Occupant votes tune socially driven HVAC to their comfort

At the federal courthouse in Phoenix, 83% of occupants reported more thermal comfort with socially driven HVAC control, and hot and cold calls to maintenance fell 59%.

  • 1 a person

    83% of occupants reported more satisfaction with their thermal environment, a gain in comfort.

  • 1 an organization

    Hot and cold calls to maintenance fell by 59% over the three-month test, which saved staff time.

  • Not rated society

    The energy savings were modeled for prototype buildings and were not metered.

Depth runs from 1, a convenience or time saved, to 5, a life or a livelihood preserved.

Status
Reported benefit.
Evidence

foundstudiedreportedclaimed

Only the evaluators support the benefit.

Compared with

The building's previous fixed-range temperature control. Measured.

The system’s part
Not isolated. The evaluators compared comfort surveys and maintenance calls before and after installation, which credits the change to the whole system, bursts of air included. GSA writes that the control software learns occupants' preferences by trend analysis of their requests, and the evaluators put "learns" in quotation marks. Neither GSA nor the evaluators show that the software is a trained model.
Limits
Not yet checked.
Costs and harms
Not yet checked.
Later
Not yet checked.

Download data Cite Suggest a correction

Benefit type
Product Personalization
System
Socially driven HVAC optimization software
Made by
Building Robotics, LLC (named in the evaluation as the technology provider)
Used by
U.S. General Services Administration (Federal Building and U.S. Courthouse, Phoenix)
Where
Phoenix, Arizona, United States
Stage
In a pilot
Date
, when the source was published
Kind
Aggregate result
Broad benefit
Controlling Machines and Processes, Analyzing Data or Information
Collections
Scrutiny and findings
Last revised
31 October 2026

Sources

  1. Evaluation report, by an evaluator that the makers or users commissioned.

    U.S. General Services Administration, Green Proving Ground. GPG-025: Socially Driven HVAC for Personal Control, findings, December 2015. GSA's summary of the measurement and verification that it commissioned from Oak Ridge National Laboratory in its Phoenix courthouse.

    Researchers found that 83% of occupants experienced an increase in thermal comfort. Modeling of large office buildings demonstrated average energy savings of 20% (cooling) and 47% (heating) over a typical GSA facility.

  2. Evaluation report, by an evaluator that the makers or users commissioned.

    Howett D, Rizy CG, Wolfe AK (Oak Ridge National Laboratory). Socially Driven HVAC Optimization: Federal Building and US Courthouse, Phoenix, Arizona. Prepared for the US General Services Administration, August 2015 (archived copy). The evaluators' full report adds the method, the occupant surveys and their own description of how the system learns.

    Over extended periods, the socially driven HVAC system "learns" through data trending how to keep occupants comfortable while saving energy, and also learns how to save energy by relaxing set points in unoccupied spaces.

Details

AI system

Workflow

Occupants asked for warmer or cooler air from a web page or smartphone, and the zone received a 10-minute burst of conditioned air. The control software widened or narrowed each zone's temperature range from trends in the requests. The 289,000 square foot courthouse had 250 thermal zones, each served by a variable air volume unit under a central building management system.

Follow-up

Conditions for replication

The building needs variable air volume HVAC with direct digital control, or wireless pneumatic thermostats. IT security staff must approve the software, which runs inside the facility's firewall. Contracts that fix a narrow temperature range limit the energy savings.

Entry details

Sector

Building energy management.

Purpose

Building Robotics built its socially driven HVAC software to let occupants request warmer or cooler air and to tune each zone's temperature range to their preferences while saving energy.

Not yet checked: model, release, provider, role, configuration, inputs and outputs, autonomy, human oversight, AI contribution, beneficiaries, reach, distribution and access, duration, every measurement question, every money question, ownership and control, dependencies, cross-border connections, geopolitical implications, replication attempts, availability, institutional decision, grounds, action and result, success criterion, every evidence question, observed or intended, alternative assessments, related entries, other identifiers, category mappings.

Cite this record

Peter H. Vartanian. (2026). Occupant votes tune socially driven HVAC to their comfort (AIBD-0882). Artificial Intelligence Benefits Database, edition of 31 October 2026. https://petervartanian.xyz/aibd/records/AIBD-0882.html (archived at https://doi.org/10.7910/DVN/AADSII, version 4.0)

BibTeX Data Field definitions