’s almost as if it were written by a reader of this blog.

IAEE Asia Oceania Conference 2020: Energy in Transition

Screen Shot 2020-01-23 at 06.22.00I am pleased to be speaking at IAEE Asia Oceania Conference 2020: Energy in Transition in Auckland February 12 – 15. My talk will by on Friday Feb 14 1:40 – 3:20.

Dual Plenary 4: Energy Transition in Transport | Chair: Professor Frank Jotzo, The Australian National University . Speakers: Professor David Levinson, University of Sydney, Dr Amela Ajanovic, Vienna University of Technology, Dr Selena Sheng, University of Auckland.
(OGGB3 | 260-092)

The 30-minute city: Small decisions for big gains

This article is adapted from the book The 30-Minute City: Designing for access, and is about those small local decisions that are often overlooked as planners and engineers focus on major infrastructure policies and programs.

Cities are organised so that many people reach one another in a short amount of time. Residents reach other people, places, goods, and services on foot, or by bike, bus, train, ferry, or car. People don’t need planes or very fast trains to travel between places within a city, even though planes and very fast trains are faster than walking, bikes, buses, trains, ferries and cars. Cities optimise what people can reach in a given amount of time, in the face of modest speeds.

We see this when we compare the average speed of travel inside Sydney — about 30 km/h by car after considering traffic signals and congestion — versus the 100 km/h that they can travel on rural highways. Rational people pay dearly to live in Sydney or Melbourne compared to rural Australia. This is not a criticism of Sydney or Melbourne. Despite their extreme congestion the access these and other great cities provide, and the value of that access, is what makes cities great.

Accessibility measures how many potential destinations (jobs, workers, stores, doctors, etc.) someone can reach from a particular point in a given travel time (say 30 minutes) by a particular mode at a certain time of day. The cumulative opportunities measure of accessibility is like the meter or kilogram in the metric system, it means the same thing regardless of where you are. We can talk to a politician and show her how many jobs can be reached from a location in 30-minutes by transit at 8:00 a.m., and we can compare that number with any other point, where the accessibility may be higher or lower. We can compare Los Angeles and San Francisco, or compare Los Angeles in 2019 with Los Angeles in 1973.

In Australia, the ’30-minute city’ has been adopted by the Greater Sydney Commission, the planning agency for the Sydney region, as a centrepiece of its 40-year plan. The aim is that residents of Sydney can reach one of three important regional centres in less than a half-hour by walking, biking, or public transport. Doing so requires the thoughtful application of knowledge at hand, using modes of transport technology that have been around for more than a century. This includes wise choices about big investments in subways or elevated highways, and intelligently making small decisions about streets, intersections, and transit stops.

Not everyone works, or needs to get to or from work in a half hour. Different places have proposed 5-, 10-, 15-, and 20-minute neighbourhoods as well. For instance, the ‘pint-of-milk test’ (in New Zealand, the beverage in question isn’t milk) asks whether you can purchase a pint of milk within a 10-minute walk of your home; and a modified version of that test asks if can you do it at a place that doesn’t also sell petrol.

The related concept of a 20-minute neighbourhood is about ‘living locally’ by giving residents the opportunity to access all the services they need with a 20-minute round-trip walk, cycle, or public transport trip. While the 30-minute city tends to focus on work and includes travel by motorised vehicles, these other tests ask about life’s other activities and emphasise non-motorised travel. If you can walk to a pint of milk or the local hotel or pub within 10 minutes, and get to your major services within 30 minutes, you are doing better than the average 62-minute trip now experienced by public transport-riding Sydneysiders.

Access and time

One way to examine accessibility is to measure how much additional accessibility some project or new service will provide. Another way is to examine how much accessibility is lost because an improvement has not been made.

Prospect Theory teaches that we feel losses more significantly than gains. So, for example, if destinations reachable in 30-minutes is considered to be 100 percent of accessibility, if you lose 10 minutes of time out of that 30 minutes (because of delay or circuitous routes arising from poor system design), you lose more than half of your accessibility.

That loss is so large because accessibility increases as a non-linear function of time. The area of the accessibility ring from 20 to 30 minutes (blue in Figure 1) is much larger than from 0 to 10 minutes (green), or 10 to 20 minutes (red). If we lose five minutes, we lose 30 percent of our accessibility, as shown in Figure 2. Every second counts.

