[OpenTRV-interest] Count crowds with Bluetooth Connectivity Kit

[Simon Hobson]

I wrote:

> So if you are in the business of counting people, I don't think X is hard to determine (and it'll vary by situation and demographic). Once you know X, you just knock up a device to count Y and you've got a pretty good idea of total numbers. Not "there were 54,157" people there" accuracy, but at least "there are 50-55k people there" levels of accuracy which is probably more than accurate enough for this sort of situation.

Just to add, this method has a really really useful additional option.
Traditional methods will only tell you that "about X,000" people came" and the best you'll know is who came in via which entrance. With a bit more logging and data analysis you can track the BT identifiers an work out what routes people took. Again, not to high levels of accuracy, but a fairly good indication of how many came by tube, which stations they used, which way in they used (I'm thinking of something like the Trafalgar Square celebrations), how many came off the tube and went straight in vs how many went via a bar or food outlet, and so on. And of course, when it's all over, where they all went.
That sort information is hugely useful to planners.

<cough> decades ago when I was still at school, I got to do a traffic survey for the local council. This was long before ANPR and stuff like that - it was groups of us sat in a site hut at pre-defined locations counting vehicles by type. In addition, they'd setup "checkpoints" where the police would pull over a small subset of vehicles for interview - where have you come from, where are you going, what's the purpose of the journey type of stuff - to get an idea of what the routes were that people were using.
That's very labour intensive. IIRC there were something like 4 to 6 of us at each point, plus police and council people - and multiple counting points round the town. Other than finding out about purpose of journey, pretty well all of that can now be done automatically with strategically placed cameras, ANPR, and data crunching.

None of this is about "Mr Smith does ..." This is all about "A thousand people do ..., B thousand do ..."