Theme Park Crowd Control Book?

Inspired by a trip to Disneyland yesterday, I was hoping that fellow Dopers could recommend decent book(s) dealing with crowd control and/or theme park management. Of course I tried Amazon, but I struck out there.

Specifically, I was wondering if there was something out there on the processes of managing the queue lines - something recent enough that it would deal with newer techniques like Disney’s FastPass, and other things they use to keep the lines moving and the people flowing.

Note, however, that I am not limiting responses to those dealing solely with the Mouse. Anything addressing the subject is fine.

Thanks,

  • PeterWiggen

Write Disney.

Seriously.

They’re usually quite nice about that kind of thing.

It sounds like the type of thing that a computer could simulate easily enough. Not entirely different from studying traffic patterns on a freeway system or the behavior of large groups of insects. What’s a generic name for that type of study?

Sounds like you’re looking for information on queuing theory. Amazon has some books on the subject.

Bosda — I’ll try contacting Disney to see what they’ll give me. Good idea.

headshok, that’s almost exactly what I was looking for! Thanks!

  • Peter Wiggen

Crowd control and other logistical issues at Disney World, et. al. is indeed an interesting subject. As far as I can tell (and I’d love to see a book!), when you see a planter at Disney World, you can be sure that the planter is located here and not over there because somebody sat down and figured out the optimal placement to facilitate the movement of the crowd.

Fascinating.

It sounds like you’re looking more for theory-type stuff, as opposed to practical-type stuff. However, continuing my long tradition of adding posts to threads that sort-of-but-don’t-quite-address-the-question, I offer:

The Unofficial Guide to Walt Disney World, by Bob Sehlinger. Sehlinger, along with some scientists, has devised a computer program that will spit out the optimum touring plan for various types of families visiting Disney World. For example, families with small kids will likely want to do different things than families with older kids. Folks who have no kids will likely want to do different things than either of those. Et cetera.

Sehlinger doesn’t go into specifics about the computer program (although the scientists he mentions in the book might be willing to do so), but he claims that field-testing the touring plans generated by the computer have cut down the amount of time necessary to visit all the attractions by three hours or more. The program does this by predicting traffic patterns in the park based on historical data and guiding the tourist to attractions that aren’t as crowded as others.

BrotherCadfael — that’s almost exactly what I was interested in: Some sort of explanation as to why they put this here, and that there. Why do they configure these lines this way and those lines that way. I guess I am looking for both theory and practical application.

Sauron — I came across Sehlinger’s book on amazon as well, but like you said it doesn’t look like it gives much explanation.

And after looking at the queuing theory books available (and remembering days of statistics and calculus) I think I am looking for something a little bit more practical.

Anything out there like Queuing Theory for Dummies? Queuing Theory as Applied to Theme Parks? Anyone have any other ideas?

  • Peter Wiggen

Bffff.

While the new parks avoid crowd congestion better than the old, I can attest that sheer inertia keeps some badly designed bottlenecks persistently crowded, e.g., the narrow passage in front of the Haunted Mansion (Orlando). That bottleneck has always been there and has always been bad for the past 28 years since I’ve first experienced it.

Fast Pass has helped. Knocking down that useless restaurant opposite the Mansion would help moreso.

Not to mention all the shops that get in the way of good flow because they want you to buy stuff.

Peace.

I hope you’re not dissing the Liberty Tree Tavern, because that’s where Mrs. HeyHomie ate a meal that was so good she actually cried! OK, maybe it wasn’t the food itself, but the whole experience. It was her first trip to Walt Disney World; it was 11:00 and, thanks to my expert planning, we’d ridden three mountains, a mansion, and a space ranger spin before the place was even starting to get busy, and we were having a sit-down meal at an upscale restaurant. She was a bit overwhelmed.

A little bit off on a tangent, but the excellent book Why Is the Foul Pole Fair?
which is an exploration of trivia revolving around a day at the ballpark - has lots of fascinating information, much of it concerning logistics–like the optimal number, placement and gender distribution of restrooms, etc.