Even if a policy or design sacrifices only 30 seconds, this extra delay costs people not only their travel time, but a sacrifices opportunities they could have reached within that travel time.

Figure 2: Estimate of % accessibility loss per minute of excess travel time

In busy and crowded cities decisions are made routinely about features of neighborhood streets and transit stops. These might appear to engineers and planners to influence the quality of life in the nearby community. The accessibility framework shows us that they also change our ability to benefit from more distant opportunities by enhancing or reducing access to the entire region. I demonstrate this by presenting three rather different examples of apparently modest urban design and traffic engineering decisions that affected regional accessibility.

The following sections address accessibility changes resulting from the placement of exits at a transit station in Sydney, the design of a Bus Rapid Transit (BRT) line in Minneapolis, and the timing of traffic signals at almost any urban location. While quite different from one another, each illustrates an apparently local decision that turns out to be surprisingly important to people who want to access the entire community.

Access to train platforms

Sydney’s 813 km commuter train system is one of the best in the world, providing high frequency service from many suburbs to central Sydney. However, 44 of 175 stations have entrances at only one end of the platform. A traveler wanting to board or alight the train from the other end of the platform has to walk parallel the station the full length of the platform, which given the length of trains, makes for a 2 minute walk. Figure 3 maps access to Erskineville station. Passenger catchment areas are drawn around the station in 5, 10, and 15 minute bands of walking time. In 2016, some 1,389 people live within a five-minute walk (about 400 metres) of the station platform.

Figure 3: Accessibility comparison of Erskineville station

A long history of research and common sense tell us that people who live near stations are more likely to use public transport than those living farther away. With the current configuration, many people live or work on the wrong end of the platform. The person unfortunate enough to be traveling between two stations both having gates on only one end of the platform averages an extra 3.25 minutes each way, due to the mismatch of their entry and exit locations with their origins and destinations, compared with the ideal case where everything is aligned conveniently.

Erskineville is among the most extreme cases in the Sydney Trains system; 5-minute accessibility to the platform by residents would increase by 89% if a southern entrance were added because there are relatively new, very large, apartment blocks near the southern end of the station. The number of people and jobs within 5-, 10-, and 15- minute walks of the station translates into riders, which translates into land value, which translates into real estate tax revenue. The increase in ridership resulting from an added entrance could produce sufficient revenue that it could pay for itself.

There also is an equity rationale for adding the additional entrance. Modern standards for providing access for the disabled require elevators to serve each platform. Elevator installation should be coordinated with additional entrances.

This kind of investment could be made for most of the other 43 stations with similarly lopsided configurations, for which there are not yet plans. And of course, this logic could be applied for similarly configured transit stations in other cities around the world. This is easily-picked low-hanging fruit to improve accessibility that can be done now.

Rapid bus

The Minneapolis – St. Paul A Line, which opened in 2016, is an arterial rapid bus line from suburban Rosedale, providing access to the Green Line and Blue lines of the light rail transit (LRT) system. The A Line is the first rapid bus line in a network planned by the Twin Cities region MetroTransit. Several of its features save a few seconds of time for each passenger at each stop.

  • Prepay: Passengers tap on before boarding the bus. This saves 1.5 – 6 seconds per boarding passenger vs. tapping or paying in cash and coins.
  • All-door boarding: Passengers can board at any door not just the front. This halves the boarding time.
  • Fewer stops: There is longer spacing between stops. Stop spacing increased from about 200 m to about 800 m. This results in less dwell time at stops and less acceleration-deceleration delay.

The faster turnaround results in higher frequency of service with the same number of buses and hours of driver time, so the service is more productive. It’s a win-win change for almost everybody, except for some people who have to walk longer distances to get to or from the nearest bus stops. Figure 4 maps the winners and losers in terms of accessibility, and clearly shows there is many more winners (green) than losers (yellow ). Overall this rapid bus configuration increased accessibility by 5 percent for people who live near the A Line. We could redesign bus networks in many cities to increase access to destinations.

The faster turnaround results in higher frequency of service with the same number of buses and hours of driver time, so the service is more productive. It’s a win-win change for almost everybody, except for some people who have to walk longer distances to get to or from the nearest bus stops. Figure 4 maps the winners and losers in terms of accessibility, and clearly shows there is many more winners (green) than losers (yellow ). Overall this rapid bus configuration increased accessibility by 5 percent for people who live near the A Line. We could redesign bus networks in many cities to increase access to destinations.

