The mill assessment workbook
The exact structure we use in a ninety-minute assessment: loss ranking, instrumentation gaps and payback arithmetic. Run it yourself if you prefer.
Resources
Everything we know, ungated
Guides, methodologies, templates and benchmarks for rolling autonomy. No email walls, no drip sequences — if it is useful, it is public.
Nothing gatedWritten by the deployment teamTemplates you can actually use
exit gauge1.982 mm
flatness I-unit3.1 IU
F5 roll force18.4 MN
interstand tension9.6 MPa
chatter margin0.24 · safe
anneal energy-11.4%
pass 6/7autonomy: assisted-writetwin delta 0.004 mmaudit id 7f31c
Used by rolling and quality teams at these producers
NordStrip Steel
Kalyon Metals
Ferrata Group
Aurum Flat Rolled
Meridian Coil
Volkan Çelik
Guides
Practical guides
Long-form, specific, and written from deployments rather than desk research.
Baseline methodology template
How to define a prime-yield, energy or scrap baseline your finance team will sign and a vendor cannot argue with. Includes attribution and normalisation rules.
Instrumentation requirements by line
Required and recommended signals for hot strip, tandem cold, reversing, plate, anneal and coating lines, with sampling rates and mounting notes.
OT architecture review pack
What a security review of a mill-edge autonomy system should cover: segmentation, write guards, watchdogs, identity and audit.
Autonomy promotion gate checklist
The four gates, the evidence each requires, and who should sign. Use it even if you never buy anything from us.
Rolling autonomy glossary
AGC, TMCP, I-units, third-octave chatter, crown, wedge, edge-wave, centre-buckle — defined for the IT and data people now sitting in mill meetings.
Benchmarks
Benchmarks and models
Numbers you can sanity-check our claims against.
Prime-yield loss benchmark
Typical distribution of prime-yield loss across off-gauge, shape, surface, property and handling causes, by line type. Anonymised across design-partner assessments.
Energy intensity reference
Reheat and anneal energy per tonne by line type and grade family, with the levers that actually move it ranked by effect size.
ROI model spreadsheet
The payback arithmetic we use, with your tonnage, coil value and measured losses as inputs. No lead capture; it is a spreadsheet.
Latency budget reference
Where the milliseconds go between a camera frame and a Level-2 acknowledgement, and which of them you can actually reclaim.
Defect family taxonomy
The surface and shape defect families we classify, with severity definitions and typical root causes by process.
Grade family coverage
Which automotive, appliance, electrical and structural grade families have validated property models today, and which are in progress.
Formats
How the library is organised
Six formats, one rule: nothing behind an email form.
Guides
Long-form, practical, written from deployment experience and updated when we learn we were wrong.
Templates
Baseline definitions, gate checklists and assessment structures you can use without us.
References
Instrumentation, taxonomy, glossary and latency budgets — the things engineers look up.
Benchmarks
Anonymised, aggregated data from assessments, with methodology published alongside.
Models
Spreadsheets and notebooks for ROI, payback and loss attribution.
Webinars
[PLACEHOLDER] Recorded technical sessions with design-partner engineers, published without registration.
Why ungated
Why none of this is behind a form
Two reasons, one of them self-interested.
Gating a PDF buys you an email address and costs you the trust of the exact person you need most — the rolling engineer who will decide whether your system is credible. In heavy industry that trade is obviously bad, so we do not make it.
It also imposes useful discipline. If a document has to survive being read by a sceptical process engineer with no salesperson attached, it has to be genuinely correct. Everything in this library is written to that standard, and the things that were not have been removed.
If you find an error, tell us. We correct in place with a dated note, and we will say so publicly.
Using them
How most teams use this library
In roughly this order, over a quarter.
- STEP 01
Run your own assessment
Use the workbook to rank your own losses before talking to any vendor, including us.
- STEP 02
Write the baseline
Use the template with your finance team, before anyone installs anything.
- STEP 03
Review the architecture
Give the OT pack to your security team early; it is the longest pole in most evaluations.
- STEP 04
Set the gates
Agree the promotion gates before shadow mode starts, so nobody negotiates them under pressure later.
Library
The library, in numbers
18Published items
0Behind an email form
6Formats, from guides to spreadsheets
2New items per month, roughly
Feedback
What readers tell us
“We ran Steelira in shadow for eight weeks on the tandem mill. It called eleven of the twelve off-gauge tail events before our AGC reacted. That was the moment the argument ended.”
“Our best roller retires in two years. Steelira is the first system that captured how he sets up a hard automotive grade — and then explained why, with the coil history to back it.”
“Prime yield moved 1.7 points on the galvanizing line and reheat gas fell double digits. Nothing else we bought this decade did both.”
Context
The systems these references assume
Most references are written against these stacks because that is what our design partners run.
Level-2 & control
- Primetals Level-2
- SMS X-Pact
- Danieli Automation
- TMEIC mill drives
- ABB Ability
- Siemens SIMATIC / PCS 7
Gauge, shape & surface
- IMS X-ray gauges
- Vollmer thickness
- Shape-meter rolls
- ISRA Vision surface
- Parsytec inspection
- Thermal line scanners
Plant systems
- Level-3 MES / QMS
- SAP PP-PI
- OSIsoft PI / AVEVA
- OPC UA & Classic
- Historians (IP.21, Canary)
- CMMS / EAM
Security
OT-safe by construction
Steelira writes to mill equipment, so security and safety are product requirements, not paperwork.
SOC 2 Type IIISO 27001IEC 62443 alignedGDPRSSO / SAMLData residency
- Segmented OT pathEdge runtime sits in a Purdue Level-2/3 DMZ with signed, rate-limited write-back and hard setpoint clamps.
- Deterministic fallbacksEvery control loop degrades to the mill's existing Level-2 setup on watchdog timeout — the mill never stops because Steelira does.
- Grade-IP isolationRecipes, grade specs and coil histories are tenant-scoped, encrypted at rest with per-tenant keys, and never used to train another producer's models.
- Assurance-grade auditEvery perception, recommendation, approval and actuation is written to an immutable, exportable log keyed to coil ID.
Get the monthly mill brief
One email a month with new library items, what we published and one thing we got wrong. Unsubscribe in one click.