Skip to content

SCADA-informed emissions & operational intelligence

Methane reductions, lower operating costs, more throughput and fewer unplanned trips, through operational efficiency.

Using only the SCADA data you already have.

AroIQ turns pressures, temperatures, flows and valve states into operational and emission events: compressor trips, process upsets, venting, flaring. Each comes with its measured impact (volume, downtime or cost) and a likely cause. Ranked by what they cost, so your teams fix the few that matter. Nothing to install.

See it on your data How it works

Detect · Quantify · Explain — every event. Triage — every site.

2018

Founded out of the FluxLab at St. Francis Xavier University

7,000+

North American facilities supported by our modelling and data work

10+

Operators across Canada and the US

0

New field hardware — a historian connection is the whole footprint

AICPA SOC for Service Organizations

SOC 2 Type II

AroIQ in scope

For operations teams

Find what costs you production. Fix it once.

AroIQ finds the events that cost you production in the SCADA you already have, counts the throughput each one deferred and ranks them by what they cost. Repeat failures rise to the top with a likely cause, so they get fixed once instead of chased every week.

What it finds

  • Compressor trips
  • Process upsets
  • Hung dump valves
  • Unplanned shutdowns
  • Deferred throughput

Station 07 · Unit 2

Last 7 days · 3 trips

Illustrative

3 trips , grouped into 1 pattern

Compressor trips on high discharge temp

Likely causeAftercooler fouling on hot afternoons

Act now

Increase revenue

  • Production lost to compressor trips and shutdowns shows up as deferred throughput, next to the emissions they cause.
  • Flared gas is product. AroIQ shows when a flare is burning sales gas — and why.
  • Setpoint and operating changes measured for what they did to throughput.

Lower costs

  • Fewer call-outs and less windshield time: the cause is known before the truck rolls.
  • Maintenance and capital aimed by data, not by the loudest alarm.
  • Levers: put your own cost on an event type, like a dump valve replacement, and size the replace-versus-observe call.

Mitigate risk

  • A hung dump valve blows gas through to the tanks — venting now, a spill risk next — and it shows in SCADA first.
  • VRU back-pressure building before a failure, flagged as a precursor condition.
  • Repeat failures, grouped: the same trip on the same unit becomes one pattern with its ranked likely causes, fixed once.

Make operators' lives easier

  • The few events worth acting on instead of thousands. An end to event fatigue.
  • Less manual vent and flare logging — the record builds itself from measured events.
  • Plain-English cause notes on every event, so the morning call starts with answers.

Throughput up. Emissions down.

Emissions are a leading indicator of operational health. The SCADA patterns that find a vent also read what's behind it: the trip, the back-pressure, the setpoint drift. Operations and environment shouldn't have conflicting targets — fix the mechanism and both numbers move.

For emissions teams

Emissions measurement is not the bottleneck. Acting on it is.

What emissions teams have today

  • Measurements that say where gas escapes — a leak list, not a cause.
  • Alarms that say something happened — rarely which equipment, or why.
  • Emission factors standing in for what actually vented — nothing a work order can act on.

What changes a maintenance plan

  • The recurring cause, ranked by what it costs.
  • Evidence a maintenance team trusts enough to budget on.
  • A measured before-and-after, once the fix is made.

A tonne is avoided when someone changes something.

AroIQ tells you which change to make first, and what it's worth.

Beyond detection

Most emissions tech stops at detection and measurement. AroIQ keeps going: quantification and localization, causation, and triage — above the line is where the value is.

  • A maintenance plan
  • Early warning
  • Less event fatigue

  1. Triage

    What to act on, and what can wait.

    a plan you can fund

  2. Causation

    Why it happened.

    a cause you can fix

  3. Quantification & localization

    How much, and on which equipment.

    a number you can defend

  4. Measurement

    When, and for how long.

  5. Detection

    Something happened.

Emissions: measured, not assumed.

Per-event volumes instead of whole-facility assumptions. Fixing recurring causes — flaring, tanks, engines, start-ups — is where real abatement lives.

Fingerprints, not thresholds

Every methane event has a fingerprint.
AroIQ knows them.

One facility can have 500+ sensors reporting every minute. To the eye it's spaghetti. AroIQ learns each site's normal from its own history, and the shapes that break from it. A blowdown doesn't look like a separator vent. A stuck dump valve doesn't look like a PSV lift.

Learn a fingerprint once and AroIQ finds it every time — across your whole history, and continuously from here on.

Built in-house by machine-learning data scientists.
This is not your ChatGPT.

  • Detect continuously on every SCADA site
  • Rank the likely causes: equipment, operating condition, underlying cause
  • Quantify and audit every event
  • Read the operational story in the same data

No new field hardware. We read the process.

