An energy saving that cannot be defended is only a claim. When a steel plant reports that a project cut energy use, finance, auditors and sometimes lenders ask the same thing: compared with what? The answer is the baseline, a model of how much energy the plant would have used without the project, under the same production and conditions. Building it carefully is the difference between a verified saving and an argument. Teams preparing a measurement and verification project can see how iFactory AI builds and tests an energy baseline on live plant data before the first savings report is due.
Energy Baseline Establishment for Steel Plant M&V Projects
Build a production-normalised baseline that holds up in review, so the saving you report is the saving you can prove.
The Questions a Baseline Must Survive
Reviewers rarely argue with the maths. They question the assumptions behind it. Expect these four.
Was the baseline period representative of normal operation?
Did production, mix or weather change enough to explain the saving?
Is the model statistically sound, or fitted to flatter the result?
Were other changes made at the same time that could share the credit?
A baseline built with clear answers to all four is far easier to defend than one patched after the challenge arrives.
Choosing an IPMVP Option
The International Performance Measurement and Verification Protocol, known as IPMVP, defines four options. The right one depends on the project's boundary and how easily its energy can be isolated.
| Option | What Is Measured | Suits |
|---|---|---|
| A: Retrofit isolation, key parameter | A key parameter is measured and the rest is estimated | Drives or lighting where operating hours are known |
| B: Retrofit isolation, all parameters | All energy for the affected system is measured | A furnace, compressor system or pump set |
| C: Whole facility | Plant-level meter data is modelled against production | Projects with broad effects across the plant |
| D: Calibrated simulation | A simulation model is calibrated to measured data | New builds, or where no baseline data exists |
Build a Baseline You Can Defend
Book a 30-minute session and iFactory AI will walk through how your baseline period, model and adjustments would be set up for a verifiable savings claim.
Picking the Baseline Window
The baseline period must cover the full range of normal operation. A short, unusual window gives a model that fails as soon as conditions move.
Include high and low output periods, so the model is not forced to guess beyond what it has seen.
Twelve months is common practice, which captures seasonal effects on cooling and utilities.
Avoid periods with major outages or unusual faults unless they are modelled.
Check calibration and gaps before fitting, since poor data cannot be fixed by good statistics.
Five Gates Before You Lock the Baseline
Treat model building as a series of gates. A baseline passes each one or goes back.
Thresholds for the statistical measures differ between protocols and contracts, so agree the required values with the verifier before fitting the model.
Routine and Non-Routine Adjustments
Conditions change after the baseline is set. Adjustments keep the comparison fair, and each type is handled differently.
A Worked Example
The numbers below are illustrative. The baseline model says energy equals a fixed load plus a rate per tonne, and the reporting period is compared with it.
Without the production adjustment, a busier period would hide the saving and a quiet one would exaggerate it. The model keeps the comparison fair either way.
Where iFactory AI Fits
iFactory AI links meter, production and cost data, so baselines are built, tested and reported on one consistent basis.
Production-normalised models
Baselines are fitted against tonnes and mix, with fit statistics shown for review.
Saving tracked every period
Adjusted baseline and actual energy are compared automatically, so savings are tracked, not rebuilt each quarter.
Adjustment log
Non-routine events are recorded with their effect, giving verifiers a clear audit trail.
Ask in plain language
Ask what the verified saving was last quarter and receive the figure with its basis.
iFactory AI arrives pre-configured on an NVIDIA server that ships racked and ready with software pre-loaded. Scope covers cabling, network, ERP and MES integration, team training and 24x7 remote monitoring.
Frequently Asked Questions
What is an energy baseline in M&V?
It is a model of how much energy the plant would have used without the project, under the same production and conditions. Savings are the gap between that adjusted baseline and measured energy after the project. You can watch a baseline fitted on sample plant data in a live session.
How long should the baseline period be?
Twelve months is common, because it captures a full seasonal cycle and a wide range of output. Shorter periods can work for isolated equipment if they cover normal operating conditions. Confirm the requirement with your verifier or contract. To plan yours, request a baseline planning session with the iFactory AI team, or ask support what data to collect.
Does weather matter in a steel plant baseline?
It can, mainly for cooling, compressed air and other utilities. Heavy process energy is driven mostly by production and product mix. Include weather only where there is a physical reason and a statistical gain. A short product tour of driver selection shows how this is tested.
What if the plant changes during the project?
Record the change as a non-routine adjustment, estimate its effect by an agreed method and keep the documentation. Large changes may mean re-baselining. Raising the issue with the verifier early avoids disputes later. See how adjustments are logged and applied in a guided walkthrough.
How does a baseline support finance and lenders?
A documented baseline turns energy savings into a figure finance can book and lenders can test, because assumptions and statistics are visible. It also protects the plant if results are questioned later. Schedule a walkthrough of savings reporting to see the view finance would receive.
Turn Energy Savings Into Verified Savings
iFactory AI builds production-normalised baselines and tracks verified savings every period. Book a walkthrough to see it built around your own plant.
