New paper out: Disparities in travel times between cars and public transport

I am very glad to share our new paper looking at the travel time gap between private and public transport at high spatial and temporal resolutions. The study combines real-time traffic data, transit data, and travel demand estimated using Twitter data to compare this travel time gap in four cities (São Paulo, Stockholm, Sydney and Amsterdam).

Despite remarkable differences between these cities in terms of transportation networks, area, and population size, we found travel times of transit and vs. driving are surprisingly similar across cities: R < 1 for trips shorter than 3km, then increases rapidly but quickly stabilizes at 2 (figure below). Moreover, using public transport generally takes on average 1.4–2.6 times longer than driving a car. The share of area where travel time favors public transport over car use is also very small in all cities. As Giulio Mattioli noted on Twitter, these results 'would confirm that car dependence is much more than just a question of culture & attitudes'.

The paper is open access and it was written in collaboration with a great team led by Yuan Liao and Sonia Yeh at University of Chalmers, Sweden.

Liao, Y., Gil, J., Pereira, R.H.M. et al. Disparities in travel times between car and transit: Spatiotemporal patterns in cities. Scientific Reports 10, 4056 (2020). https://doi.org/10.1038/s41598-020-61077-0

Abstract:

Cities worldwide are pursuing policies to reduce car use and prioritise public transit (PT) as a means to tackle congestion, air pollution, and greenhouse gas emissions. The increase of PT ridership is constrained by many aspects; among them, travel time and the built environment are considered the most critical factors in the choice of travel mode. We propose a data fusion framework including real-time traffic data, transit data, and travel demand estimated using Twitter data to compare the travel time by car and PT in four cities (São Paulo, Brazil; Stockholm, Sweden; Sydney, Australia; and Amsterdam, the Netherlands) at high spatial and temporal resolutions. We use real-world data to make realistic estimates of travel time by car and by PT and compare their performance by time of day and by travel distance across cities. Our results suggest that using PT takes on average 1.4–2.6 times longer than driving a car. The share of area where travel time favours PT over car use is very small: 0.62% (0.65%), 0.44% (0.48%), 1.10% (1.22%) and 1.16% (1.19%) for the daily average (and during peak hours) for São Paulo, Sydney, Stockholm, and Amsterdam, respectively. The travel time disparity, as quantified by the travel time ratio R (PT travel time divided by the car travel time), varies widely during an average weekday, by location and time of day. A systematic comparison between these two modes shows that the average travel time disparity is surprisingly similar across cities: R < 1 for travel distances less than 3 km, then increases rapidly but quickly stabilises at around 2. This study contributes to providing a more realistic performance evaluation that helps future studies further explore what city characteristics as well as urban and transport policies make public transport more attractive, and to create a more sustainable future for cities

The relationship between travel distance and travel time ratio R (PT travel time divided by the car travel time)

Travel time to closest healthcare facility in Rio de Janeiro

The map shows how long it takes (in minutes) to travel by public transport and walking to the closest healthcare facility across the city of Rio de Janeiro. The analysis is disaggregated for facilities providing low-, medium- and high-complexity services. 

The first thing to note here is that physical accessibility to public health is relatively high in Rio. Approximately 94% of Rio`s population could reach at least one facility providing low-complexity services under 30 minutes. Under the same time, medium- and high-complexity services could be reached by 81% and 72% of the population, respectively. This is explained to some extent by the spatial planning of healthcare in the region, which has been relatively successful in spreading low- and medium-complexity facilities across the city. The map also gives a good sense of how the distribution of healthcare facilities vis-à-vis the public transport network varies across space, and how access to public health tend to be much lower in the west and particularly in the urban fringes of the city.

ps. This is a map I created for my PhD research but I didn't include it in the thesis in the end . To create this map I used a 2015 dataset of healthcare facilities and the GTFS of Rio's public transport network from March 2017. The dataviz and data wrangling were done in R. In case you're interested in doing similar analyses, I've created a simple tutorial with reproducible example on how to use OpenTripPlanner (OTP) to estimate travel times.