Events, likely causes and volumes, all from the SCADA data you already collect. AroIQ ranks what to fix first.

For both teams

How AroIQ works

From raw signal to a decision the operator makes.

Six steps from historian to decision, each one inspectable. Step two is the one most tools skip.

Inside the engine

Step 01Ingest

Send everything.

Every tag, the finest interval you have. Historian exports or a live feed — no tag list needed to start.

Step 02Clean

Self-tested before it's trusted.

The model hides readings it already has, fills them back in and scores itself — thousands of times — before it trusts a gap fill.

Step 03Fingerprint

Learn normal. Find what breaks it.

Any recurring pattern in any sensor becomes a signature. Tag names don't matter; behaviour does.

Step 04Match causation

One confirmation, every similar event.

Matched against a causation library built over years with facility engineers. An analyst confirms the first; the model carries it to every event like it.

Step 05Quantify

A volume, with its range.

A 95% range, your methodology or ours. GHG and VOC separate, every assumption recorded.

Step 06Triage

Costed. Ranked. Routed.

Events grouped into patterns, costed against your KPIs and ranked — then sent to AroIQ or your own stack.

Triage → patterns → ranked causes

Triage first. Then the cause. Then the fix.

Nobody fixes thousands of flare spikes. Operators fix one mechanism.

AroIQ groups every event into the pattern behind it, puts your cost on each one and ranks them across every site — so your team can fix the ten things that matter instead of chasing the hundreds that don't.

Triage

All sites · last 12 months · patterns

Rank by

Sorted by count — the noisiest patterns first.

Ranked by cost — the few that matter first.

  1. 01

    Dump valve hung open

    Separator · Battery 3

    Likely causeSand on the valve seat

    9events 41.6t CH₄ $214k/yr Strong 0.91
    Act now
  2. 02

    Compressor trips on high discharge temp

    Station 07

    Likely causeAftercooler fouling on hot afternoons

    31events 12.4t CH₄ $168k/yr Good 0.82
    Act now
  3. 03

    Slug-catcher dumps flashing in tanks

    Battery 5 · Tanks

    Likely causeHigh-pressure dumps, no flash stage

    212events 27.3t CH₄ $96k/yr Good 0.77
    Watch
  4. 04

    Planned blowdowns — segment B

    Station 07

    Likely causeScheduled maintenance depressurizations

    48events 18.2t CH₄ $41k/yr Strong 0.94
    Normal operation
  5. 05

    PVRV lifts at normal setpoint

    Tank farm · Site 12

    Likely causeTank breathing on daily temperature swings

    640events 4.9t CH₄ $12k/yr Fair 0.63
    Watch
Top 5 of 37 patterns · 14 sites Illustrative

  1. 01

    Triage

    Every site in one ranked list: where to look first.

  2. 02

    Patterns

    Repeat events collapse into the one mechanism behind them.

  3. 03

    Causes

    Ranked likely causes, each with its confidence and its SCADA evidence.

  4. 04

    Costs & levers

    Put your own cost on each event type, then decide: replace, repair or keep watching.

  5. 05

    Measured change

    The fix goes in. AroIQ measures the before and after.

Walk through the workflow

Built to be trusted

Your SCADA never leaves your rules.

Every number shows its working, and every engagement starts with a technical and security session with your IT team.

How data moves

Regulators & frameworks we work with
EPA · US state regulators · ECCC · AER · BCER · Saskatchewan ER · OGMP 2.0 · MiQ

AICPA SOC for Service Organizations

SOC 2 Type II · AroIQ in scope

Independently audited controls, with the platform itself inside the audit boundary. Report available under NDA.

A range, not a point

One site baseline and three scores on every event: detection, volume range, causation. One score would be easier. It would also be wrong.

Independently evaluated

METEC at Colorado State University evaluated our SCADA-based methodology (January 2026): operating states from SCADA alone were statistically distinct and correlated with measured emissions.

How to start

Start with a trial. Move to commercial.

A structured trial on history you already hold, inside one budget cycle. Your team's time: about thirty minutes every two weeks. You finish with a costed, ranked list of what to fix and the evidence behind it — then move to the AroIQ dashboard and model-as-a-service together.

How a trial works

Trial

3–6 months
  • 12–24 months of historical SCADA
  • A handful of facilities, focused scope
  • No integrations required
  • Graded at every phase gate

Commercial

you get both

AroIQ dashboard

Triage, patterns, events and causes — reviewed and actioned by your team.

Model-as-a-service

Raw outputs into your own stack: Snowflake, ops centre, work orders.

You know where.
AroIQ tells you why.

Bring 12–24 months of historical SCADA from a handful of facilities. We'll show you what AroIQ finds — events, likely causes, volumes with their range, and what each pattern costs you.

See it on your data How a trial works