Skip to content
gigabiome
Try Plant Check

How it works

Tune the condenser-water loop as one system. Prove it every month.

The deliverable isn't a dashboard. It's a monthly verified report that an owner's engineer, finance team or lender can check.

  1. Sources: BMS / BACnet, Tower controller, Meters, Lab results
  2. Read-only gateway: reads, never writes
  3. Point mapping: you confirm, physics checks
  4. Baseline: weather- and load-normalized
  5. Optimizer: proposes within guardrails
  6. Operator approves: the only path onward
  7. Monthly verified report: kWh and gallons, verified
  8. A dashed branch from the approval: supervised writes to the plant, only if the site asks

Getting started

Onboarding takes an export, not an IT project

Four steps from a trend export your BMS already keeps to your first monthly verified report. The technical detail behind each step follows below.

  1. Step 1

    Send a trend export

    Your BMS already logs chiller power, loop temperatures and flows. Export a month or more as CSV; we send point-by-point instructions for common BMS platforms. No hardware needed to start, no network changes.

    Your timeTypically under an hour

  2. Step 2

    We map and baseline

    We identify every point (the AI proposes, an engineer checks), run the physics checks and build your baseline.

    Your timeOne short call to confirm the point list

  3. Step 3

    Recommendations you approve

    Specific setpoint and water-side recommendations. Your operators accept or reject each one. Nothing is written to your equipment.

    Your decisionAccept or reject each one

  4. Step 4

    Monthly verified report

    kWh and gallons against your baseline, guardrails, and dollars at your tariffs. When you'd rather not send files, a cellular gateway we install, off your IT network, replaces the monthly export.

    You receiveA verified report every month

What we need from you
  • A BMS trend export (or read-only access later)
  • Water-treatment service logs, if you have them
  • Your utility tariffs
What we never do
  • Write to your controls without approval
  • Replace your BMS, your water-treatment company or your qualified person
What we add where it helps
  • Read-only flow meters on make-up and blowdown lines
  • Conductivity sensing where your tower controller doesn't log it
  • A cellular gateway, off your IT network, so reports run without exports

Most exports cover the chiller side well and the water side poorly. We supply and install these, priced separately from the service, so water savings can be measured, not estimated. None of it is needed to start.

Behind each step

What we do with your data

The engineering behind onboarding, and what each phase leaves behind for you to check.

  1. Phase 1

    Map the points

    It starts from your BMS trend export. Point names are mapped to standard names, and an engineer checks the mapping. Later, if you choose, a read-only gateway (BACnet) can read the BMS, tower controller, meters and lab results in place of the monthly export.

    ArtefactAn engineer-checked point map, made with the same mapping step as Plant Check

  2. Phase 2

    Baseline

    Several weeks of data across the load and weather range, modeled against ton-hours and wet-bulb. We state how well the baseline fits, to the industry standard, before any saving is claimed.

    ArtefactA baseline with its accuracy stated: CV(RMSE), NMBE and savings uncertainty

  3. Phase 3

    Recommend and approve

    Condenser-water reset, tower fan staging, chiller sequencing, cycles targets against the scaling index, blowdown and dosing targets, and tube-cleaning timing from the approach trend. Operators approve each change.

    ArtefactRecommendation cards, each accepted or rejected by an operator

  4. Phase 4

    Report monthly

    The verified report below, every month, plus an early warning when approach, fan speed or chemistry starts to drift.

    ArtefactThe monthly verified report

What a baseline has to normalize

Chiller kW/ton moves with outdoor wet-bulb and with load, so one month can't be compared with another until both are accounted for. Here is that spread on a sample plant with no faults.

Chiller kW/ton against load and wet-bulb · fault-free run

CALCULATED FROM THE DATA
Sample data: LBNL's public simulated chiller plant. About the data Hourly averages while running, July. No fit is drawn: a pilot's baseline is fit on the site's own data before any saving is claimed.

Where AI is used, and where it isn't

Where AI is used in Gigabiome, and where it isn't

  • Identify your points

    In Plant Check
    Done by
    AI proposes what each point is; you confirm every row
    Does it produce numbers?
    No. It only labels points. Nothing is calculated until you confirm.

  • Physics checks

    In Plant Check
    Done by
    Fixed engineering rules
    Does it produce numbers?
    No. They flag a suspect mapping; they never change it.

  • Performance figures

    In Plant Check
    Done by
    Standard engineering formulas, applied to your data
    Does it produce numbers?
    Yes. This is the only step that produces numbers.