[click on the image to enlarge it]

The geography of Manhattan distorted by travel-times

The figure below was created by Stefan Musch from Gradient Metrics (hat tip Jean Legrand). The figure was created using R and ggplot2 based on travel-time estimates from Google Maps API. There is a bit more info about the creation process in this post and perhaps Stefan will share his code at some point.... please? :)

Echoing the comments of others on Twitter.  The figure does a great job illustrating how it is much harder to cross Manhattan from east to west than from north to south. Finally,  it would be great to see how this shape has changed over the last decades using historical travel time estimates.

credit: Stefan Musch (Gradient Metrics)

Assorted Links

1. Thanks to Geoff Boeing's great work, you can easily create isochrone maps for anywhere in the world automatically with Python and its OSMnx package
   
   
2. Christian Brand (TSU/Oxford) is one of the co-authors of a new WHO report on "Health economic assessment tools (HEAT) for walking and for cycling"
   
   
3. An AI experiment to find unsafe housing using street-level images, by Jonathan Jay
   
   
4. The Right to the City: Free Ebook Download
   
   
5. Rio's mega-event transportation investments did not address inequalities (or even improve mobility)
   
   
6. call for papers on supply chain and logistics:  2018 MIT SCALE Latin America Conference
   
   
7. The Harvard University Data Science Initiative is seeking applications for its Harvard Data Science Postdoctoral Fellows Program for the 2018-2019 academic year
   
   
8. Google has been mapping air pollution using Google Street View cars. So far, they've measured over one billion air quality data points and now air quality scientists can request access to their data

Assorted Links

1. Isoscope: work by Flavio Gortana (Twitter), showing how isochrones by car vary across space and time
   
2. Tutorial: Spatial Datasets and Urban Applications (code, slides + data)
   
   
3. Beautiful photographs of Forgotten Places, via Darran Anderson
   
   
4. Five tools that help you master map projections
   
   
5. The unfinished cities of Spain via Urban Living Lab
   
   
6. 15 years of urban growth in China, via Lazaro Gamio
   
   
7. Surnames and ancestry in Brazil, by Leo Monasterio
   
   
8. "Seeing Theory – A visual introduction to probability and statistics.”
   
   
9. paper: Combining satellite imagery and machine learning to predict poverty, by Neal Jean et al [code]

How much faster you could travel today compared to 1914

A couple of months ago, Intelligent Life magazine published a great isochronic map from 1914 showing how much time (in days) it used to take to travel to anywhere in the planet if you departed from London.

The team at Rome2rio have recently updated the map to 2016, which does a great job in showing how the world has gotten incredibly "smaller", though I would rather say "faster". I haven't found any information on which datasets they have used to do this though.

Related Links:

• How far can you go from any European capital to anywhere else by train? 
• Isochronic maps of American railways in the 1800s
• Old and New Isochrone Maps
• A Travel Time Map of the Roman Empire

Walking Tube map, or the most important map when you are in London

Map of the  last  week: Transport for London (TFL) has recently published new map that shows how long it takes to walk between stations on the London Tube (a.k.a. underground or subway).

Click here for the full size map.

How far can you go from any European capital to anywhere else by train?

Stephan Franziskus runs an interesting blog where he has written a great post about the history Isochrone Maps. These maps use color gradients and contour lines to visualize the places one can reach from a single destination within different time windows (isochrones).

obs. We have already posted about these maps in this blog to show some isochronic maps of American railways in the 1800s, a similar map departing from Rome during the times of the Roman Empire, and a contrast between Old and New techniques of isochrone maps.

Stephan Franziskus shows a very precious map, created by Francis Galton (yeap, the same Galton who demonstrated the central limit theorem using a 'bean machine'), which shows the number of days to get to different places in the world if you were to departure from London in 1881.

More recently, Peter Kerpedjiev has applied the same idea to estimate the places one can reach from different European capitals within different time windows using only trains and walking. There is a very good piece by Lazaro Gamio published in The Washington Post covering Kerpedjiev's work. You might like it.

click on the image to enlarge it

[image credit: Peter Kerpedjiev, Lazaro Gamio, WP]

Taxi vs Public Transit in major US cities

Nate Silver has recently published a nice piece in his blog comparing the travel time performance of taxis (usin data from Waze) vs public transit (using GTFS data behind Google Maps). Nate compared average journeys from the Airport to downtown during weekdays in major US cities. Here is the result:

[image credit: Allison Mccann 538]

Average Commute Time Worldwide

This chart was taken from this study here. In general, commute times are relatively long in Brazilian metropolitan areas, especially when their wealth/population size is taken into account.