I think moriah must be referring to the Columbia Harbor House, or maybe the Pinocchio Haus.

My wife and I ate at Liberty Tree Tavern a couple of years ago, and I agree – I was pleasantly surprised at the quality of the food.

I wanted to put in a plug for Sehlinger’s book too (for those going to the park). My mom picked it up before a family vacation and planned out when we would hit each ride, and everything worked out really well. We would hit the rides at the off peak times and it significantly cut the time we had to wait.

As for just learning more about queue management, if you can’t find anything on specifically, maybe check out an operations management text book. There was a whole chapter on queue management that was covered in an operations management course that I took.

Hi PeterWiggen,

I write the computer software the Unofficial Guide uses to create its touring plans. (I’ve also been a big fan of Cecil’s writing for more than 15 years.)

Thanks very much for the comments on the Guide. As Sauron speculated, I’ll be happy to answer any high-level questions about how the software works. Feel free to drop me a line at testa@ncat.edu.

For those of you interested, my Masters thesis describes an early version of the code, and explains some of the potential pitfalls that one may encounter in touring plan construction. I’ve also co-authored a couple of academic papers on some techniques for solving these kinds of scheduling problems. I continue to work on the general problem as part of my PhD.

With regard to the question regarding FASTPASS, the patent for FASTPASS is available for reading at the U.S. Patent Office. That’s a good explanation of how it works.

As to the question of theme park design and managing queue lines, Disney employs a number of techniques beyond FASTPASS. One such technique is to have the “headliner” attractions at opposite ends of the park, to better distribute crowds throughout the areas. That’s why Space Mountain, for example, is at a distance from Splash Mountain (Big Thunder is another story).

If I may throw a semi-plug in for the Guide here: FASTPASS alone isn’t as useful as a good touring plan for minimizing your waits in line. Using the computer software, we’ve created touring plans that have allowed individuals to see every attraction, show and live performance in the Magic Kingdom in a single day. Some folks have done this without using a single FASTPASS. The current record for a “complete park” tour is 41 different attractions in 10 hours, 40 minutes. If anyone’s interested in trying to beat this record, let me know.

Thanks for the opportunity to discuss this.

Sincerely,

Len Testa

Well shoot len!!! You’re like some sort of celebrity! (Seriously!) Thanks for speaking up and letting us know what a resource we have on board!

Let me just run a few questions by you: (or anybody else)

(1) You mentioned spreading out headliner attractions, and we already covered FASTPASS, but what are some other techniques that Disney uses for crowd control and queuing? And not just hi-tech stuff - what about lo-tech stuff like trash can placement?

(2) The corollary to that - while we have you here - is how do you counteract those techniques in your program? And I’m not looking for confidential information - just generally how do you handle Disney’s techniques?

What I really mean is this: It seems like Disney puts in detail after detail to control the people in the park. To herd them in an orderly manner. And while the techniques they use might be optimal for crowd control, it’s doubtful that they are beneficial for an individual trying to hit every ride. I’m interested in how you counteract Disney’s tactics.

(3) And even more off-topic from the original post: With respect to the Guide - how much assistance / cooperation does Disney give you in your development? Do they provide things like ride numbers or detailed distance maps?

Thanks!

  • Peter Wiggen

Hi PeterWiggen,

Thanks very much. The Guide is a lot of fun to put together. I hope it’s fun (and useful) to read.

(1) A paper you might be interested in reading is “Managing Capacity and Flow at Theme Parks” by Reza Ahmadi (Operations Research Volume 45, Number 1, January-February 1997). It’s a good overview of the techniques used in another theme park (Six Flags Magic Mountain), and many ideas apply to Disney as well.

Disney generally knows the way in which their guests tend to tour their parks. That’s why, for example, you’ll see Toontown and Adventureland in the Magic Kingdom open an hour later than the rest of the park most days. Disney knows that their guests don’t get to those lands until after touring somewhere else, so it doesn’t pay to operate those attractions for the first 60 minutes the park opens.