Figure 4: Change in number of jobs within 30 minutes by transit

Rethinking traffic signals

Everyone is familiar with traffic signals. They operate in phases, with green lights given to alternating directions (for instance north-south, then east-west). Traffic signal engineers don’t normally think about accessibility the way I present it here. They think about traffic signal time and vehicle delay.

A traveller in a car approaching a signal faces a cycle which comprises a red indicator, a green indicator, a yellow indicator, and perhaps an all-red period (during which signals are red in all directions). Imagine that a car arrives at this intersection when the light is red. It waits for the red light to change to green before it moves on. There’s a delay associated with that. If cars arrive randomly, the average stopping time is half the red time, and the probability of stopping is the likelihood of arriving when the light is red (the ratio of red time to the duration of the whole cycle).

Traffic engineers apply the same treatment to pedestrians, but the times differ because people on foot systematically get less green time than people in cars. It takes longer to walk across the street than to drive across, so pedestrians are assigned a longer ‘yellow’ period during which the flashing ‘don’t walk’ signal is displayed. Unless a walker arrives during the very brief walk signal window, they must wait. At a rather typical traffic signal, the light indicates ‘walk’ for as few as 6 seconds out of up to 2 minutes. Except for the lucky 5% of pedestrians who arrive during that brief 6 second window, they must wait an average of 57 seconds, and as long as 1 minute, 54 seconds. And that of course assumes the pedestrian actuator push button was depressed on time and registered by the traffic signal controller.

Another source of pedestrian delay is that ‘adaptive’ signals – which adjust green and red time in response to current traffic – give varying amounts of green time to approaching cars, depending on how many are approaching. Thus adaptive signals may extend a phase compared with a fixed time allocation. But pedestrian walk phases are not similarly extended because walkers cannot be guaranteed to clear the intersection quickly enough. There is an obvious inequity with this practice.

I have estimated that the delay experienced by pedestrians because of this is 27% of their time stopped at intersections. At 27% delay due to signals, a pedestrian can reach in 30 minutes what they could reach in 22 minutes if there were no traffic to worry about and no traffic signals to slow them down.

Traffic signals were installed for the benefit of people in cars, not people on foot. Pedestrianised districts don’t require traffic signals to ensure walkers don’t bump into each other. As traffic signals were steadily deployed over a century, they increasingly gave priority to cars and pedestrian conditions become significantly worse. I wrote earlier about accessibility loss. A pedestrian losing 27% of her time on a 30-minute walk is losing 8 minutes. This amounts to losing 45% of her accessibility because of waiting at traffic lights. She reaches only just over half as many opportunities as she would in a world without traffic signals, a world like 1920.

We encourage people to walk more and drive less, yet we design traffic signals to favor people in cars rather than on-foot. There are many things that could improve this situation, short of eliminating private car traffic from busy urban districts, which should also be on the table.

Pedestrian phases in places where there are many walkers could be automatic, without requiring the push of a button. Instead a pushed button could recall the cycle so the pedestrian phase comes sooner and the pedestrian walk signal is lit for a longer period of time.

Smart intersections could sense and count pedestrians automatically. New camera technologies are available but not widely deployed. We could also make much more effective use of the pedestrian actuators to estimate pedestrian flows.

Traffic signals could prioritise pedestrians to give them the maximum rather the minimal amount of green time necessary to cross the street.

We could give pedestrians a leading interval, meaning that the walk signal would be lit before cars get a green light to cross their path. This would increase the visibility of pedestrians because they would already be in the right-of-way before cars begin to move.

We could provide more “all pedestrian” phases. These are sometimes referred to as a “pedestrian scramble” or a “Barnes dance,” in honor of the New York City traffic commissioner who pioneered their use. The traffic signals could be set on “walk” by default, and only change to “don’t walk” when enough cars show up.

These are all things that we could do, but usually don’t. Instead we systematically design traffic signals to be hostile to people on foot throughout most of the United States and Australia.

Changes of these sorts in traffic signal timing would dramatically reduce accessibility loss for pedestrians. Since most transit trips require walking at the start and end of each trip they would make the larger city more accessible within thirty minutes by walking and transit. This would also expand transit ridership, but of course it might impose an accessibility loss on automobile travelers who would have to wait a bit longer at those intersections.

