The HVAC ran full tilt on empty floors.

The HVAC ran full tilt on empty floors.

The HVAC ran full tilt on empty floors.

That is the kind of waste a model can see. Energy, equipment, and empty rooms. A heating plant that starts to fail weeks early. A spike that always hits at 3 p.m. A floor nobody uses next to one that is packed. Facilities already has the history. A model reads it.

One plant cut energy 20 percent in six months. HVAC at full blast when the building was quiet. They changed the schedule. Paid for itself in a year. Sensors you already have, plus weather and occupancy, are enough. The team still decides. The rec is the draft.

Fix it on Tuesday, not at 2 a.m.

Vibration, heat, how the machine has been running. A 50-building campus watched 200 chillers, boilers, and elevators. In three months, 15 likely failures. They fixed them on planned downtime and skipped about $300,000 in emergency work.

Badge swipes and booking data show the empty desks. One finance shop found 40 percent of the office vacant most days. They stacked onto fewer floors and subleased the rest. About $2 million a year. That is a space decision, not a science project.

One building. Clean data. Then the next.

A messy file makes a confident wrong rec. Train people on the tool. Write who reviews and who acts. Energy and a longer life on a chiller also cut carbon, which is a slide the CFO and the sustainability group can both sit through.

High bill? Start there. Rising repair spend? Watch the machines. Mystery occupancy? Map the badges. Use a vendor that has done buildings, not a generic dashboard. One system, a number, then scale.

Subscribe to NetNerd AI

Sign up now to get access to the library of members-only issues.
Jamie Larson
Subscribe