houston: transit, reimagined

Yesterday, Houston Metro began seeking public comment on what may be most transformative transit plan in its history.  I'm honored to have been a part of it, as the network design lead* on the consulting team.  Read all about it, in as much detail as you want, here.  Explore the detailed map here.  Note that the pulldown menus in the black bar lead to lots of cool maps and diagrams, as well as extensive data about the plan.

The plan shows that without increasing operating cost, Houston's frequent network — the network of services where the bus or train is always coming when you need it — could grow from this …

Frequent Network Existing


to this:



This cool page toggles between the two, so you can see the system growing.

How on earth could we grow a network that much without new money?  There are two answers:

1.  That's how much waste there was in the existing system.  Waste in the form of duplicative routes, and due to slow meandering routes created due to a few people's demands.  

2.  Hard choices are proposed about expensive service to very small numbers of people.  The plan devotes 80% of Metro's resources to maximizing ridership, which all of these frequent lines do, and only 20% to providing access to people living in expensive to serve places.  Currently only about 50-60% of resources are devoted to services where high ridership is a likely outcome.   (See here for my paper on this analysis method.)   This shift in focus will have negative impacts on small numbers of riders who rely on those services, but these were small numbers indeed .  (About 0.5% of existing riders end up over 1/4 mile of service, and most of them are just over that threshold.  Often, their longer walk is to a better service, a tradeoff that most people are willing to make in practice.)

The exciting thing is not just the massive growth in frequent services proposed, evident above, but the shape that they'll take.  The core idea of the new network is the high-frequency grid, designed to enable anywhere to anywhere travel with a single fast connection.  Everywhere on the proposed network of red lines, that kind of easy access will be possible.

Obviously, too, the whole geographic focus of the network had to shift.  Houston is one of the biggest US cities that still has a radically downtown-oriented transit network despite decades of decentralization.   The core area where the existing network converges has only 25% of the region's jobs, and while transit must favor the jobs that are in dense and walkable settings, there are now many highrise clusters around Houston that answer to this description to some degree. 

Houston has been growing mostly westward and northward in the last few decades.  Its densest residential neighborhood, for example, is Gulfton, located 7 miles west of downtown.  Not far from there is its second-densent employment and activity center, Uptown-Galleria.  Houston is a constellation of centers, and the transit network needs to be more decentralized to effectively service all of those centers where the density and walkability make transit viable.  The high-frequency grid, shown above, reaches all of those places.

Houston also features a fascinating patchwork of incomes.  There are rich and poor neighborhoods but there's no longrer a rich or poor side of town.  That means that low-income people, too, will find the whole network useful.  We have done our best to retain useful service on the historically low-income and minority eastside, despite declining population in some areas.  The key strategy there was the anchor most services to the main universities in that area (University of Houston and Texas Southern U.) which are the surest drivers of all-day demand.

The huge no-cost expansion of useful service may remind you of a plan I worked on two years ago for Auckland, New Zealand, where it was also possible to massively expand the frequent network by redeploying duplicative services.   Not all  transit agencies have this much waste, so your city's mileage may vary.  But if you suspect that transit could be doing more in your city, read all about the Houston plan.  You'll be amazed, as we were, about how much is sometimes possible.

Finally, if you're in the Houston Metro service area, remember to submit a comment even about things you like.  Sadly, most of the public comments received on transit plans are negative even if the plan is broadly popular, becuase people who like it falsely assume it will happen anyway.   This plan will not be implemented if it does not attract strong support.   We welcome constructive comments about the plan, which will be used to make the final plan even better.  But if you like the plan, it's important to say that as well!  Instructions for how to comment are here.  


* This term means I led the design workshop that developed the design, but it does not mean I get all the credit.  These plans are collaborations among many players, both on our large consulting team and of course at Houston METRO.  The team was headed by TEI of Houston — Geoff Carleton was the excellent project manager — and included Carol Lewis, Nancy Edmondson, Dan Boyle, and Asakura Robinson.  Kurt Lurhsen, METRO's head of planning, shepherded the project internally with the support of a great teams.  The plan would also not have gotten to this stage without the preliminary support of METRO's Board, including Chair Gilbert Garcia and Strategic Planning Committee Chair Christof Spieler.  Spieler has been an especially tireless advocate of this project from its earliest days.  All of these people and organizations contributed substantially to the plan as it appears.

23 Responses to houston: transit, reimagined

  1. Dono May 9, 2014 at 11:10 am #

    please come fix BART for us! we desperately need you or at least someone thinking like you!

  2. tacony May 9, 2014 at 12:34 pm #

    Wow, looks excellent.

  3. Herb Curl May 10, 2014 at 11:32 am #

    Like many cities, Seattle is increasingly too expensive for middle to lower income people to live there, although that’s where they work. Has any plan been developed to provide bus service specifically to concentrations of lower income people who commute to work in city cores?

