Transitmix: a new tool for armchair transit planners (and pros too?)

Screen Shot 2014-06-10 at 10.06.05 AM

Another interesting web transit app, this time from a group of Code for America developers:

Transitmix is a sketching tool for transit planners (both professional and armchair) to quickly design routes and share with the public. 

Transitmix is simple way to think about transit in terms of bus requirements and real costs. Basically, the user draws a route on a map and plugs in span and frequency. The app then calculates a vehicle requirement and cost in both hours and dollars, factoring in an adjustable layover ratio, average speed and dollar cost per service hour.  

Screen Shot 2014-06-10 at 10.06.13 AMScreen Shot 2014-06-10 at 10.06.17 AM

 

 

 

 

 

 

 

Transitmix is very similar to (though much simpler and prettier) the sort of cost estimation methods used in transit design processes , and as it stands is a fun sandbox to think about transit in your city. It's still very much a beta, but the prospects are intriguing. 

Its clear that the developers of Transitmix see it as much more than a curiosity. They've actively sought feedback from people in the industry, and are working hard to build an app that could one day replace some of the tedious documentation work of network design with an interactive, visually attractive interface. Apparently functions like summary tables, GIS file exports, and the ability to save multiple iterations of one design are all in the works.

I can imagine all sorts of possibilities for a tool like this, particularly if secondary data sources were incorporated. How about a public or stakeholder  involvement process that would actually give people a way to view demographic and ridership data and make real, financially constrained transit choices with a familiar, modern toolset? Or an update to our transit network design courses, where participants are given the same information for a fictional city and asked to design a transit network from the ground up? It's great to see transit concepts picked up by a talented group from an organization as reputable as Code for America. A project worth keeping an eye on!

17 Responses to Transitmix: a new tool for armchair transit planners (and pros too?)

  1. Neil June 19, 2014 at 2:46 pm #

    Very slick. I heart CfA in a big way.
    Do you think apps like this are viable as start-up businesses, or must they always be charitably funded? How much do you think a consultancy would pay for this?
    Say it needs two or three $70k pa coders plus 15% overheads. That’s about a $quarter million a year you’d like to be bringing in. Or say it costs $100k in coder time to build, and VC/angel investor wants to see $100k profit. How do you get back $200k, with some lower ongoing maintenance costs..
    Are there 200 consultancies that would pay $1000 for this? Or 1000 consultancies that would pay $200? Or are there consultancies that would hire a developer for $100k to develop and own the app for their own use?
    I’m genuinely curious to understand the potential market, since I can see two or three apps I’d love to (make) exist, but I can’t think how to make the economics work. How do those numbers compare to kinds of spare cash consultancies have, or can build in to client bids? And how do they compare to your understanding of the size of the transit/urbanism consulting market? How many transit agencies are there in the (first / new / whole) world?

  2. Daniel Howard June 19, 2014 at 4:11 pm #

    <3 <3 <3 but looking forward to the next few versions. :)
    I like that the app links over to this blog to explain transit costs. :)

  3. Sam Hashemi June 20, 2014 at 7:53 am #

    Hi folks, I’m part of the team building this tool. It’s still early days, so we’re looking for all the feedback we can get. I’ll keep an eye on this page for comments, but you can also reach out to the us directly: transitmix@codeforamerica.org
    Thanks!

  4. Matthew June 20, 2014 at 8:37 am #

    I played with a similar idea using the Google Javascript Maps API a while back. This seems to do something like that. One of the problems I ran into are: finding places for the buses to turn around is tricky when you are using the Google path tools; for a real bus system they often use little bits of roadway that cars are not allowed on so the driving directions tool will not route onto them. It seems that this TransitMix dodges the whole problem by not considering the need to turn around buses at either end point at all. That’s a missing feature. Layovers should also be in there, and location of bus depots, when you get a chance.
    I also recall some scalability issues that perhaps they have figured a way around: when I made a long enough route, with too many waypoints, Google Maps Javascript started to choke on it.

  5. J June 20, 2014 at 11:13 am #

    Maybe consider a way to submit to local “agencies” or public sharing forum?

  6. J June 20, 2014 at 11:31 am #

    Also, perhaps a tool for light rail/subway proposals?

