Furnace Monitoring AI

Watch every tap, flame, and charge, find shell hot spots, and keep people clear, on every furnace, melter, and kiln.

Molten metal pouring in a foundry with a worker in protective gear and AI object detection overlays labeling elements like "person," "tap_stream," and "shell_hot_spot."

Furnace Monitoring AI for the Toughest Melt Shops, Smelters, and Glass Plants

  • Deploy Anywhere, Run Everywhere

    Run furnace monitoring on heat-shielded visual and thermal cameras at tap holes, peepholes, doors, and charging floors, on the edge, on-prem, in your VPC, or via API.

  • One Platform, Full Adoption

    Tools every hot-end team can adopt, from furnace operators and melt shop supervisors to process, refractory, and safety engineers, no separate ML team required.

  • Secure, Compliant, and Audit-Ready

    Data stays safe with SOC 2 Type II compliance, encrypted data, and an uptime SLA, with a timestamped clip of every tap, alarm, and hot spot.

  • Tapping Stream & Slag Carryover
  • Shell & Refractory Hot Spots
  • Flame, Arc & Burner Condition
  • Batch Blanket & Foam Coverage
  • Door, Charge & Scrap Bucket Events
  • People Near Molten Metal Zones
“The vision AI solutions we developed with Roboflow have been a core component of connecting our teams and unlocking their ability to collaborate.”

Lou Stocco

Director of Manufacturing Advanced Analytics, USG

Talk to a Vision AI engineer who's shipped in metals and materials.

Bring us your toughest furnace monitoring problem and we'll map a working solution.

Ask us about:

  • Solution architecture that fits your furnaces, cameras, and level 2 systems
  • A live demo on your own furnace and melt shop footage
  • Deployment options: edge, on-prem, air-gapped, or VPC, on heat-shielded visual and thermal cameras
  • ROI modeling against breakouts, refractory costs, off-spec heats, and energy per ton

Over 80,000 organizations build with Roboflow.

  • Rivian
  • Pella
  • Chobani
  • USG Corporation
  • BNSF Railway
  • American Woodmark
Start where you are

Vision AI is transforming businesses

Customers across the board are solving complex challenges and driving meaningful impact.

  • Automotive customer

    $10 million

    Saved by automatically detecting defects on the production line

  • Logistics & freight company

    90%

    Less time spent manually tracking shipping inventory

  • Building materials supplier

    60%

    Lower customer return rate with improved product quality

Watch Every Furnace, from the Charge to the Tap

Add a real-time monitoring layer to every furnace with Roboflow vision AI for furnace monitoring. Built for electric arc and blast furnaces, aluminum and copper melting furnaces, reheat furnaces, glass and mineral wool melters, and kilns, where the process runs behind a peephole or a door at temperatures no one can stand next to, an operator can watch only one screen at a time, and a missed change shows up as a breakout, an off-spec heat, or an injury.

Tapping and pouring:

  • Watch the tap hole and tapping stream for splashing, abnormal flow, and early slag carryover.
  • Detect when each tap starts and ends, and log its duration with a clip for every heat.
  • Track each ladle under the furnace by number, alongside ladle monitoring.

Inside the furnace:

  • Classify flame shape, arc behavior, and burner condition from peephole and tuyere cameras.
  • Measure batch blanket and foam coverage on glass and mineral wool melters, and spot unmelted batch.
  • Log scrap bucket charges and door openings, and flag skull and build-up at doors and openings.

Shell, refractory, and people:

  • Map shell temperature with thermal cameras and flag hot spots that point to thin refractory.
  • Detect water, steam, and smoke where they should not be around the furnace.
  • Detect people entering molten metal zones during tapping and charging, and check their PPE.

Bring intelligence to every furnace today. Stop missed changes behind the peephole from becoming breakouts, off-spec heats, and injuries.

‍

Frequently asked questions

What is furnace monitoring with Vision AI?

Furnace monitoring with Vision AI uses heat-shielded visual and thermal cameras at tap holes, peepholes, doors, and charging floors and a trained computer vision model (opens in a new tab) to watch tapping, flames, and charging, map shell temperature for refractory hot spots, and keep people clear of molten zones. It pairs with billet tracking out of the reheat furnace and overhead crane monitoring across the melt shop.

Which furnaces does it work on?

Electric arc furnaces, blast furnace tap holes and tuyeres, basic oxygen and ladle furnaces, aluminum and copper melting and holding furnaces, reheat and heat treat furnaces, glass and mineral wool melters, and cement and lime kilns. Any furnace with a camera view of the tap, the flame, the shell, or the floor around it is a fit.

Can cameras run next to a furnace?

Yes. Furnace cameras sit in water- or air-cooled housings with purge air to keep the lens clean, and thermal cameras watch the shell from a safe distance. Roboflow runs on the video those cameras already produce, over RTSP streams.

Can it integrate with our level 2 systems, PLCs, and historian?

Yes. Roboflow sends tap events, temperatures, and alerts over OPC UA, MQTT, or REST (opens in a new tab), or writes them to a database, so they feed level 2 and MES systems, a PLC for alarms, and plant historians.

Roboflow serves as the monitoring engine, and your melting, refractory, and safety teams own the limits, alarms, and response.

‍

Stay connected

Get the Latest in Computer Vision First

Unsubscribe at any time. Review our Privacy Policy.