Algorithms are buildings’ best friends

February 17, 2021Real Estate
I have been involved in commercial buildings’ technical management since… 2017 only. So I am quite a clean sheet without any “outdated” habits in this matter. And what I have seen so far (I would concentrate on software solutions)? Most of advanced proptech using AI (Artificial Intelligence) and ML (Machine Learning) are developed by SMEs while most of R&D (human) resources are concentrated to global giants. Or may-be it just seems for me…

Some time ago a new BMS-version (Building Management System) of one global player was released. 3D vs older version’s 2D – nice! More opportunities to calibrate every single mechanical component – cool! And some more developments. But is it easier to use for operators? Will it increase energy- or maintenance efficiency? Can it validate building’s technical systems’ “health” (incl BMS itself) better?

I do not think so. I have to say that I have not seen any building (even the new ones) with perfectly working technical systems. Many critical faults are not even ever detected nor solved! This often results in shorter equipment lifespan, indoor climate problems and higher energy consumption. And it makes me sad because there are new advanced solutions available in the market that enable to give a transparent overview of the systems’ construction and performance quality!

Sausage & Ketchup

A typical commercial building (shopping mall, hotel, office building) with ca 10 000m2 has ca 500 mechanical components and ca 2000 automation system’s data points to manage HVAC (Heating, Ventilation, Air Conditioning) system to achieve the optimal indoor climate and keep energy costs not too high.

BMS is a very good tool to manage HVAC. To manage by us – humans. As a result, everything seems perfect and a combination of strong owner – smart operator – advanced BMS and efficient mechanical components make our tenants satisfied. BMS is a sausage!

But still indoor climate can be better, discovering technical problems can be easier and faster, operating can be more efficient, more energy costs and CO2 emissions can be saved!

BMS has thousands of parameters for a human operator to observe and calibrate. The larger and more advanced is a building, the more place there is for inevitable human errors and the more complex is the management of the systems. The severity of the problem is increased by the fact that buildings are always tailored with tight construction budget and schedule.

Ketchup is missing! In our case – AI based on advanced algorithms that analyse in detail building HVAC data to detect faults and write the most optimal parameters for achieving comfort with minimal costs 24/7. Free your personnel for trying to be robots; there are real robots for that work already!

Believe me – we (humans) never can calculate fast how for example will a 57,6 Pa pressure instead of 57,9 Pa inside one AHU unit influence heat from a boiler plant, indoor climate in room X and energy costs. I even do not speak about understanding all the rooms, heating, ventilation and air conditioning as a whole. Artificial Intelligence (for example R8 Digital Operator) can.

Sausage can be eaten without ketchup. Vice versa… I would not. Thus there is my recommendation: use real robots instead trying to become a robot yourself. Algorithms came to stay and help!
(The content published in this session doesn’t necessarily reflect the opinion of the GRI Club or its members.)

Imre-Gustav Vellamaa
Board Member