How do they figure out queue times?

[Log Flume's Dennis]

it was the greatest thread in history. The storyline arc of the guy who wanted to create a perfect utopia in block form, that instantly crumbled when he became a full-blown dictator at the slightest hint of power. It was beautiful.

 

[imanautie]

it was the greatest thread in history. The storyline arc of the guy who wanted to create a perfect utopia in block form, that instantly crumbled when he became a full-blown dictator at the slightest hint of power. It was beautiful.

What post is this? Link or screenshot?

Sent from my Swift 2 Plus using Tapatalk

 

[Log Flume's Dennis]

it was thread that was something like "Who wants to create a Minecraft Alton Towers and live in it and manage it" and he had a list of jobs, then a couple of people wanted the top job, which he took exception to. The rest is hazy but I think a genocide happened

 

[Rob]

If this thread doesn't stay sensible then it may too be locked!

 

[DiogoJ42]

it was the greatest thread in history. The storyline arc of the guy who wanted to create a perfect utopia in block form, that instantly crumbled when he became a full-blown dictator at the slightest hint of power. It was beautiful.

You should have been around for the infamous "would the world be a safer place if Hitler won the war?" thread. Now THAT went down in forum history....

 

[MattyH]

https://towersstreet.com/talk/threa...a-safer-place-if-hitler-had-won-the-war.1251/

 

[Lurker]

One way in which Thorpe tried to combat was trialled on Saw. Basically, the cameras which oversaw the queue line had some software installed (I believe it was called Headmapper or similar). As the name might suggest, this maps people's heads as they go through the queue, and can log how long they're in the queue for. This can also take into account Fastrack and RAP systems, and work them into it. This can then throw up a very accurate estimated queue time based on how many people are in the queue. You may have seen this actually; the queue time for Saw was advertised to the minute (eg 53 minutes) during these trials.

In my experience, this technology was largely accurate, but has some problems:

-It needs constant, full coverage of the queue line (ie: cameras have to be in a fixed position and can't move, and can't have other things to focus on). Most cameras at any theme park will be focused on safety critical areas (parts of ride areas and pathways), with queue lines only in view at the side / background.

I don't see why this is true. [Requires cameras throughout the queue line]

If it is truly mapping people as they go through the queue it only needs 2 cameras, one at the entrance and one at the cars which are being borded. Any faces successfully tracked could be marked and an average time taken.

If it is counting the number of people in the queue it could use a simple semaphore. It counts people in past the entrance, and counts people out (boarding the ride).

Simpler still, assuming it is possible to get dispatch information (from the ride systems or from image recognition of cars leaving the station) using an average number of people per car it's possible to estimate to the nearest person?

I see the disadvantage of course of not having coverage of the whole queue, as there is no error correction. If it miscounts upwards, that would remain wrong until the system resets, as you can't recount the whole queue. But I think it's possible to work around.

I think facial recognition is the best way to go, and one assumes it's possible to get financial assistance with something like that, as I doubt all of the shopping centres I've seen using facial recognition do it by themselves. Given the predictability of where people will stand it should be easy to do.

 

[JoshC.]

I don't see why this is true. [Requires cameras throughout the queue line]

If it is truly mapping people as they go through the queue it only needs 2 cameras, one at the entrance and one at the cars which are being borded. Any faces successfully tracked could be marked and an average time taken.

If it is counting the number of people in the queue it could use a simple semaphore. It counts people in past the entrance, and counts people out (boarding the ride).

Simpler still, assuming it is possible to get dispatch information (from the ride systems or from image recognition of cars leaving the station) using an average number of people per car it's possible to estimate to the nearest person?

I see the disadvantage of course of not having coverage of the whole queue, as there is no error correction. If it miscounts upwards, that would remain wrong until the system resets, as you can't recount the whole queue. But I think it's possible to work around.

I think facial recognition is the best way to go, and one assumes it's possible to get financial assistance with something like that, as I doubt all of the shopping centres I've seen using facial recognition do it by themselves. Given the predictability of where people will stand it should be easy to do.

It's a good question, and one that sadly I can't really answer.

I'd imagine there's some sort of benefit / need to having a continuous track rather than a discrete track of guests. As you say, error correction is probably one of them. Possibly that it allows more data to be gotten too? There's possibly more.

Aside from that, I can't give any insight into the why - that's just what I've been told!

 

[imanautie]

It's a good question, and one that sadly I can't really answer.

I'd imagine there's some sort of benefit / need to having a continuous track rather than a discrete track of guests. As you say, error correction is probably one of them. Possibly that it allows more data to be gotten too? There's possibly more.

Aside from that, I can't give any insight into the why - that's just what I've been told!

You can see which bits of the queue might need redesigning.

Sent from my Swift 2 Plus using Tapatalk

 

[Lurker]

You can see which bits of the queue might need redesigning.

Sent from my Swift 2 Plus using Tapatalk

Because queues frequently get a redesign? Best we can hope for is them adding a gate to save us a walk, but even that isn't exactly likely.

Thanks for your response @JoshC.

Very curious about this now, and also curious about how accurate is the right amount. I don't think guests want to the minute, as queues change too much too quickly. I think an overestimation to the closest 10 is quite fine, nobody cares if a queue goes 10 minutes faster.

 

[Godscrasher]

I think a pressure pad would work better. Just like what traffic lights and car park barriers have underneath the tarmac.

They're not pressure pads. There's wires underneath the tarmac, when a vehicle passes over it the magnet detects the metal and creates a loop informing the control panel.

 

[Godscrasher]

It would be easier than a turnstyle, you would just need to lay it under the block paving.



They can usually work out from the weight pressing on it, how many people are standing on them. While not perfect it does work pretty well. They are pretty clever actually, they can even differentiate between two children standing on it rather than a single adult, by the amount of pressure being applied at any given area.

The question is though, is there a need for such a thing? Probably not. Not unless they wanted to keep the data for something else.

I do wonder however, if they artificially adjust que times to try and help spread the crowds around the park on busy days. As people may be put off by a long que time and go somewhere else. Seeing as the times are on boards around the park.

Sorry like but the info you're providing above about pressure pads is rubbish. Please post something to back this claim up. Pressure pads on all queue lines will have an astronomical cost to install and maintain.

My take on queue lines is that Alton encourage people to download the app, now not everyone downloads the app but I can guess a fair majority of visitors do to enable them to navigate the park initially. The app will use your data to know where you are in the park and for how long you've been in a set queue for. Think how google maps work and how it calculates the traffic on roads based upon how fast or slow traffic is moving. A simple calculation can then be done and a decent estimate for the queue will be displayed.