  • Written explanation

    In Plant Check
    Done by
    AI, citing only figures we calculated
    Does it produce numbers?
    No. If it writes a number of its own, the explanation is rejected.

  • Baseline and verified savings

    In the monthly service
    Done by
    A weather- and load-normalized baseline, verified to IPMVP
    Does it produce numbers?
    Yes, calculated from each pilot site's own data each month

  • Recommendations

    In the monthly service
    Done by
    An optimizer within guardrails
    Does it produce numbers?
    Will propose changes; none reaches the plant without operator approval

The reports

One-off check, or monthly verified service

Plant Check gives you a one-off report from a single export. The monthly service adds the monthly report, verified to IPMVP, the industry standard for measuring savings (how verification works). Below is its format; each section fills with your plant's data.

CALCULATED FROM THE DATA

Plant Check report · Sample plant, cooling-tower fouling (July)

MetricValue
Chiller efficiency (energy-weighted)1.710 kW/ton
Whole-plant efficiency (chillers + towers + pumps mapped)2.909 kW/ton
Chiller kW/ton, median running interval1.655 kW/ton
Chiller kW/ton, worst 10% of running intervals2.042 kW/ton
Tower approach to wet-bulb (median while running)8.7 °F
Tower approach, worst 10% of running intervals10.7 °F
Share of running hours with the mapped tower fan at ≥95% speed85.9%
Lift (median while running)40.9 °F
Chilled-water delta-T (median while running)7.1 °F
Cooling delivered66,248 ton-h
Chiller electricity113,300 kWh
Tower fan energy per ton-hour96.9 Wh/ton-h
Period covered31.0 days

Explanation

WRITTEN BY AI · 2026-09-30 · UNEDITED
This simulated plant runs at 1.710 kW/ton at the chillers and 2.909 kW/ton whole-plant, with pumps and a tower running near full fan speed pointing to the biggest areas to examine.
Get one for your plant →

Report format — filled in with your plant's data each month

Monthly verified report

  1. 01Energy, verified

    Plant kWh against a weather- and load-normalized baseline, verified to IPMVP (the industry standard for measuring savings), with its uncertainty. kW/ton by load and wet-bulb band, so weather isn't mistaken for performance.

    Plant kWhfilled in from your plant's data
    Adjusted baseline kWhfilled in from your plant's data
    Baseline fit CV(RMSE)how closely the baseline tracks your plantfilled in from your plant's data
    Baseline fit NMBEwhether the baseline runs high or lowfilled in from your plant's data
    Fractional savings uncertaintyhow sure the saving isfilled in from your plant's data
    kW/ton by load and wet-bulb band

  2. 02Water, verified

    Makeup and blowdown gallons, cycles of concentration, and savings against a baseline at the same evaporation.

    Make-up galfilled in from your plant's data
    Blowdown galfilled in from your plant's data
    Cycles of concentrationfilled in from your plant's data

  3. 03Guardrails

    Scaling index and chemistry within limits, biocide residuals, and Legionella results where they're required. Savings that cost fouled tubes or a compliance breach don't count.

    Scaling indexfilled in from your plant's data
    Biocide residualfilled in from your plant's data
    Legionella resultswhere requiredfilled in from your plant's data

  4. 04Actions and acceptance

    Every recommendation, whether operators accepted it, and what changed as a result. Advisory systems only save what people act on.

    Recommendationsfilled in from your plant's data
    Acceptedfilled in from your plant's data
    Rejectedfilled in from your plant's data

  5. 05Dollars, reported last

    Energy, demand and water-and-sewer savings at your actual tariffs. Reported last, and only once everything above holds up.

    Energyfilled in from your plant's data
    Demandfilled in from your plant's data
    Water and sewerfilled in from your plant's data

Report format — filled in with your plant's data each month

One recommendation, as an operator sees it

Change type (one of)

  • condenser-water reset
  • tower fan staging
  • chiller sequencing
  • cycles target
  • blowdown and dosing targets
  • tube-cleaning timing
Rationalecites the calculated figures behind itfilled in from your plant's data

Guardrails checked

Scaling indexfilled in from your plant's data
Biocide residualfilled in from your plant's data

Operator decision

Approve or reject, drawn here as an outline only

Every decision is logged in the monthly report

What we don't do

  • Interlock 01

    Write to your equipment without approval. Advisory first; supervised setpoint writes only when a site asks for them.

  • Interlock 02

    Replace your water-treatment program or your qualified person. We measure and recommend; they stay accountable for chemistry and compliance.

  • Interlock 03

    Claim savings without a baseline, or report them without the uncertainty.

Talk to us about a pilotTry Plant Check