Obs.1: If you know of data sources (preferably household surveys) with commute time information for other metropolitan areas across the globe, I would to receive a message from you.

Obs.2: If you spot a name in the chart you've never heard of, it's probably a Brazilian metropolitan area.  

[image credit: Pereira and Schwanen, 2013]

Related Link:

• Parking Rates Worldwide
• Wendell Cox recently wrote a post about the 10 cities with the worst traffic in the world, according to the Tom Tom Traffic Index.  

Old and New Isochrone Maps

Apparently, this is one of the oldest isochrone maps, circa 1920. It shows the “minimum” travel time into the city of Melbourne via suburban railways and tram lines (via Daniel Bowen and Transit Maps).

 Transit Maps also points out to this isochrone map of Manchester in 1914.  Finally, the Atlas of the Historical Geography of the US of 1932 showed also published a few isochrone maps of American railways in the 1800s. These maps comprise only a small sample of our old obsession with time.

Train and Tram Travel Times in Melbourne, Australia, c. 1920

More recently, some people/projects have been applying new technologies to this old obsession with travel times, and the results include some pretty amazing maps. Among many of these new projects, I would highlight two:  The great Mapnificent (by Stefan Wehrmeyer).

Ant the amazing work Xiaoji Chen and her maps of Singapore and the isogreenic (!) map of Paris

Related Links

• A Travel Time Map of the Roman Empire 
• Accessibility Observatory + GTFS data

Accessibility Observatory + GTFS data

David Levinson, Andrew Owen and their team at the University of Minnesota have recently created the Accessibility Observatory. If you're interested in the topic, you may read a nice piece written by Emily Badger for TAC.

One of the greatest aspects of this project is about the data sources and methods they use for accessibility evaluation. Instead of relying on traditional household travel surveys, they combine open source data from OpenStreetMap (OSM) and from General Transit Feed Specification (GTFS), developed by Google. With this sort of data it is possible to analyze jointly the public transport system as a whole (routes, stops, and schedules) plus pedestrian routes and walking times. There is a more detailed explanation of how it works here. 

I've bee thinking a lot about this 'empirical strategy' since I'm planning to dedicate part of my PhD to research socio-spatial inequalities in the distribution of accessibility. The more I think about this, the more I become a passionate skeptic, more passionate than skeptic.

To be honest, I believe that these data/methods are probably not going to revolutionize accessibility research in transport studies. However, this new 'empirical strategy' allows us to expand accessibility analysis so as to incorporate hundreds of cities where no travel survey has been undertaken. It also provides a tool with great potential to compare accessibility levels across space and time and to assess how certain modifications in the public transport system can affect accessibility levels. Not to mention the research questions you could address by combining these data/methods with Census data.

By the way, the use of GTFS in accessibility analysis is quite new. I have found a few applications that have been using GTFS since 2010 but none seems to be as advanced the Accessibility Observatory at Minnesota. Here is a list with some of those other initiatives:

• Stefan Wehrmeyer started in 2010 a project called mapnificent. Today it has perhaps the most comprehensive collection of cities so far.
  
  
• Comercial initiatives by Trulia in the US and by Mapumental in the UK.
  
  
• In 2011, Brookings Institution published the report 'Missed Opportunity - Transit and Jobs in Metropolitan America' analysing job accessibility based on GTFS for 100 metropolitan areas in the US.
  
  
• Renato Vieira and Haddad (2012) proposed a similar approach to build an accessibility index to São Paulo.

In 2013 we saw many more initiatives:

• Yuval Hadas published a paper in the JTG: Assessing public transport systems connectivity based on Google Transit data
  
• WNYC came up with this map of Transit times in NYC
  
• Andrew Hardin has also created his version of the map covering four other cities. Here is a recent study where he explains how he calculates city wide travel times by simulating walking and public transit
  
• and finally, Sean Barbeau and Aaron Antrim wrote a nice piece on the many uses of GTFS data (slides)

I must be missing some other initiatives and publications, so please leave a comment to this post with suggestions and links if you remember any.

