Evalu8Pro
Sign in Get started

Process and plant intelligence

See your plant the way your best engineer does.

Your pumps, motors, tanks and lines already produce the data. Evalu8 Pro joins it to your maintenance, production and quality records, spots drift before it turns into downtime, and suggests the fix for a person to approve. Your control system stays in charge throughout.

Book a demo How it stays safe

Works alongside your PLCs, SCADA and historian, from the readings they already keep. It never sits in the control loop.

Example figures. One pump’s page: readings from the historian, hours from the maintenance system, and a likely cause with its reasons for an engineer to confirm.

The problem

Your plant knows more than it can tell you

  • The data is in silos

    Readings sit in the historian, work orders in the maintenance system, quality results in a spreadsheet. Nobody sees them side by side, so nobody sees the pattern.

  • Alarms without reasons

    The control system says a value crossed a line. It doesn’t say why, what to check first, or that it has been creeping that way all week.

  • Know-how walks out of the door

    Your best engineer can hear a pump going wrong. When they are off shift, or retire, that instinct goes with them.

How it works

Connect, watch, diagnose, act

Evalu8 Pro sits above the systems that run your plant. It reads what they know, joins it up, and turns it into advice your team can act on.

  1. Connect

    Read what you already have

    Read-only queries on your databases, the APIs of your maintenance and production systems, and gateways or drives posting readings directly. Spreadsheets are welcome too.

    Historian · maintenance · ERP · gateways

  2. Watch

    Every series, checked as it arrives

    Gone quiet, too high or too low raises an alert by itself. Every asset gets live charts, and every run sits against the range your past runs stayed within.

    Alerts · charts · runs

  3. Diagnose

    Reasons, not just readings

    When something drifts, Evalu8 Pro weighs the evidence across sensors, running hours and maintenance history, and names the likely cause for an engineer to confirm.

    Likely causes · advice · history

  4. Act and learn

    Approve to act

    A suggested fix can be a button: approved by a person, sent to the controller, and confirmed by the readings. Every answer and outcome is kept to learn from.

    Controls · follow-ups · predictions

What you get

The instinct of your best engineer, on every shift

Drift, caught before it trips

Limits catch failures. Trends catch them early. Evalu8 Pro looks at how each reading is moving, how it compares with the asset’s other sensors, and how long it has been since the last service, and names the likely cause with its reasons.

Engineers confirm or correct each one. Their answers are kept with the readings behind them, so the advice gets sharper with every shift.

Pump P-201Flow, last 14 days

Flow, m³/h, from the historian

Raised 4 days agoSuction strainer blocking

  • The control system’s alarm has not been reached yet
  • Advice: clean the strainer at the next planned stop

Every run against the runs before it

For batch and run-based processes, each run is charted against the range your past runs stayed within, lined up from the start. You see a run going off course while there is still time to recover it, and every finished run adds to what normal looks like.

Readings follow the line a run was on, for the time it ran, so moving a sensor or swapping an asset never rewrites history.

Run R-2291Mixer torque, hours from the start

This run, N·mWhere past runs stayed

Energy and output, side by side

Meter readings add up to the energy each line used, next to what it produced. Energy per tonne, availability, stoppages and first-pass quality are each a few plain steps on the dashboard, with targets and an alert when they slip.

Anyone can open a number and see exactly how it was worked out, so the figure in the board pack and the figure on the shop floor are the same figure.

Energy per tonne 41.6 kWh/t ▼ 6.2% on September
  1. divideEnergy used ÷ Tonnes produced

Energy used

  1. totalpower from the line meters over the period, in kWh

Tonnes produced

  1. filterruns where status = finished
  2. sumtonnes

Target: under 44 kWh/tMet

Approve to act, from anywhere

When a fix is a change the equipment can take, it becomes a button: “Reduce line 2 to 80%.” Someone allowed to presses it, the request goes to the controller, and it is only marked confirmed when the readings show the change.

Ask, the built-in assistant, can propose the same fix in answer to a question, and runs it only when told to go ahead. Nothing is retried by itself, and a double press is refused.

Line 2Line controller

Set line speedNow 100% · 46 m³/h
Set speed…
  1. Asked for: line speed 80%[email protected] via Ask · 14:02 · after “ok, do it”
  2. Sent to the line controller14:02 · the controller replied “accepted”
  3. Confirmed by the readings14:05 · speed 80%, as asked

Never in the control loop

Evalu8 Pro directs. Your control system controls.

