what makes a good “planning game”?

People learn from doing more than by being told, so public outreach is increasingly turning to various kinds of simulations, which can broadly be called planning games.  A planning game is any interactive exercise that allows people to play with the tools of planning, under some kind of budgetary constraint, and thus to experience the hard choices that arise from the material. 

You've probably played with online tools, often published by newspapers, that invite you to "balance the government budget yourself."  Planning games drill down to a more detailed level than that, but they share the same spirit:  They invite the citizen to express her views through grappling with the actual problem that government is facing.  When used as outreach tools, the rule of these games is: Listen and educate at the same time. 

Planning games can be group exercises in public meetings, but of course they can also happen on the web.  Web-based versions are better for allowing an individual to explore at her own pace, but they don't fully expose the user to the diversity of opinions and needs in a community.  Public workshops tend to be much better for that.

I'm inspired to return to this topic because of an intriguing style of planning game developed by Community Transit in suburban Snohomish County, Washington (north of Seattle).  I'll talk first about the kind of planning game I do — and the one on which my network design course is based.  Then I'll talk about the very different approach that Community Transit used.

Geography-Based Planning Games

To build stakeholder consensus around a transit plan in a difficult area, TransLink in Vancouver BC uses a geographical planning game.  I helped them hone this tool for their South of Fraser area in 2006 and they have used it successfully in other areas since then.  It can be helpful in many other contexts.  This planning game technique, applied to a fictional city, is also the core of the interactive course in network design that I teach.

Picture1 For a long-range infrastructure plan, stakeholders would be gathered in groups of about six around a map of their community, with a layer of clear acetate over it.  We'd give them some tools:  the red tape is an elevated metro, the blue tape is light rail, the green tape is frequent bus service.  Here, fellow citizens:  We have 24 kilometers of green tape that we can lay out.  You can trade five km of green tape for a km of blue, or ten km of green tape for a km of red.  Design your own system, but experience the process of making hard choices as you do.

A similar exercise can be done when doing a short-range bus service redesign.  Here, you're just dealing with frequencies of service rather than technologies, so the costs are more obvious.  Red tape is a bus every 15 minutes, blue is a bus every 30 minutes, green is a bus every 60 minutes, so a kilometer of red is worth two km of blue or four km of green.

Self at SOFA wkp  At the end of the work at tables, different groups of citizens would have come up with different networks for the same area.  So we'd put the resulting maps on the wall and I'd lead a discussion about how different groups had solved the problem differently, and get citizens talking to each other about why they approached the problem in various ways. 

At the end, we understood their views, but more important, they understood each other's views.  They also understood the underlying problem facing the transit agency, so they could form more useful and constructive ideas in the future.

In my course, I add a number of other features to this tool, including different ways to analyze the resulting networks, and also ways to humanize the issues.  I also emphasize how the geography of transit generates choices among competing values, which is why citizens and their elected officials ultimately need to make the decision. 

I advocate geographically-based games because the network design problem happens in geographical space, and you can't really see either the obstacles or the opportunities unless you look at a map.  But Community Transit did it differently …

The Community Transit "Transit Values Exercise"

Like many US agencies, Community Transit is having to make a steep service cut this year, around 20% of their total service.  They wanted a way to engage citizens in thinking about the choices that this implies, so they invented their own "Transit Values Exercise."

Instead of talking about geography at all, they forged a set of 15 fictional people who represent slices of their market.  The question for participants was:  "Which of these people should we no longer serve?"

Each person is represented by a card:


The narrative on the left is Cathy's story, but we're to understand that there are many people "just" like her.  Services that meet the needs of these people have a total cost of 7 points (lower right) out of a total of 125 for all the cards.  Icons on the left tell us that Cathy doesn't have a car (or a bike, or a wheelchair) and gives a sense of the diversity of trip purposes that she uses transit for. 

