Extruon

Physical AI for aluminium & polymer extrusion

The extrusion plant,running itself.

Extruon is the autonomous operations layer for extrusion manufacturing — autonomous billet heating and press control, AI profile-dimension and surface-defect vision, die-flow optimisation, quench, stretch and aging control, robotic handling, and a press-and-die twin that hits target profile and yield before the push.

Deployed at the plant edge · Works with your press PLC, furnace, gauge, quench, saw, aging oven and MES.

Built alongside extruders serving construction, EV, solar and industrial markets

NORDHAUS ALUVantera ProfilesSolFrame SystemsKestrel ExtrusionsRhein LeichtbauAurora Billet Co.

Design-partner names are illustrative pending public reference approval. [PLACEHOLDER]

The problem

Extrusion is stilla black art.

A preheated billet is pushed through a hardened steel die at hundreds of degrees, where metal flows and welds around the die ports to form a precise cross-section. Billet taper, ram speed, container and die temperature, and exit temperature all interact nonlinearly.

  • Push a touch too hot or too fast and the profile tears, blisters or picks up die lines.
  • Too cold or too slow and it comes out under-strength or off dimension.
  • The die itself behaves unpredictably, so plants still endure slow, expensive die trials — scrapping metal and press time to correct twist, bow and flow imbalance.
  • After the press, quench rate and stretch-straightening set temper and flatness, and aging ovens develop T5 and T6 properties. Each is another chance to miss spec.
  • Press recovery, butt and offcut scrap, and oven energy dominate cost under thin margins and decarbonisation pressure.

What plant managers run blind on today

  • Real-time dimensions
    Spot gauging between pushes, not continuous truth.
  • Surface quality
    Die lines and pickup found at packing, not at the die.
  • Twist and bow
    Corrected at the stretcher, if it can be corrected at all.
  • Temper
    Confirmed days later in the lab, after the oven ran.
  • Recovery and energy
    Reconciled monthly, long after the metal is gone.

The push, instrumented

Watch one billetbecome a profile.

Scroll to follow the push: the ram advances, the billet upsets and flows through the die, and the profile emerges on the runout — with Extruon sensing exit temperature, wall thickness and surface the whole way.

Extruon closing the loop on an extrusion press A heated billet is pushed by the ram through the container and die; the finished profile emerges on the runout while Extruon senses dimensions, surface quality and exit temperature and adjusts ram speed and billet taper in real time. RAMBILLET · CONTAINERDIEPROFILE · RUNOUT EXIT TEMP508 °C WALL±0.06 mm RAM SPEED6.4 mm/s RECOVERY87.9 %

Illustrative schematic of an instrumented press: ram, container, die and runout with press-exit sensing. Values shown are representative of a 6063 architectural profile.

How Extruon works

Perceive. Plan. Push.Prove. Learn.

One closed loop runs at the plant edge on every billet. Nothing is advisory-only unless you want it to be — and nothing acts outside an envelope your engineers signed.

01Perceive

Sense the billet, the profile, the temper

Fused line-scan and RGB vision, thermal imaging, laser gauging and press PLC telemetry describe the push as it happens — dimensions, wall thickness, straightness, surface, exit temperature.

02Plan

Set the taper, the speed, the die, the quench

The agents plan billet-heating taper, ram speed, pressure and temperature, die-flow correction, quench rate, stretch, cut length and aging recipe as one coupled decision, not five isolated setpoints.

03Act

Run with adaptive control at the press

Approved moves write back into the press PLC, the furnace and the quench inside bounded action envelopes, with rollback and alarm interlocks wired to the same fail-safe stop your line already trusts.

04Prove

Predict defects and recovery before they cost metal

The loop predicts off-spec dimensions, surface defects, twist and temper misses, and flags die-trial risk early enough to change the push instead of scrapping it.

05Learn

Every correction trains the plant

Engineer approvals and corrections are logged to an immutable, assurance-grade audit trail and fed back into training — so the site's craft compounds instead of retiring.

Seven agents, one orchestrator

The plant runs as a crewof specialised agents.

Each agent owns a physical stage of the line, senses it directly, and acts inside an approved envelope. A plant orchestrator arbitrates between them so nobody optimises recovery at the expense of temper.

Heat-and-press agent

Controls billet temperature taper, ram speed, breakthrough pressure, container and die temperature, and exit temperature to keep metal flow balanced and the profile on dimension.

Profile-and-defect agent

Senses and predicts cross-section dimensions, wall thickness, straightness and surface defects — die lines, pickup, scoring, blisters — from fused vision and laser gauge.

Die-and-flow agent

Optimises die temperature, correction and flow balance to cut die trials and the twist and bow that send profiles back to the stretcher.

Quench-and-stretch agent

Controls press-quench cooling rate and stretch-straightening so temper and flatness land inside spec on the first pass.

Cut-and-age agent

Controls cut-to-length, batching and aging ovens to reach T5 and T6 mechanical properties without over-soaking the oven.