Think of it as a manager who sees the whole plant and gives good advice, but never grabs the controls. Real-time control stays exactly where it is.

  • Your PLCs and SCADA

    Control in real time

    Loops, interlocks and safety limits stay where they are. If Evalu8 Pro is offline, the plant runs on without noticing.

  • Evalu8 Pro

    Reads and advises

    Watches, compares and diagnoses across the whole site. It sends a command only when a person approves it, and the controller can still refuse.

  • Your people

    Decide

    Every press is logged with who pressed it, what was sent, what came back, and what the readings did next.

The model behind it

Your plant, in your own words

Nothing in Evalu8 Pro is written for one industry. A single description of your plant says what it is made of, how the parts connect and what each one measures, using the names your team already uses. The screens, alerts, diagnoses, numbers and API are all built from it.

New line? New sensor? New way of measuring a run? Describe it, and it appears everywhere at once.

  • Water and wastewater
  • Food and drink
  • Chemicals and coatings
  • Packaging
  • Materials handling
  • Mining and aggregates
  • Energy and utilities
  • Building services

In the model, in plain English

A line is made of assets: pumps, motors, tanks and valves.

A sensor is placed on an asset for a while; its readings stay with the asset it was on.

It reports flow in m³/h, pressure in bar and current in amps. Raise an alert above 4.5 bar, or if it goes quiet for 10 minutes.

A run starts and finishes on a line, and produces so many tonnes.

Energy per tonne is energy used divided by tonnes produced.

  • A page for every asset and runlive charts, alerts, work orders and controls
  • Monitoringwhat is quiet, out of range or drifting, and why
  • Numbers for the whole siteenergy, availability, stoppages, quality
  • An APIevery record and reading, for your own tools and models

Who it is for

For the people who run plants, and the people who build them

Operations and maintenance teams

Fewer surprises, fewer wasted visits, and numbers everyone trusts.

  • Maintenance by condition and running hours, not just the calendar
  • A likely cause and advice before anyone walks to the asset
  • Follow-ups that check a fix worked, without anyone remembering to look
  • Energy, output and stoppages for the board, from the same records as the shop floor
Book a demo

OEMs and system integrators

Monitoring and advice for every customer, without a custom build for each.

  • Describe your equipment’s API or device cloud once, as connection rules
  • Ship diagnoses for your machines that your customers’ engineers confirm
  • Devices and gateways post readings to one documented endpoint, safe to resend
  • Every customer in their own account, with their own keys
Talk to us about integrating

How we work

Start with one line. Prove it. Then the site.

We start where the downtime hurts most, connect read-only, and show results on your own data before anything else changes.

  1. Walk the plant

    We sit down with your engineers to find the assets, failures and numbers that matter most.

    Assets · failure modes · KPIs

  2. Connect, read-only

    We connect your historian, maintenance and production systems, and fit gateways or sensors where readings are missing.

    Databases · APIs · gateways

  3. Describe and tune

    We describe the plant in the model and tune alerts and diagnoses against your history, with your engineers.

    Model · limits · diagnoses

  4. Act and grow

    Add approve-to-act controls where they pay off, then more lines and sites, all in the same model.

    Controls · lines · sites

Questions

Does it replace our SCADA or historian?
No. It reads from them and adds what they don’t do: readings joined to maintenance, production and quality records, likely causes with reasons, and numbers for the whole business.
How does it connect to our equipment?
Through what you already have: read-only queries on a PostgreSQL historian or database, the JSON APIs of your maintenance and production systems, or a gateway posting readings to Evalu8 Pro’s readings API. Machine protocols come in through the gateway.
Can it change setpoints?
Only through controls you describe, only when someone allowed to presses the button, and always as a request the controller can refuse. Every press is logged, and only counts as done when the readings show it.
Is our process knowledge safe?
Your model, your diagnoses and your data belong to you, in your own account. Download the whole model at any time, and connection keys are encrypted and never shown again.
Can we use our own prediction models?
Yes. Every confirmed or corrected diagnosis is available through the API with the measurements behind it, ready to train on. A model you train elsewhere can write its predictions back as ordinary fields, where pages and dashboards use them.
What if readings arrive late, or out of order?
They slot into place. Readings can be sent in batches, resent safely after a timeout, and kept by a gateway through an outage and sent afterwards.

Put your plant’s data to work.

Tell us about one line that costs you the most downtime. We will show you what Evalu8 Pro would see in it.

Book a demo See other solutions