I think one of the most common techniques Disney and others use for managing crowd flow is the timing of live performances. A popular show such as Beauty and the Beast at Disney-MGM Studios will draw thousands of people to one corner of the park three to five times per day. Disney also puts small-scale “street performers” around each park many times per day, drawing in hundreds of people at a pop. Finally, parades are another useful way to gather people at pre-determined points in the park.

You’ll notice that when the park is very busy, such as at Christmas, the number of live entertainment offerings increases dramatically. You can’t swing a child without hitting some costumed character from Disney’s latest film.

I’m sure Disney’s put some thought into trash can placement, as well as a host of other park features. But as anyone who’s ever encountered the stroller flood between Fantasyland and Liberty Square in the Magic Kingdom can tell you, they’re not in complete control of everything. As with any large-scale endeavor, what Disney wants to build and what ends up getting built are often two different things. A good website that details how the parks ended up the way they are is JimHillMedia.com.

Finally, I might add that we’ve seen instances in some theme parks (not necessarily Disney’s) where the posted wait times of some attractions have been set artificially high at certain times of the day, presumably to dissuade people from queuing up. I’m not going to speculate on motives, but it’s happened on a regular enough basis that I doubt it’s coincidence.

(2 and 3) We collect our own data for almost everything, rather than rely on someone else to tell us what’s happening. Disney does tell us approximate capacities for the attractions, but we measure everything else. For walking distance between attractions, we count paces between attractions and use Dijkstra’s Algorithm to make a fully-connected graph. For ride times, we ride the rides over and over again.

To measure wait times, we send teams of researchers to each park several times per year. The researchers measure the wait times at every attraction, show, FASTPASS booth and restaurant every 30 minutes from park opening to park closing. We’ll also send people into lines to make sure, for example, that a 20-minute posted wait time is actually 20 minutes. (Disney, I’m told, likes to keep the actual wait time around 80% of the posted wait time. A customer satisfaction issue. Dumbo’s posted wait time, IMHO, is pure fiction most days from midmorning on.)

On any given day, each researcher will walk 18-24 miiles and collect 500 pieces of data. We employ a professional statistician (a bloody good one, I might add) to help with the analysis.

We’ve been collecting this kind of data for more than four years now, so we have a pretty good idea of what large numbers of people do in the parks on a regular basis. Feeding these data into the software, we can identify strategies that help avoid standing in line. I wish I could boil all this down into a single ‘ah ha’ moment, but it’s been years of work to get where we’re at.

Hope this helps. Let me know if there’s anything else I can do.

Sincerely,

Len Testa

Well, might I suggest playing a game? Roller Coaster Tycoon is all about implementation of queuing theory.

The Columbia Harbor House is opposite the Haunted Mansion, and backs up to the Hall of Presidents - it does create a bottleneck because of the overpass/overhang/fake bridgey-looking thing. Pinocchio’s restaurant is next to ‘it’s a small world’ (currently closed for much-needed major overhaul), and IIRC, the Liberty Tree Tavern (across from the Hall of Presidents) is now a reservation-only character-meal restaurant (no longer just a ‘drop-in’ site).

len,

A quick word of thanks – I’ve become a huge fan of Sehlinger’s books over the years. You guys have saved me a metric buttload of time and money in my last few visits to DisneyWorld. It’s a pleasure to interact with someone who has had a hand in producing those books.

I can’t even begin to comprehend the amount of research you guys do to put the books together. On the plus side, at least it’s not sitting in a lab trying to get tse-tse flies to mate or something.

Hmm… How about a virtual Disney Word game? You set the parameters (time of year, time of day, holiday attractions, etc, and the game simulates the typical crowd. The object is to do all of the attractions in the minimum time.

At each attraction you see the ride from the viewers’ point of view, and then between times an arial view of the entire park.

Throw in pee breaks, food stops, crying kids, heatstroke…

Maybe this isn’t such a good idea.