On the right, the five bars describe the consequences, for the whole network, of serving Cathy and people like her.  Note that she lives in a rural community, so it's not surprising that (compared to a whole network that also serves a larger urban area) she tends to need services that have high cost to operate (first bar) and require high subsidy (second bar.)  Like all rural services, the routes Cathy rides cover a large area (high "Coverage") but have very, very low ridership by system standards.  The middle bar, "Efficiency," refers very narrowly to schedule efficiency: the amount of dead running required.

(Obviously there's a little redundancy among these five, since subsidy itself is a result of ridership and cost [as well as fare levels, not modelled here].   Logically, subsidy could have been omitted, but it does tend to pique the interest of more conservative participants.)

Here's another card:


You get the idea.  At the APTA Multimodal Operations Planning workshop on August 17, a roomful of professionals had a chance to play with this tool, choosing different people to "discard" (harsh, but that really is what we're doing) and then seeing how that choice effects the system's cost, subsidy, efficiency, coverage, and ridership.  For example, if you discarded only the patrons associated with low-ridership routes, you can cut almost 50% of the system's total coverage area while cutting only 20% of cost and barely 10% of ridership.  Rural coverage is a lot of area, and almost no riders.

It was clear that the Community Transit exercise would teach citizens about basic budgetary tradeoffs, and obviously it gives those tradeoffs a human face — or at least a set of demographics with a name [though not an ethnicity, an annual income, or a political persuasion].  The bracing task of jettisoning human beings, like deciding who gets the lifeboats on a sinking ship, certainly must have impressed participants with the gravity of the problem, and perhaps, in aggregate over time, such tools could motivate more support for funding sources that could change the picture.

But it's still about geography …

In the last stage of the Community Transit exercise, participants are finally given a map of the fictional community where these people live:

TVE map

Knowing this, we were to make one final assessment about whom to discard.

For me, this part was a problem, because the geography revealed so much more than the game wanted us to think about.  Now, for example, it was apparent that Cathy lives on the way to where Pat lives, so if we "discarded" Cathy but kept Pat we knew Cathy would still have some service.  Yet there was no way for the game model to capture this obvious fact.  I would probably have introduced the map much earlier in the exercise, and thought about how to integrate its geographic information into the "human" discussion about which cards to discard.

Ordinary citizens often assume that a transit agency's service to them reflects an assessment of what they need or deserve.  In fact, this is only true if you're at the outermost end of a line, where the line is serving only you.  Everywhere else, transit efficiency lies in combining the needs of diverse ranges of people onto a single vehicle, and this relies, above all, on taking advantage of situations where multiple people, often with very different demographics, are travelling along the same reasonably direct path.  If you're on such a path you'll get better service than if you're not, even if you are exactly the same person in all other respects.

I wonder if, by steering us away from noticing the way transit combines diverse markets, the Community Transit game may have missed an opportunity to educate about how transit efficiency actually works.  By presenting us with people whose needs are cheap or expensive to serve, the exercise may confirm a widespread and false assumption that transit planning really is about assessing the merits of interest groups and communities. 

That's not what it is at all.  Your transit service isn't a judgment about who you are, but about where you are!  The barriers and opportunities presented by your location (and the location of places you go) are what determine how much service you can expect.  We need to put a human face on our work, and help people understand the human consequences of changes, but we also need to help customers focus on the geometric fact of life.  More than anything else about you, your location matters.

18 Responses to what makes a good “planning game”?

  1. Dan W. August 19, 2011 at 1:43 pm #

    Wow, Jarrett, what an amazing article. I would love to play such a game.
    One other factor not included and perhaps it shouldn’t be for purposes of the exercise is the effect of transit modes on traffic.
    For example, at-grade rail may end up slowing down auto traffic in certain places. Heavy-rail tends to have some sort of grade separation, but light-rail can be built either at-grade or above/below grade.
    Perhaps in an “advanced” version of the game, light-rail could have a below-grade length and an at-grade length.
    Should there be a variable for impacts on auto travel?
    One question. Is there any sizable transit authority you are aware of that said, “Okay, while we cannot move our existing rail lines, our current bus alignments have been in place for decades and need to be rethought from the ground up. We are going to redesign our bus service alignments from scratch.”