  7. Cyclelicious June 20, 2014 at 3:35 pm #

    This looks interesting, but it doesn’t seem to work for me using either Chrome (35.0.1916.153 m — up to date) or I.E. (11.0.9600.17126 also up to date) on Windows.
    I get the “choose a city” dialog, accept the San Francisco default, press “Start” and … nothing happens.
    Looking at the Javascript console, it appears the mapbox library might be broken (or perhaps has an invalid key for whatever service it uses?). The error message says: “Unable to geocode city. Womp Womp. undefined” followed by: “Uncaught TypeError: Cannot read property ‘geometry’ of undefined”

  8. Sam Hashemi June 20, 2014 at 5:58 pm #

    @Cyclelicious Sorry about that, we got way more traffic than we expected and hit some of our server limits. It’s been fixed now!

  9. Jeffrey Bridgman June 20, 2014 at 6:24 pm #

    Very neat! I’ve already modeled up how I’d like to see my small city’s transit system reorganized 😉

  10. mike0123 June 20, 2014 at 6:52 pm #

    Feature requests:
    1) Add the ability to load the relevant parts of a GTFS (so that existing networks can be calculated and compared against new networks).
    2) Add an option to use the base map’s travel time for cars (if Google provides it) instead of the input field for speed.
    3) Add a typical stop spacing input field and a dwell time per stop input field, and add the product to the car travel time.
    4) Allow routes that are not constrained to roads (rapid transit) and that have stops.
    5) Add a trip time calculator between two user-specified points.
    5) Add an average trip time calculator that plots average trip times, including headway and walking, by time of day and day of week. In a too-simple model, the calculator could use 100 points and sum up the travel time between them. In a more realistic model, the calculator could use origin/destination weighting from another data source like the census.

  11. Austinonyourfeet.wordpress.com June 20, 2014 at 8:20 pm #

    Building on what mike0123 said, it would be great if instead of using the basemap’s travel time for cars, it used the actual system’s travel time for existing buses. I want to design a realistic system, using realistic travel times. No more realistic travel time than the time that exists right now.
    For example, if you have a bus that takes 10 minutes to go from A to B and 14 minutes from B to C. You also have a bus that takes 13 minutes to go from A’ to B. It would be great to be able to build a bus that goes from A’ to B to C, and know that it would take 27 minutes.

  12. anonymouse June 21, 2014 at 12:39 pm #

    I guess this might be a good time to plug the suite of GTFS tools I’m working on. I’ve mostly been doing this for my own amusement, but I can code up something that gives average travel time per segment retty quickly if people want it. And if you’ve got some example GTFS you want to run it on, let me know, so that I can make sure it parses correctly.

  13. Eric Goodman June 24, 2014 at 12:24 am #

    I’m really excited about this tool!!! It shows the potential for engaged and informed planning very well. Refinement over more iterations will add capabilities and solid supporting data. It will facilitate better communication with citizens and decision makers by showing trade-offs clearly, based on realistic operating scenarios. More ideas will be tested and the public will gain a better understanding of service cost. It’s a step beyond what we’ve done with the Transit Values Exercise at Community Transit. I could use it to help identify and demonstrate small cost / large benefit speed and reliability improvements, and for route and network design.
    I’m starting work on a similar project this fall and would like to offer any help I can. If you’re involved, please get in touch. I’m very happy with this beta version. It’s an excellent foundation…

  14. Nelson Galeano June 30, 2014 at 1:45 pm #

    I like it!
    The number of stations/stops impacts travel time and therefore operating costs. Capacity of vehicles plays also a critical role in costs determination. Routing alignment does not need to be extremely accurate, however some new links would be interesting to be used like tunnels and bridges.
    Different modes (BRT, LRT, Metro) need to be clear and identifiable.
    Nelson Galeano

  15. Alan July 3, 2014 at 4:34 pm #

    This is a prospective tool that fills a real need. As a one-person professional transit planning office, I need a sketch planning tool to map bus route options and simplify the miles, time and cost estimation for various service scenarios. I suggest means to develop routes in segments between designated timepoints, setting avg. speed by segment. Also, provide means to identify stop locations and set avg. dwell time per stop. I agree that means to identify off-street routing and stops would be useful. Means to generate concept schedules using timepoints would also be appreciated. Finally, the maps and schedules should be exportable to Office software such as Word and Powerpoint to make presentation ready. When will you be ready to release for professional use? I need it yesterday!

  16. Nathanael July 11, 2014 at 12:55 pm #

    Good thing to make open source.
    The ability to make one-way loops is important, unfortunately.

  17. Suzi Nasby April 14, 2016 at 10:16 pm #

    Thank you for the auspicious writeup. It in fact was a amusement account it. Look advanced to far added agreeable from you! However, how could we communicate?

Leave a Reply