As we remake cities to reduce, if not eliminate, automobile dependency, making that tradeoff reduces the existing inequities that favor the movement of cars over pedestrians and cyclists. In addition to fairness, such a change would contribute to reductions in greenhouse gases, air pollution, crashes and other well-known “externalities” that are created by policies that prioritise cars over people.

Cities are made of places, not points

When developing plans for cities or intersections, planners often represent intersections, transit stops, and entire communities as dots on a map. Drawing lines to connect them with new roads, buses, or trains ignores important details. While large scale community plans are very important, we must recognise that, to achieve improved accessibility, the details hidden inside each dot matter.

A station is not a point — it’s a place that can be designed for efficiency and equity. An intersection is not a point – it is a space of flows that contains points where people going in different directions, using different travel modes, come into conflict with one another. Those flows and conflicts can be managed differently depending on our priorities and they should be a focus of policy.

We are blinded by regional plans that cause us to see places, small and large alike, as points. We are divided by the deep professional chasms within the transport community – traffic engineers and planners have similar objectives when it comes to safety and equity, but they often fail to communicate effectively with one another. Maps abstract away details, but the map is not the territory. We have ‘big thinkers’ who fail to consider the interaction of small places with their surrounding community, and ‘bounded thinkers’ who narrow the scope of work so much that they don’t think about the interaction of their place with the wider community.

Stations are designed so that passengers can exit, while ignoring what happens once they leave the station. The relationship between the station and the neighborhood around it can easily be neglected.

Bus networks are designed looking at how much of the map is within a measured distance of the stops, rather than how easily and how frequently they connect people to places.

Many of these misdesigns are not intentional and result from neglect, errors, or oversights. Designs that may have once been fit for the environment they served have not kept up with changes in the world around them. The goals of planners and engineers must change continuously to serve the people who live in the cities they create.

Planners, engineers, and — especially — politicians like to focus on building new things rather than repairing, restoring, and reshaping existing systems. Saving time is not pointless. Small amounts of time savings matter at bus stops, traffic signals, train stations, and everywhere else. Small time savings accumulate into large time savings, and increase the number of opportunities that can be reached in a given time budget. Reaching opportunities is why so many people live in cities in the first place.

This article is based on the 12th Annual Martin Wachs Distinguished Lecture in Transportation, given by the author at UCLA in May, 2019. A more extensive discussion appears in the book: The 30-Minute Version: Designing for Access. It is licensed under the Creative Commons BY-NC 4.0 license.


Levinson, D. M., & Lahoorpoor, B. (2019). Catchment if you can: The effect of station entrance and exit locations on accessibility. TransportLab

Levinson, D. M., Marshall, W., & Axhausen, K. W. (2017). Elements of Access: Transport Planning for Engineers, Transport Engineering for Planners. Network Design Lab.

Noland, R. B. (1996). Pedestrian travel times and motor vehicle traffic signals. Transportation Research Record1553(1), 28-33.

Norton, P. D. (2011). Fighting traffic: the dawn of the motor age in the American city. MIT Press.

Wachs, M., & Kumagai, T. G. (1973). Physical accessibility as a social indicator. Socio-Economic Planning Sciences7(5), 437-456.

Walking and Talking: The Effect of Smartphone Use and Group Conversation on Pedestrian Speed

Recently published:

Walking speed, walking in group, and phone use
Walking speed, walking in group, and phone use

Distracted walking due to smartphone use is on the rise resulting in growing concern over pedestrian safety and well-being. Our study measured the walking speeds of pedestrian groups differentiated by their smartphone use in two different environments – a wide pedestrian bridge at a university, and a narrow footpath on a busy commercial street. The results show that groups of people, phone users, and often followers of phone users, walk significantly slower than solo walkers uninfluenced by phone. Especially on the narrow street, people in groups and phone users are seen to not only slow themselves down but also slow the people walking behind them.

Should Road Rules be Rewritten to Put Pedestrians First | 2SER

I was interviewed on 2SER radio last week on Weekend Breakfast about Should Road Rules be Rewritten to Put Pedestrians First.

Listen here (mp3).

Now, let’s face it most drivers on our roads choose to cheat the concept of ‘giving way when permitted’ to other vehicles at the best of times.

However, what happens when we’re talking about giving way to pedestrians in unmarked areas? What do our road rules say and what does the law state? We were joined by Professor David Levinson from the University of Sydney’s Transport Engineering School of Civil Engineering to give us his perspective on the issue.