  2. Bradley Wentworth August 19, 2011 at 7:07 pm #

    Great overview of planning games Jared; I especially liked the idea of exposing people to difficult trade-offs in planning.
    I broadly agree with your concluding remarks; ‘geography is destiny’ is a byline I use. I’d like to add some complexity though. I would say that transportation planning in the long-term does judge us to a certain extent. To flip your phrase around, we are where are (to an extent) because of who we are.
    In my own city’s recent transit planning imbroglio this map http://3cities.neighbourhoodchange.ca/files/2011/04/map_1.jpg ought to have gotten more attention. (Side note: I agree, in a planning game show the map up front and overlay the human data on it.) What it shows is how the subway lines in Toronto (Yellow, Green, Purple) go through neighbourhoods that have seen a significant rise in income level over the past 35 years. The yellow and green lines are older than that. Don’t live close to a subway? Chances are your neighbourhood has seen a decline in income.
    All planning has to take into account that it is a force of geography itself, both human and physical. Building a massive trunk line along a dense corridor is great, as long as you’ve considered that 30 years from now land values there will have doubled. The people in the Community Transit exercise are where they are in part because of the existing transportation infrastructure, and the moment you change it they (or the groups they represent) will again shuffle around in response.
    Apparently even the location of something as simple as bike share station is affecting real estate prices in London!

  3. Matt August 19, 2011 at 8:03 pm #

    The scenario games you have described sound useful in describing some of the service tradeoffs facing operators. e.g. should we try to serve the whole city, or just the areas where there is high demand. However I don’t think it’s a useful approach for planning of capital improvements.
    Every city has a legacy of transit infrastructure and services. Moving rail corridors is not an option. Bus lines can theoretically be moved, but as Dan W. indicates above, cities don’t usually redesign the network from scratch but instead tinker around the edges. Public involvement in the process is usually limited to those being affected.
    In my experience, there is rarely a slush fund of $X million/year for general capital improvements to the transit network. Five kilometres of frequent bus service can’t be traded for a kilometre of light rail. Instead, a particular project will be adopted because it can be shown to meet the needs of the corridor and provide economic value to the community (measured by a cost-benefit analysis). Also the project must have broad stakeholder support from politicians, planners and the public. A new light rail line is more likely to be competing with funds for a hospital or convention centre than a doubling of service frequency on a bus line.
    So in summary I’d argue that the main challenge is increasing the size of the transit pie rather than dividing it up.

  4. Jonathan Hallam August 20, 2011 at 2:02 am #

    I don’t disagree with the idea of putting the geographical information in earlier – it certainly seems like a useful tool. But it’s worth noting that people’s decision to take transit can be quite binary state. In the above case, Pat drives to a park and ride, the bus for which is an express into the city. That wouldn’t provide Cathy any useful service because she can’t drop off her kids and can’t get to the park and ride base. Cathy needs a bus that stops near to her home and her children’s school. Compared to that comparatively slow multi-stop service Pat would probably chose to drive or vanpool with her bus-friends instead.
    One can imagine being clever – the bus winds through their hometown, before doing a last stop at the park and ride before dashing into the city. But that kind of cleverness is situation specific, and is really something that an expert (like you!) should be employed for. The ‘input’ for that is the tradeoffs between people, which this kind of exercise can start to provide.

  5. Eric O August 20, 2011 at 4:50 pm #

    Forgive the teensy quibble, but I’m having a hard time seeing the “game” in these. Seems the games are simulations of plan-making decisions. For a game (or a good one anyway) you need competitors, the bounds of play, a level start, rounds, and – for a fun game holding the interest participants – a winner. Ideally. When does Kathy get a chance to duke it out with Chris? You need to insert a spinner here. You can’t just arbitrarily elect a winner like gods from Mt. Olympus. That would be no fun.

