Contact us

Case study · Optimization (Pilot Completed, Production Implementation in progress)

From next-morning KPIs to same-day predictions

A virtual sensor and advisory optimizer for the Pak Arab Fertilizers (PFL) ammonia plant. It predicts daily Production and Energy Index from live process signals. A capability field-validated in a pilot, and now being implemented at production grade. It recommends safe setpoint moves that cut energy while holding production.

  • Industrial AI
  • Process optimization
  • Virtual sensor
  • Energy efficiency
3D illustration of a process plant with columns, tanks and piping linked to a central glowing data cube and three chart panels

At a glance

Energy Index Optimizer, in four facts

Product
Energy Index Optimizer
Industry
Chemicals and fertilizer
Built for
The Pak Arab Fertilizers (PFL) ammonia plant
Outcome
Daily production predicted with 99.4% accuracy (proven in pilot)
Predict the day. Recommend the move. Keep the engineer incharge.

Energy Index Optimizer · The design goal

The case file

From challenge to outcome.

Step 01

The challenge

Production and Energy Index were known only the next morning, with no view of whether the operating point was efficient and no quantified basis for trading production against gas burned.

3D illustration of plant columns and pipework around a glowing data cube with chart panels

Step 02

Our approach

Predict, then optimize. 127 DCS tags, KPI and lab reports and weather feed one daily pipeline; each day becomes about 890 features; a Ridge and LightGBM ensemble predicts production, and Energy Index is computed from an energy balance.

Step 03

The solution

An advisory layer, built and pending field validation, that scores thousands of setpoint combinations in about 10 seconds. Moves are capped at ±3% per day, nothing writes to the DCS, and the engineer decides.

Step 04

From next-morning KPIs to an in-day virtual sensor results, field-validated on 49 unseen days

  • 0.59%Average error predicting daily production.
  • 2.1% Average error predicting Energy Index.
  • 0.93 R²Variance in daily production explained, on unseen data.

FAQ

Questions about Process Optimizer.

The short answers, in the order teams usually ask them.

What is the Energy Index Optimizer?

A virtual sensor that predicts the plant’s daily Production and Energy Index during the day, plus an optimizer that recommends safe setpoint moves. It is currently a pilot project

Does it control the plant?

No. It’s advisory only. Nothing writes to the DCS, and an engineer approves every move.

What data does it use?

127 DCS tags from the historian, daily KPI and lab reports, and ambient weather.

How accurate is it?

In Phase 1 field validation, 0.59% average error on daily production and 2.1% on Energy Index, tested on 49 consecutive unseen days.

What if a recommendation is wrong?

Moves are capped at ±3% per day and stay inside proven operating ranges, so the downside of a bad day is an ignored suggestion.

Is it live?

It’s a pilot project. Phase 1 prediction is field-validated; Phase 2 recommendations are built and pending field validation.

Pilot the Process Optimizer
on one process unit.

Bring the challenge. Bring the ambition.
We’ll show you what it looks like on your own data.