Robot-and-handling agent

Drives robotic pulling, stacking, racking and packing of long, hot, delicate profiles — the handling nobody wants to staff at 3am.

Yield-and-throughput agent

Optimises press recovery, scrap, butt and offcut loss, and energy per tonne across the whole line rather than one station at a time.

Extrusion-knowledge agent

Answers metallurgy and die-design questions with citations into your die books, profile drawings, press recipes and alloy specs.

Press-and-die digital twin

Simulates billet heating, metal flow through the die, quench and profile properties — hitting the target profile and yield before the push, not after the scrap.

Measured on the press

The numbers a plant directoractually gets paid on.

Every Extruon engagement starts with a baseline and ends with an audited delta. These are the target bands we underwrite in a paid pilot.

+4.2 pts

Press recovery uplift, saleable vs charged metal

−38%

Surface-defect and dimensional rejects

−61%

Die-trial pushes before a die is signed off

−12%

kWh per tonne across heating, quench and aging

Target outcome bands modelled from design-partner baselines. [ASPIRATIONAL — to be replaced with audited pilot results.]

The console

One screen for the press,the profile and the money.

Press and process engineers approve, correct and override from a review console. Every correction trains the models — which is how the site's craft stops walking out the door at retirement.

87.9%Press recovery (shift)
508 °CExit temperature
±0.06 mmWall deviation
0.9%Reject rate
  • Ram speed −4% · blister risk at breakthrough
  • Billet taper adjusted +12 °C front-to-back
  • Die 44-118 flow imbalance flagged · trial avoided
  • Quench rate held for T6 on rack 19
  • Aging batch 7 consolidated · 340 kWh saved

Representative console view. Data shown is illustrative of a single shift on one press.

Why nobody else closes the loop

Gauges measure.Simulators guess.Extruon acts.

Incumbents sell a press-control system, or an in-line gauge, or a furnace controller, or a flow simulator, or an aging controller. None unify them into one plant-edge platform that acts on the process and compounds the data.

System of action

Closed perceive–decide–act, at the edge

Extruon does not stop at a setpoint or a measurement. It writes back into heating, press, quench and aging inside approved envelopes, then proves the result on the same push.

Data moat

The craft, encoded

Every supervised correction from a press operator or die engineer becomes training data. The scarce craft compounds instead of retiring.

100%

of approvals and corrections logged and reused

Fleet learning

Rare faults travel

A die-line signature learned on one plant's 6063 architectural family transfers — privacy and IP preserving — to every extruder running similar alloys and profiles.

Before the push

Hit profile and yield in simulation first

The press-and-die twin simulates billet heating, metal flow, quench and profile properties, so a new die enters the press with corrections already applied rather than earning them through trials.

Who it is for

Three people have to say yes.We answer all three.

The plant director is the economic buyer. Press and process engineers and quality and metallurgy managers are the champions. IT, OT, quality and security approve.

Recovery, throughput, scrap, energy — measured

You own press recovery, throughput, scrap and energy, and you cannot hire your way out of a press-operator shortage. Extruon lands on one press with a hard success metric and an audited baseline.

  • Hard ROI proven in a paid pilot before you standardise
  • Recovery, scrap, butt loss and kWh per tonne on one dashboard
  • Expansion press by press, module by module, site by site
  • No dependency on hiring scarce veteran operators

Assurance-grade by default

Trust is the gateto autonomy.

Extruon never asks for control it has not earned. Shadow mode proves accuracy against your operators and gauges; advisory mode proves ROI; bounded autonomy only follows once the gates are met.

Isolation

Per-tenant everything

Data, models, retrieval indexes and memory are scoped per tenant and per entity. On-prem deployment is available for sensitive producers.

Die IP

Your die geometry stays yours

Die designs, customer drawings and recipes never leave your boundary. Fleet learning shares defect signatures and process priors, never geometry.

Audit

Immutable action log

Every perception, recommendation, approval and setpoint write is recorded with its evidence, ready for quality and customer audits.

Controls

SSO, RBAC, encryption

SSO and role-based access, encryption in transit and at rest, SOC 2 programme in progress, and fail-safe interlocks on every write path.

From the floor

What changesin the first quarter.

The first thing that changed was the argument. We stopped debating whether the die or the billet caused the twist, because the system showed us both on the same push.

Process engineering leadArchitectural profile extruder, 2,750 t press

We used to burn three pushes proving out a new die. Getting that down is press time we sell instead of scrap we melt.

Die engineering managerMulti-site aluminium group

Recovery is the number my board reads. An audited baseline and an audited delta was the only proposal that survived the capital committee.

Plant directorSolar-frame and EV enclosure extruder

Design-partner quotes are illustrative of the target outcome. [PLACEHOLDER — to be replaced with named references at GA.]

The finale

Every push,perfectly profiled.

Start with one press and one workflow. We baseline your recovery, scrap and energy, run in shadow, then earn every increment of autonomy against a number you choose.

Land on one press. Expand press by press, module by module, site by site.