  6. Miles Bader August 21, 2011 at 7:51 pm #

    @Eric O
    “Game” is a word with a very broad range of meaning … certainly none the things you mention [in particular, competition] is necessary for a good/fun game.

  7. Cap'n Transit August 21, 2011 at 10:58 pm #

    To expand on what Dan W. said, I want to see a game that takes into account the effects of car infrastructure on transit, and vice versa. Ideally, one that treats the entire transportation network as a single system, with private and public components, individual and small group and collective arrangements, rubber tires and steel wheels and feet.
    The goal is to provide the most access to the most people and promote social and economic interaction without treating any groups unfairly, while minimizing emissions, carnage and resource use. Your job is to find the right balance of all these.
    I wouldn’t be surprised if transit staff and consultants can get fired for pointing out that transit competes with subsidized car infrastructure, but that doesn’t mean that you have to pretend that that competition doesn’t exist.

  8. Michael Lewyn August 22, 2011 at 5:26 am #

    I notice raising fares isn’t an option, and that cutting service instead is a given. Is that really such a good idea?

  9. Tom West August 22, 2011 at 6:44 am #

    @Bradley Wentworth :
    I think you have cause and effect the wrong way round. You are implying that subway lines make people near them richer. What actually happens is that only rich people can afford to be near subway lines. (Everyone eants to be near, pushing up property values).

  10. Andrewbnash August 22, 2011 at 7:47 am #

    Great article and comments.
    We developed a game called BusMeister that tries to explain public transport operations. We too struggled with the question of how to make the project feel like a “game” rather than a simulation.
    BusMeister is pretty good, although its user interface needs improvement, party because transit operations is so complex. Here’s a link to the game: http://www.greencitystreets.com/busmeister.
    Here’s a link to a draft “Teachers Manual” that explains the game and how to play in detail (www.andynash.com/nash-publications/2011-BusMeister-in-school-22aug11.pdf).
    We want to use games to teach citizens about public transport so they can help identify good ideas (crowd sourcing) and also provide political support for difficult decisions (taking parking to build a bus lane). So we combine the game with a best practices wiki/blog and a social network so players can comment on how to improve public transport in their own cities. The prototype application is on-line at (www.greencitystreets.com)
    Now we need to refine the system and find a progressive public transport agency who wants to test it … your ideas welcome!

  11. Bradley Wentworth August 22, 2011 at 9:11 am #

    @Tom West
    I agree to an extent, rich people move to desirable neighbourhoods, which probably underlies most of the cause of income increase on the map I linked to.
    However, there are causative factors going the other way too. A strong majority of North American households own their homes, which will be their life’s largest asset. A new rapid transit line increases the value of nearby properties. It also affords better access to jobs. For business owners and developers in particular, the opening of a new rapid transit line can be quite a boon.
    My main point isn’t so much about income, wealth, etc. Rather, it’s important to recognize that transportation planning has long term arbitrary and/or unforseeable effects on the geography of an urban area. If you plan without regard to that, you may end up with a few expensive, high-profile lines that, while very effective and efficient, have the effect of slowly driving away low-income households to the poorly-served periphery of an urban region. A broad network should be planned right from the start, even if the outer edges don’t seem like they justify it right this moment.

