May 4, 2026
4 min read
A good supervisory controller should leave an audit trail
If AI adjusts HVAC plant behavior, operators need to know what changed, why it changed, which constraints were checked, and how to reverse it.
Read field noteField Notes
Practical notes on AI HVAC optimization and building operations.
Field notes for teams evaluating whether AI can safely reduce HVAC energy use in existing buildings, especially central plants already running on BMS infrastructure.
Archive
Latest field notes
Page 2 of 2. 12 posts per page.
May 2, 2026
3 min read
HVAC optimization should be sold on savings, not seats
The outcome of HVAC optimization is not another dashboard. It is lower utility bills, a better-running plant, and accountability tied to measured performance.
Read field noteApr 30, 2026
4 min read
Most HVAC analytics do not change energy bills
Visibility is useful, but energy bills move when recommendations become safe, accepted, operator-visible control actions.
Read field noteApr 28, 2026
4 min read
The best HVAC optimization starts with constraints, not algorithms
In a real building, the first optimization question is not mathematical optimality. It is what the system is allowed to change safely.
Read field noteApr 26, 2026
4 min read
The hardest part of HVAC optimization is earning the right to change a sequence
Facility teams are not rewarded for clever risks. AI HVAC optimization has to prove it understands the building before asking for control authority.
Read field noteNew notes are added as the archive grows; pagination keeps each index page scannable.