[image credit: Transit Accessibility for the Minneapolis St. Paul region from the Accessibility Observatory]

A Travel Time Map of the Roman Empire

A neat isochronic map! via Patrick Chovanec:

Travel times from Rome, at the time of the Roman Empire in number of days (each shade represents one week). This map is part of the The Stanford Geospatial Network Model of the Roman World. There is also a short video showing how to export data and to navigate the interactive maps of the projecet.

[image credit: orbis.stanford]

Related Links:

• Isochronic maps of American railways in the 1800s
• Travel-time Maps and their Uses
• Isochronic London Underground
• Netherlands Isochronic Map

Bus speeds visualized in real time

As you can see here, Andy Woodruff is a quite talented cartographer! In june this year he released this interactive map where you can visualize (almost in real time) the bus routes and speeds in Boston.

More information at Bostonography blog.

[Image Credit Andy Woodruff]

Related Link:

• Shortest-path routing through geotag density

13th WCTR + Commute Time in Brazil

Dear readers, next week I'll be in Rio for the 13th WCTR. I'm presenting a working paper Tim Schwanen and I have recently published. In this study we explore 18 years of commute time trends in Brazil. I was planning a dedicated post to talk about this study but I've been quite busy these days. 

So here it is. You may download it here:

Commute Time in Brazil (1992-2009): Differences Between Metropolitan Areas, by Income Levels and Gender

[Portuguese version]

ps. It would be great to meet some of you in case there is anyone out there that's also attending the conference. If there is anyone out there, just leave a comment of drop me an email.

Crowdsourcing road congestion data

Nathan Yau points out to this interactive map showing county-level commute time estimates for 2011 in the US. The data source is the American Community Survey organized by the United States Census Bureau. Pretty good job!

Now imagine if you could have acess to real time data on traffic conditions on arterial roads in several cities around the globe. There is one company that generates these data. I know what you're thinking: "Damn these guys from Google are awesome!"

Acutally this is not a new project. Google has started it aroud 2009 and now it covers. several cities around the globe. They basically track anonymous locations from smartphones to gauge traffic conditions in real time:

When we combine your speed with the speed of other phones on the road, across thousands of phones moving around a city at any given time, we can get a pretty good picture of live traffic conditions. We continuously combine this data and send it back to you for free in the Google Maps traffic layers. It takes almost zero effort on your part — just turn on Google Maps for mobile before starting your car — and the more people that participate, the better the resulting traffic reports get for everybody.

The Google Maps Traffic Layer (web version) also inclues traffic prediction for any day of the week and time of the day, based on past conditions. Neat! Now it should be very interesting to compare/validate Google's traffic data to "real data" from more traditional sources.

Any volunteers?

Isochronic maps of American railways in the 1800s

We have already seen some Isochronic maps here at Urban Demographics Blog. Here are a couple of travel time maps for rail mode in the US using New York City as a starting point (1800, 1830, 1857 and 1930). Via Nathan Yau.

Time to get from New York City to almost anywhere in the US, in 1857.

[Image Credit: the 1932 Atlas of the Historical Geography of the US, via Michael Graham Richard]

Travel Time Maps

We have already seen some Isochronic maps here at Urban Demographics. This post shows a few more Travel Time Maps for other cities around the world.

• A very good analysis on travel time variation across Melbourne and Victoria (Australia)
  
  
•  a few  UK Cities

• Mapnificent for more than 20 cities (tutorial video here)
  
  
• New York - Travel Times on Commuter Rail via (visual complexity)
  
  
• Isochronic London Underground (interactive)
  
  
• and the most attractive: Netherlands Isochronic Map (by Vincent Meertens via wired)
  

TIMEMAPS from graphsic on Vimeo.

Assorted links

• Urban Network Analysis: A Toolbox for ArcGIS 10 (via @jmichaelbatty)
  
  
• G-Econ - Geographically based Economic data (via devecondata)
  
  
• ‘Middleweight’ Cities as Growth Engines
  
  
• NYT tool for Mapping the 2010 U.S. Census (via UPB)
  
  
• A fresh look at US population trends
  
  
• Mapnificent (via The Map Room)

Mapnificent from Stefan Wehrmeyer on Vimeo.

Commute Map

What could you do if you had origin-destination commute data, zip codes and Google Maps API ?

Great job Harry Kao !

(via The Map Room)