  4. Peaton May 10, 2014 at 12:59 pm #

    “80% of Metro’s resources to maximizing ridership” What would look different if this were 100%? What would the projected impact on ridership be and what are the barriers to making this switch (legal, political, moral)? Do coverage based routes really help anyone in relation to their costs?

  5. jack horner May 10, 2014 at 5:44 pm #

    Concerning the balance of ridership vs coverage services: transit agencies need to have clear geographical criteria for providing service, and they need to have the strength to say to property developers things like: ‘If you want to build a retirement village at the end of a 1km culdesac, please understand that the new residents will *never* have a public bus service, and if they ask us that’s what we’ll tell them.’
    And when the new residents come knocking a few years later, the agency needs to have the strength to say, ‘Sorry – no. Here are our guidelines, and you don’t rate.’
    Having clear guidelines concerning the geographical criteria is key, so everyone knows where they stand.

  6. Eric May 11, 2014 at 5:08 am #

    “without increasing operating cost”
    That’s amazing. Wow.

  7. Kymber May 12, 2014 at 10:31 am #

    What’s scary about the “before” map is that the Red light rail line presently has a terminal with NO high-frequency bus service.
    And the “after” map has three.
    Here’s hoping the voters of Houston see the wisdom in this and approve it.

  8. Mike May 12, 2014 at 4:29 pm #

    Looks pretty good and amazing what you can do by optimizing resources.
    The ridership vs coverage idea still concerns me.
    While I completely understand the idea of putting more frequent service where it will have the most ridership, we cannot expect transit to be an alternative to the auto if it does not serve all built up areas.
    This seems to be a trap that most American transit agencies and planners seem to get themselves stuck into. I noticed on the slides comments about some areas getting demand response because there was pockets of low income population.
    If the transit system is being designed to be a true alternative, then that should not matter, and the service should be provided regardless of the income levels.
    It also creates an issue where we are telling people in certain areas that they can’t have quality transit service because we as planners think the built form will not support it. This is the trap Paul Mees was advocating we get out of as planners, and instead focus on providing quality transit to all corners of a region.
    So great plan, but I think in future plans, I would rather see the approach that is taken in Canada, where there is a mandated goal to provide a basic quality transit service within a 5 or 10 minute walk of 95% of the population in the service area. Then you build up the main corridors from there with more frequent service.
    The only other issue I noticed with this Houston plan is that bus routes have different service spans.
    Why not just choose to operate all the routes until say midnight seven days a week? If you have parts of the system closing down earlier, then again, you are not offering a true alternative to the auto.
    And what about the other 1.5 million people who will not be on the frequent service network?
    How is transit going to address their needs and provide an attractive alternative to the auto?

  9. Mike May 12, 2014 at 4:31 pm #

    “Like many cities, Seattle is increasingly too expensive for middle to lower income people to live there, although that’s where they work. Has any plan been developed to provide bus service specifically to concentrations of lower income people who commute to work in city cores?”
    Herb Curl. If cities have to provide special routes to get low income people to work, then I think we can say the transit system is failing.
    If we are truly going to build public transit networks that maximize mobility and offer a real alternative to the auto. Then we should not have to worry about routes to serve low income populations, because all areas of the region would have an attractive transit service that meets the needs of not only poor people, but people with a choice.

  10. asdf May 12, 2014 at 8:35 pm #

    I grew up in Houston, and my biggest transit experience there now is coming from the airport to visit my family. It’s a 3-seat ride, totaling about 27 miles, eventually ending up in the southwest portion of the city. Bus from airport to downtown, light rail to the medical center, then another bus from the medical center to home. It’s a perfect example of the kind of random trip where a frequent, gridded network is so essential.
    The current network works reasonably well, provided that the plane lands between 2 and 4 on a weekday afternoon. The new network will make a smooth trip on a much wider range of times, besides speeding up the first leg by at least 15-20 minutes (by removing a bunch of zig-zagging).
    It’s a great idea. Unfortunately, this being Houston, I’m not too optimistic about it getting people out of their cars. Based on personal experience, most buses are half-full during rush hour and mostly empty during non-rush-hour. This is also a city where transit debates have strong racial overtones (an overwhelming majority of regular transit riders there are minorities), and a city whose own congressman (John Culbertson) is using his influence to keep federal transit money away from his district.
    This is also a city where (except for college dorms), the support network for living without a car is largely lacking. Many sidewalks are poorly maintained and disappear entirely in many neighborhoods. Pedestrian cut-through paths in residential cul-de-sacs do not exist. Bike lanes are almost non-existant. Arterial roads that a bus could plausibly run on are a mile or two apart. Freeways and railroad tracks divide the city, with limited crossing points. There is no real-time info for buses. Express buses are virtually non-existant outside of rush hour. Uber and Lyft exist, but are spread thin, compared to real transit cities. Car2Go does not exist at all, while Zipcar’s sole location in the entire city is restricted to Rice University affiliates only. Bikeshare technically exists, but the coverage area is so small it provides little benefit over walking.
    Sometimes, it seems like people there lack a basic sense of awareness of non-car trips. Like the grocery store in my former neighborhood across the street from a major transit center, with fences and “don’t walk” signs directing everybody to go the long way around to the nearest official crosswalk. (In practice, people ignore these signs all the time and cut through using the driveway intended for buses exiting the transit center, along with some landscaping tromping).
    While a frequent network is a great start, Houston has a long, long way to go for transit to become anything beyond an afterthought for the average person’s life.

