Factbird Connected the Shop Floor to AI Assistants With Its New Visual Intelligence and MCP Capabilities

Factbird Connected the Shop Floor to AI Assistants With Its New Visual Intelligence and MCP Capabilities

Image source: Public Domain

Factbird announced the launch of two new AI capabilities. Visual Intelligence uses cameras and AI to measure what sensors and PLCs miss. Factbird MCP lets AI assistants such as Claude, ChatGPT, and Microsoft Copilot work directly with production data in Factbird.

The losses that are hardest to explain happen around the machine. A station waits for material. A manual task runs long on the night shift. Each one costs a few minutes, and over a year they add up to real capacity.

Factbird is a System of Action, built for manufacturers who have plenty of data and struggle to act on it. It connects production data from every line, machine, and operator into one live data layer. AI built into the platform then finds the losses worth fixing and suggests the next move. The two new capabilities extend that system. Visual Intelligence brings more of the shop floor into the data layer. Factbird MCP makes that data available in the AI tools people already use.

“Machine data has always shown when a line falls behind and by how much. Visual Intelligence adds what happened around the machine, and Factbird MCP lets our users analyze it alongside the rest of their systems. That combination of data sources is what sets Factbird apart, and it gives manufacturers a big advantage,” says Søren Schønnemann, CEO at Factbird.

Visual Intelligence: a new dimension of production data

Visual Intelligence builds on Factbird’s AI product counting. Cameras film the stations and areas a manufacturer chooses, and AI on an edge device at the line turns the footage into production data. The numbers show up in Factbird next to the downtime and performance data teams already track.

On a line, that means teams can:

  • See when a station is starved or blocked
  • Measure manual cycle times and station use without a stopwatch study
  • Follow material flow to find where product stops moving
  • Count products that standard sensors can’t
  • Flag visible defects and deviations on products

Alongside the numbers, the AI highlights the events worth investigating first and keeps the thirty seconds of video around each one. Production teams still make the call on the root cause, but they start from facts where they used to have notes.

Faces in that footage are blurred on the edge device before any video leaves the line, and Factbird permissions decide who can watch it or access the data.

Getting started doesn’t take a large IT project. Factbird supplies the cameras and the edge device and trains the AI on each manufacturer’s products and stations. Manufacturers start with one station and add stations as they see results.

Factbird MCP: a direct line from production data to AI assistants

Factbird MCP is the connection that lets AI assistants work with live production data. It is built on the Model Context Protocol, an open standard. Claude, ChatGPT, and Microsoft Copilot already work with it. So will any future assistant that supports it.

A plant manager can type “Which station held Line 2 back on the day shift?” into the assistant they already have open. The assistant pulls the live numbers from Factbird and works out the answer. If the same assistant is connected to an ERP or maintenance system, it can draw on those too.

Access is read-only. Each person sees exactly what their Factbird permissions allow, and nothing more. Factbird MCP is included in every Factbird plan at no extra cost.

“Visual Intelligence turns what could only be observed into data you can track over time. With Factbird MCP, AI assistants can work with that data too. We built it on an open standard on purpose. Manufacturers should choose their AI tools freely, and their production data should work with every one of them,” says Mogens Hansen, Chief Technology & Product Officer at Factbird.