<script type="application/ld+json"> { "@context": "https://schema.org", "@type": "Service", "name": "AI solutions for foundries", "serviceType": "Artificial intelligence integration for foundries", "provider": { "@type": "Organization", "name": "StratIA", "legalName": "STRATIA CONSEIL SAS", "url": "https://strat-ia.fr", "address": { "@type": "PostalAddress", "streetAddress": "8 place Roger Salengro", "postalCode": "31000", "addressLocality": "Toulouse", "addressCountry": "FR" } }, "areaServed": { "@type": "Country", "name": "France" }, "description": "StratIA builds tailor-made AI solutions for foundries: part quality control, predictive maintenance of furnaces and stabilisation of casting processes.", "url": "https://strat-ia.fr/en/industrie/fonderie" } </script> <script type="application/ld+json"> { "@context": "https://schema.org", "@type": "BreadcrumbList", "itemListElement": [ { "@type": "ListItem", "position": 1, "name": "Home", "item": "https://strat-ia.fr/en" }, { "@type": "ListItem", "position": 2, "name": "Industry", "item": "https://strat-ia.fr/en/industrie" }, { "@type": "ListItem", "position": 3, "name": "Foundry", "item": "https://strat-ia.fr/en/industrie/fonderie" } ] } </script> <script type="application/ld+json"> { "@context": "https://schema.org", "@type": "FAQPage", "mainEntity": [ { "@type": "Question", "name": "How does AI reduce the scrap rate in a foundry?", "acceptedAnswer": { "@type": "Answer", "text": "AI inspects every part by vision (camera or X-ray) right out of the mould and identifies porosity, shrinkage and surface defects before finishing. The defect is isolated early and pouring parameters are adjusted before a whole batch drifts, which lowers scrap measurably." } }, { "@type": "Question", "name": "Can internal defects such as porosity be inspected?", "acceptedAnswer": { "@type": "Answer", "text": "Yes. For internal defects invisible on the surface, AI relies on radiography or tomography: the model is trained to spot internal porosity and shrinkage on the images, with greater consistency than manual sampling inspection." } }, { "@type": "Question", "name": "Does AI adapt to small batches and many references?", "acceptedAnswer": { "@type": "Answer", "text": "Yes. We build models that learn quickly from few examples and retrain when a new reference arrives. This is exactly the challenge in foundries, where batches are often short and varied." } }, { "@type": "Question", "name": "How does AI help with furnace maintenance?", "acceptedAnswer": { "@type": "Answer", "text": "Furnace and equipment signals (temperature, consumption, vibration) are analysed continuously. AI detects drift before failure and triggers an early alert, avoiding unplanned stops on a stage as critical as melting." } }, { "@type": "Question", "name": "How long does it take to deploy an AI solution in a foundry?", "acceptedAnswer": { "@type": "Answer", "text": "We work in stages: a short audit to scope the most profitable case, then production in 2 to 3 months depending on scope. You keep control of your data, hosted in France or on site." } } ] } </script>
INDUSTRY · FOUNDRY

AI for the foundry industry

2 to 3 months from diagnostic to deployment

In a foundry, a defect caught at finishing is value already lost. We build tailor-made AI solutions that inspect your parts right out of the mould, anticipate furnace drift and stabilise your casting processes.

Sector challenges in 2026

Foundries face heavy constraints: costly scrap, energy, rare know-how and processes that are hard to stabilise. AI acts on each of them, without changing your trades.

Defects caught too late

Porosity, shrinkage and surface defects often appear only at finishing, when the value is already committed. AI vision spots them right out of the mould and triggers correction before a whole batch turns to scrap.

Energy-intensive melting

Melting concentrates most of your energy bill. AI optimises furnace parameters at constant quality and flags abnormal overconsumption in real time.

Know-how becoming scarce

Adjusting a melt or reading a defect relies on the experience of a few people. AI captures these decisions in models and assistants, so the know-how stays available as the team changes.

Processes hard to stabilise

Pouring temperature, alloy composition, sand quality: many parameters make quality vary. AI tracks them continuously and suggests the settings that stabilise the conformity rate.

Image de Yoan Hibert, fondateur de StratIAImage de Baptiste Wieczorek, fondateur de StratIA
Book a call

Roles in the sector

From melting to finishing, every role in the foundry can rely on AI to secure quality, cut scrap and preserve rare know-how.

Melting operator

Monitors melting and pouring. AI tracks temperature and composition in real time to stabilise each melt and limit out-of-tolerance pours.

Moulder

Prepares and assembles moulds. AI vision checks mould condition and cavity quality before pouring, to prevent recurring surface defects.

Core maker

Produces cores. AI detects core cracks and defects before assembly, where a breakage often goes unnoticed until shakeout.

Quality manager

Runs part inspection. AI spots porosity, shrinkage and surface defects right out of the mould, with full traceability per batch.

Methods engineer

Industrialises new parts. AI cross-references pouring parameters and quality results to lock in the settings of a new reference faster.

Maintenance technician

Maintains furnaces and equipment. AI analyses equipment signals to anticipate furnace drift before a production stop.

Get your own use cases for free

Why StratIA

Sovereignty

Your data stays in France or on site. We rely on sovereign building blocks (OVH, Scaleway, Mistral) or an on-premise deployment, with no dependency on a foreign cloud.