  11. Sascha Claus May 13, 2014 at 3:26 am #

    > Freeways and railroad tracks divide the city, with limited crossing points.
    Sounds like an application for small-scale chokepoints; or at least an opportunity to bring multiple (paralleling) routes together for easy connections.

  12. Andre Lot May 13, 2014 at 9:15 am #

    Something that stroke me on the revamped plan is that the light rail didn’t become a focal point, with many bus routes duplicating parts of the light rail lines. Isn’t that a waste? If you have a good light rail line, why bother still running parallel buses? Shouldn’t all buses that run parallel to light rail (especially the red line) be just truncated at the first/last common stop with the light rail?

  13. Mike May 13, 2014 at 5:48 pm #

    I was looking at the Houston map in more detail, and it is amazing how currently and in the planned network, there are huge swaths of built up areas with no transit service.
    Like this area
    Areas like this in other cities would have a bus every 15 minutes – 30 minutes. Why was it decided that such areas in Houston should have no transit?

  14. Sascha Claus May 15, 2014 at 12:56 am #

    > Shouldn’t all buses that run parallel to light rail (especially the red line) be just truncated at the first/last common stop with the light rail?
    In one theory, yes. In another theory, a route shouldn’t be terminated shortly before a transfer point (Wheeler and TX Med Ctr) or shortly before a large destination (Downtown and probably TX Med Ctr).
    The purple line seems to have even more paralleling without the abovementioned reasons.

  15. J May 15, 2014 at 7:19 am #

    Why? Because it’s a shit-hole. Half the city doesn’t even have sidewalks…

  16. Jarrett May 15, 2014 at 8:40 am #

    Mike. There are large AREAS with no transit service, but not all that many people in many of those areas. What’s more, in those areas the street environment is so hostile to pedestrians that transit has little opportunity to succeed.

  17. Danny May 22, 2014 at 12:11 pm #

    This looks fantastic! Though, as a Houstonian living along Main Street in The Heights, I wish there was high frequency bus service up Main Street linking The Heights to the Red Line MetroRail.
    On the whole though, a system like this might actually facilitate my dream of being able to ditch my car!

  18. scott t May 23, 2014 at 12:09 am #

    i bet the small number of negativley affected coudl be given folding bikes to offset their longer walks/times to buses and it would still be cheaper

  19. Haruki November 11, 2015 at 6:38 am #

    the menus are not working

  20. D July 26, 2016 at 2:16 pm #


    I work for a Transit agency in the icy frigid north and we are in the midst of undergoing a transit strategy overhaul which may lead to us reallocating resources from a peak focused pulse based network to more of a connective network similar to Houston. I have been examining Houston’s network and a few key differences stick out. Houston surprisingly mostly has a grid street structure. While I am fully aware of the benefits of a connective network, my colleagues are less convinced. A few of the arguments that arise is that places like Auckland and Houston have fairer weather with no snow and freezing temperatures. Another key issue is that outside of the inner suburbs the majority of new growth has taken a SFD spaghetti street network. We have also been conducting an outreach process to try and gauge where the public stands on what they want. The results are mixed with the results slightly in favour of consolidating service hours for fewer routes in areas with higher demand that come more often. However one of the arguments that also comes up is that older adults that live in low demand neighbourhoods will have difficulty walking to transit if it farther away. While this is a valid argument, it also a very emotional argument takes the conversation hostage. So I guess my question(s) for you: Do you have a case study of a predominantly sprawled SFD and non-gridded city with a cold climate with a connective transit network? Consolidating service hours should improve mobility and access for everyone, so how do you get around the notorious “old lady can’t walk to the bus stop” argument?

  21. Mason October 6, 2016 at 8:10 pm #

    The “this cool page” for toggling between the options doesn’t work anymore, can you fix please?

  22. Sam McLeod April 6, 2017 at 8:31 pm #

    Hi Jarrett,

    You may like this article, published in JPL, where we discuss your ideas extensively:

    Warm Regards,

    Sam McLeod
    Curtin University

  23. io games December 6, 2017 at 12:06 am #

    I have read through other blogs, but they are cumbersome and confusing. I hope you continue to have such quality articles to share with everyone!