ers will be able to test the opera-
tional outcomes of any scenario.
“You can organize the data and
the way systems operate so you
have a digital twin of a building or
campus where you can test differ-
ent strategies and scenarios over
an entire year in exactly how the
building would respond,” says
Awad. “You can see how that will
This type of technology is still
emerging, but it’s sure to be a
game changer when this kind
of intelligence becomes widely
Justin Feit justin.feit@
buildings.com is Associate
Editor of BUILDINGS.
all geographic areas leads to more
opportunities to improve efficien-cies in all facilities. Sharing data
widespread allows for more AI
strategies to be employed.
One possibility for future
machine learning technology is
creating a digital twin of a building. When this level of analytics
becomes available, facilities manag-
Don’t fall behind when it comes to applying machine learn- ing in your facility. Employing analytics with the mass of data collected in your facility can help you cut costs across the board.
There are five steps key to getting the most out of machine learn-
ing, according to Ash Awad, Chief Market Officer at McKinstry, a
design, build, operate and maintain firm. Follow this blueprint to
optimize your building through data analytics.
1) Unlock Data in the Built Environment
The more data you have access to, the greater the insights you will
yield. Using a networked control setup with your building systems,
you can begin harvesting data.
“If I were a facility manager, unlocking data effectively and effi-
2) Analyze Data
ciently would be the No. 1 thing I’d be seeking out,” explains Awad. “I
would work to unlock more data than I thought I needed.”
If you can successfully unlock data from all of your building sys-
tems, you give yourself many more opportunities to identify ineffi-
ciencies across the board.
Once you have access to a wealth of data, start putting it into con-
text. Analyzing the data will help you develop key actions. Your con-
trol system will provide many insights, but you can look at the data in
some instances on your own.
“If you get the data out, you can then start to do analysis on the
data, some of which you can use outside firms to analyze and some
you might be able to do yourself,” explains Awad.
For example, your control system might not tell you whether
equipment is cycling on and off against a schedule, notes Awad.
Looking at this data over time might eventually reveal that some
pieces of equipment are cycling on and off or are staying on when
they should be off. Ultimately, it’s a practice of recognizing patterns
you typically wouldn’t.
3) Sharing Analysis and Data
More detail about the bottom line is always welcome when meet-
ing with your organization’s upper management. Getting the most
out of machine learning involves sharing this information.
“As a facility manager, you might have information that your finan-
4) Create Action
cial folks might be interested in,” says Awad. “You might have data
and analysis for utility and energy costs that you can put into reason-
able equipment replacement schedules that are not just, ‘I think this
equipment is old and failing.’ You can point your findings to other
parties so they can run their own analysis.”
The data and analysis revealed through machine learning aren’t
just valuable to your own operational goals. This data is also valuable
to other stakeholders in your organization.
If you properly enact the previous three steps, you’ll be able to
capitalize, and the savings will begin. Once your data is telling the
right story, it’s time to take action and significantly improve your
“Where analytics has failed in the past is that there has been too
5) Change Behaviors
little data or the data has not been fully understood or utilized well
enough,” says Awad. “So when it comes time to act, no one can
really act robustly on that information stream. If you get the right
data and analytics, and you share it with the right people, you can
actually act on it.”
Machine learning helps the built environment run collectively,
which is the Holy Grail for facilities, explains Awad. The right applica-
tion of machine learning can improve total operational efficiency –
not just energy – by 50%, he adds.
As you continue to take action with the insights you have received
from machine learning, you can create positive cultural shifts in
your organization. With machine learning, you will also learn a lot
about your building. The cumulative actions you take will eventually
become improved behaviors.
“A lot of resources and energy are being wasted, so if we have
enough of this information and action, we can start to see patterns
that change standard operating procedures,” says Awad. “In totality, we’re trying to change behaviors by making both buildings and
Machine Learning: 5 Steps to Optimize
Your Facility with Data Analytics