Too much information
In this month's President's Blog, CIBSE President John Field takes a look back at two presentations delivered during the ASHRAE Winter Conference and at a CIBSE Scotland/SLL meeting, discussing performance, big data and modelling.
In the last few weeks, I've been lucky enough to see a series of fascinating presentations on performance, control, feedback, BIM and big data, so I'm going to go through two of them now.
|Alastair MacGregor, Vice President |
The first was by Alastair MacGregor, Vice President AECOM Los Angeles, at Tim Dwyer's excellent workshop during the ASHRAE Winter Conference. It concerned "third generation" sports arena design - in this case for the Sacramento Kings basketball team. The first-generation was passive design supported by brute force plant, the second generation was design for peak demand and the third generation is performance driven.
Well the Sacramento arena has a capacity of 17,000 and they are all online: they've all got high-quality Wi-Fi with an app for their team and for the game and they are hooked into the controls via the app which encourages feedback about their conditions. Alastair talked about the internet of fans - I thought at first he meant the internet of ventilation fans, but actually he meant the internet of sports fans - and they are providing source data for the HVAC and lighting controls: if you get say 2000 people in one area saying they are too cold, then it's likely that that's worth acting on, so it is hooked up to the BMS via some analytics and provides an addition to traditional control paths. My belief is that larger buildings will generally be operated like that in the future.
The second presentation was at an interesting Scotland Region/SLL meeting where Friedrich Wilhelm Bremecker of DIAL presented what could be called the dummies introduction to BIM. His examples were from lighting and hot water system design. His initial point was that Building Information Modelling has four underlying principles:
1. It is a digital representation or model of the real building
2. Each real object will be represented by a digital object
3. Complete information will be exchanged by open interfaces
4. Each participant of the process can access the information which is
relevant for them
We all probably think we know that, but Friedrich cast a novel light on BIM illustrating it rather strangely with Rene Magritte's famous picture of a smoker's pipe which has the words underneath (in French) "this is not a pipe".
|The Treachery of Images by Rene Magritte|
Magritte is right it's not a pipe, it's a picture of a pipe and it doesn't have any weight, it doesn't have a hollow bowl, it's just a bit of paint on a bit of paper and that that's the difference between a drawing and an object, between a CAD drawing of a door and a BIM door object; it's a very revealing example.
The other thing is the importance of open standards and the tendency for a model to be very complicated. There will be lots of proprietary model formats, but there will always be things they can't solve. You need open access and flexible interaction so that information can be swapped with specialist models and applications. As examples, there are the IFC and bSDD vocabulary/frameworks and the DDS-CAD BIM project viewer.
A key aspect is limiting the depth of the model: you don't need to model where each photon is going in a lighting calculation, nor the detailed luminaire construction down to individual rivets. If you did, each project would have gigabytes of information most of which would be unused.
Another aspect is ownership and its protection: these models have such complete information about designs and specifications, that it's possible in principle to basically just rip a whole design or a whole process.
Well these two presentations were pretty different but they reflect on very interesting aspects of performance, big data and modelling. To my mind, they show we are making progress in areas where we have to make progress or somebody else will.