  12. Eric Goodman August 22, 2011 at 3:35 pm #

    Thanks for the review Jarrett. Your critique is appreciated.
    We struggled with whether to add the map. Our team decided it would be useful to some visual thinkers even if non-essential to the task. Bringing it in earlier without some specific use for it would have taken up limited “playing” time and been distracting from the main “values” tradeoff conversation we wanted people to have. (sacrifice coverage to become more efficient and preserve ridership, keep coverage but expand subsidy and lose ridership, etc…). This was intended to be a high level philosophical approach to the problem of dwindling resources and not a route planning workshop.
    We also contemplated the type of linkage you suggest – IF Cathy gets tossed THEN Pat must also be chosen – but we felt that would add too much complexity and development time. We were faced with a very short timeframe to produce this and a desire to keep it simple, quick and easy to understand. In the actual exercise, if someone asked, we were able to explain that cutting someone’s service means the service she uses now, not all of the potential service that she could use. For some people there are likely to be other options, especially those closer in to the urban area.
    The exercise does show that a more refined strategy tends to provide better results than just hacking away with one particular approach. So someone tried to “Just cut off the rural areas” – and they cut 50% of coverage and only lost 10% of ridership. That’s great but another group only cut 25% of coverage and lost just 12% of ridership. Which of those strategies you prefer tells us something about what you value and how strongly, and that’s the point.
    Teaching the public about transit efficiency is indeed very difficult and that’s why I read this blog to learn from you and other professionals sharing knowledge here. I agree the things you mention would improve the exercise if time were not an obstacle, but remain very pleased with the reception we’ve had from our local community participants and others with whom it has been shared. Designing games, like transit, often involves difficult tradeoffs of time and resources.

  13. Eric O August 22, 2011 at 6:45 pm #

    Is there anything preventing merging these two games together? Thinking about it, it seems the latter game would serve as a useful evaluative tool to rank the performance of the networks formulated from the network game. The locations of a Cathy or a Chris can be uncovered on the map at the end of the network design game and can therefore be used to rank the service resulting from the networks based on how well they serve the performance profile the hypothetical users. This way you move from the global problem to the particular, and give a personal look to the hard choices of “geometry”.
    I have in mind here the board game “Scotland Yard”. Remember that one? Transit options/coverage/mobility figured largely in that game, with the goal of cornering transit-hoppy Mister X, whose location is revealed at set times in the play. Might be inspirational…
    Of course, look also into the board game Settlers of Catan for ideas on how to craft a game around geography and resources, which also effectively introduces players to the art of balancing competing needs as they go.

  14. Jarrett August 22, 2011 at 10:32 pm #

    @Eric O. In my course, after groups have designed networks, I present them with some sample journeys that certain people want to make, and invite them to route those journeys over their network. This helps people see the strengths and weaknesses of their structure.
    @DanW and others. Transit agencies do sometimes seek to take a “blank slate” approach to redesigning their network. Such a process inevitably finds that large parts of the existing network are in the right place, but that doesn’t mean we’re just “tinkering around the edges.” I would define “tinkering” as making many little changes without thinking about how their impacts affect the total network.

  15. Al Dimond August 24, 2011 at 12:29 am #

    In my high school Global Studies class we played a game where we allocated something like 1000 calories of food per day among a family of five people. After we got in groups and basically all decided that it should be divided pretty equally, the teacher showed us how our answers would all result in everyone dying.
    That is an enormous exaggeration of CT’s funding difficulties, but that’s what their transit priority game reminds me of, in the context of their scope: locals and one-way expresses serving most of Snohomish County except for Everett, which is probably the one place you would serve if you had a choice.

  16. Alon Levy August 24, 2011 at 2:10 am #

    The problem is that games can influence the players with their choices of wording and metrics. Coverage is a particularly egregious example: it assumes that transit should serve a maximum geographic area, without regard to the populations. Cutting out the rural areas may shrink the transit-accessible area by 50%, but it will reduce the population of the transit-accessible area by much less. In this case, there’s an easy fix: a different metric, measuring how many people live within 500 meters of the route instead of how many hectares of dirt, would be more useful. But in other cases – for example, how to assign comparative costs of modes – every choice is political.

  17. Steve Yaffe August 29, 2011 at 7:51 am #

    Hate to quibble, but I see one unlikely option (vanpooling only one day a week) and several missing options – organizing a vanpool for Cathy to get to work with a guaranteed-ride-home taxi option as well as a midday twice/week circulator for her parents to get to the store.
    By the way, Everett has its own local independent transit system.

  18. Nathanael February 22, 2012 at 8:29 pm #

    “I notice raising fares isn’t an option, and that cutting service instead is a given. Is that really such a good idea?”
    Raising fares is easy enough to include in the Community Transit game; raising fares abandons each person who can’t afford the new fare.