French squad

A dedicated French team, from scoping to production. People who understand your industrial context and stay available after deployment.

Framed ROI

Return on investment is framed from the scoping stage. We prioritise the most profitable use case and set measurable targets before writing a single line of code.

Ils nous font confiance
Logo Paymetrust, client de StratIA
Logo Union des Marchands de Biens, client de StratIA
Logo Accessite, client de StratIA
Logo Bastide Manutention, client de StratIA
Logo Hello Prépa, client de StratIA
Logo Comité Olympique, client de StratIA
Logo Paymetrust, client de StratIA
Logo Union des Marchands de Biens, client de StratIA
Logo Accessite, client de StratIA
Logo Bastide Manutention, client de StratIA
Logo Hello Prépa, client de StratIA
Logo Comité Olympique, client de StratIA
Logo Paymetrust, client de StratIA
Logo Union des Marchands de Biens, client de StratIA
Logo Accessite, client de StratIA
Logo Bastide Manutention, client de StratIA
Logo Hello Prépa, client de StratIA
Logo Comité Olympique, client de StratIA

Expert insights

Foundries hold remarkable know-how, yet too much value is still lost to scrap because defects are caught too late. AI does not replace the founder: it puts a camera and a model on the line to spot porosity or shrinkage right out of the mould, and closes the correction loop before finishing. That is where, within a few months, the return on investment is decided.

Yoan Hibert
Yoan Hibert Co-founder & Managing Director Profile
Baptiste Wieczorek
Baptiste Wieczorek Co-founder & CTO Profile
Image de Yoan Hibert, fondateur de StratIAImage de Baptiste Wieczorek, fondateur de StratIA
Réserver un échange

Our method: from entry point to continuous use

based on your maturity level...

01

Formations IA

Des équipes prêtes pour l'IA

1 à 2 jours

02

Audit Flash IA

Vos cas d'usage priorisés

1 jour

03

Quick Win IA

La preuve par un premier gain

2 à 6 semaines

04

Développement sur mesure

Du brief au produit en production

2 semaines à 3 mois

01

Formation IA

Pour qui : vos équipes doivent d'abord comprendre ce que l'IA change dans leur métier, avant tout projet ou en parallèle.

En 1 à 2 jours, sur site ou à distance, on forme vos équipes de direction, de production ou vos fonctions support : fondamentaux de l'IA, usages concrets observés dans votre secteur, prise en main d'outils. Aucun prérequis technique. Vos équipes repartent avec des réflexes applicables dès la semaine suivante.

Tarif : certifié Qualiopi, finançable par votre OPCO, souvent intégralement.

J'ai un projet de formation

02

Audit Flash IA

Pour qui : vous savez que l'IA est un sujet, vous ne savez pas par où commencer, ou vous avez une idée précise mais personne pour l'exécuter proprement.

En 1 jour, nos experts analysent votre organisation et vos processus pour identifier les cas d'usage IA les plus pertinents et les plus rapides à déployer. Vous repartez avec une roadmap priorisée et des fiches opérationnelles par cas d'usage.

Tarif : 2 000 € HT (déductible si vous lancez un projet avec nous).

Voir plus

03

Quick Win IA

Pour qui : vous avez un irritant précis et vous voulez une preuve avant d'investir plus. Lecture de documents, chiffrage, contrôle qualité, tri, reporting.

En 2 à 6 semaines, on livre une solution en production sur un périmètre resserré, branchée sur vos outils existants (ERP, MES, GMAO). Vous mesurez le gain réel, heures gagnées, erreurs évitées, marge protégée, avant de décider de la suite.

Tarif : à partir de 6 000 € HT (l'Audit Flash déjà réalisé est déductible).

Voir plus

04

Développement sur mesure

Pour qui : vous savez déjà ce que vous voulez. Cas d'usage identifié, périmètre clair, ROI cadré. Vous avez besoin d'une équipe technique sérieuse pour passer de l'idée au produit en production.

En 2 semaines à 3 mois selon le périmètre, notre équipe française conçoit et déploie votre solution : agent IA, plateforme métier, système de vision, RAG, automatisation. Cycles courts, démos hebdomadaires, validation utilisateurs continue. Mise en production directe.

Tarif : entre 15 000 € et 80 000 € HT selon la complexité, devis ferme après cadrage.

Discuter de mon projet

One client case

Client case Aluminium foundry

Detecting porosity before machining on cast parts

An aluminium foundry producing safety parts in medium batches was catching porosity after machining, too late. We deployed vision inspection at the moulding line output, coupled with a model trained on its recurring defects.


−21 %scrap rate
96 %defects caught before machining
8 monthsreturn on investment

Estimated savings between €90,000 and €130,000 per year on material and avoided machining hours.

Discuss a similar project

Let's discuss your AI project

Image de Yoan Hibert, fondateur de StratIAImage de Baptiste Wieczorek, fondateur de StratIA
Book a call

To go further

Industry

Our AI approach across the whole industrial sector.

AI Flash Audit

Identify your priority use cases in 1 day.

AI Foundations Audit

Build your AI strategy in depth.

Tailor-made development

From scoping to the deployment of your solution.

FAQ