Monday, 12 February 2018

Data, the cloud and the internet of things

In terms of data, the cloud and the internet of things there are many examples of how this revolutionising the world around us. Whether this is Fleet Management of TrucksSmart Retail or Smart Motorways - 'things' (trucks, products, cars) are given a data structure. Sensors then report on the real-time performance of these things by push information into this data structure in the cloud. Decisions can then be made based on past and existing performance to improve existing or future performance.

I enjoy running. And one thing that has made this hobby more enjoyable in recent years is 'seeing the data'.

Last year I bought a Garmin Watch and I am completely fascinated by the data analysis it offers.

For example, the runner, the route and the conditions are given a data structure. For example:

  • The runner
    • Age
    • Weight
    • Heart rate
    • Steps per minute
  • The route
    • Distance
    • Elevation
    • Location
  • The conditions
    • Temperature
    • Wind
    • Time of day
It's then the job of runner to go for a run, and a sensor to report the data up to the cloud.

Example 'run stats' below show a distance of 10.07km, a time of 55:14, an elevation gain of 207m (thanks Durham!) and a temperature of 2.8 degrees. The only manual effort from the runner here was giving the run a name 'Four hills challenge' and tagging it by the type of run 'training' (oh, and the small effort of running it!).
Key data
In terms of adding an x axis for duration it is possible to plot elevation, pace, heart rate, cadence against time. In the example below, there is a clear correlation between heart rate and when the extra effort was put in for the hill climbs.
(One criticism for the Garmin Team - The scaling of the elevation is poor for the y-axis. The chart in green below doesn't show how horrible this run really is - 42m to 88m climb out of Durham city centre is agony!)

More detailed data
The ability to break down a race into 'splits' - for example 1km splits - allows the runner to monitor real time progress. For example, if you want to run a 25:00 parkrun. If you do your first km in 5:30 then you know you have to speed things up a bit.
Breaking down the data
The heart rate monitor is also a nice touch to see how hard your body is actually working.
Time spent in each heart rate zone
It's good to see the data from one run, and as every run has the same data structure, there is a lot of power in the ability to look across a number of runs. In the example below the filter query is 'Show all runs classified as 'training', 'between the date range of 1/1/17 and 31/12/17' and sort by 'most hilly'.
Querying across the wider data set
With the data being on the cloud - it means that access is possible from any device. Your watch, your phone or your computer. The same data - just different user experiences based on context and available screen size.

One set of data - access anywhere
Going one step up the data tree - from run, to runs to everyone's runs - if you are prepared to share your Garmin data - then you get lots more data and insight back in return.

One very nice feature is the ability to create 'Segments'. On a route, someone records a start and finish point, and creates a 'Segment' on the cloud. All runners that then run through that segment at any time then have their best time recorded.

The example below shows a lap of the Consett Parkrun circuit. The data shows every Garmin runner who has gone through that segment. It then allows you to benchmark yourself against them. Any part of the world can be set up as a 'running race' where you are racing against virtual ghosts of previous runners - which is pretty cool. In the future, I am sure augmented reality running races will be taking place around the local parks where you can race against yourself!
(Note - A little more care needed for cyclists).
Benchmarking over a specific segment of land
Finally, over a given time period, you can benchmark your efforts against the running world. How did running for around about two hours last week match up to the rest of the world?

Benchmarking against the other runners in the Garmin community
...and how is this relevant to construction???...

Most of my blog posts are around digital construction. So it is a fair question to ask what this has to do with other items on this blog?

Well consider 'the run' equivalent to a building or a manufacturer product. Consider a shared data structure. Consider a sensor pushing performance data to the cloud.

Would it be useful to be able to see how this is performing over time? Would it be useful to compare this performance against the rest of a client's portfolio? Would it be useful to share this data securely with a community and benchmark the performance against similar buildings/products regionally, nationally or globally?

The answer of course is 'yes'.

Data, the cloud and the internet of things is going to change the construction industry. We'll have the most incredible feedback loop which will be used to improve how we operate assets in the existing built environment and how we design when working on the next generation of projects.

- and if anyone fancies a run - hopefully see you at AU5K later in the year!

No comments:

Post a Comment