Visual Intelligence Summit: Oct 22 in San Francisco Get your ticket (opens in a new tab)

Roboflow + Process Historians

Log detections, counts and pass or fail results in AVEVA PI, Canary, FactoryTalk Historian and other historians, collected over OPC UA from your edge device.

A process historian stores time-series data from plant equipment, so teams can chart trends, compare shifts and investigate events. AVEVA PI, Canary, FactoryTalk Historian and Proficy Historian all collect data from OPC UA servers through their own connectors or collectors.

Roboflow's OPC UA Server service turns your edge device into one of those servers. A Workflow's OPC UA Writer Sink writes results, such as a defect count, a reject flag or the last confidence score, to tags on the device. The historian's OPC UA collector then subscribes to them and stores them alongside your process data.

How to connect Process Historians to Roboflow

6 steps
  1. Set up an edge device

    Register a device in Deployment Manager on a network the historian's collector can reach, with port 4840 open between them. The device runs your Workflows on site.

  2. Add the OPC UA Server service

    On the device's Configuration tab, add the OPC UA Server service and create a folder and a tag for each value you want to store, such as an Int32 defect count or a Float confidence score. Turn on Require authentication if the collector should log in.

  3. Write results from the Workflow

    Add one OPC UA Writer Sink per tag. Enter the device's endpoint URL, the namespace http://opcua.roboflow.run, the folder.tag path as the Object Name with Variable Name left empty, and a value type that matches the tag.

  4. Point the historian's OPC UA collector at the device

    In your historian's OPC UA connector or collector, add a connection to the device's endpoint URL, with the user name and password if you set them.

  5. Select the tags to store

    Browse or discover the device's tags from the collector and add the ones you want as historian points.

  6. Check the data

    Trend the new points in your historian's tools. The device page in Deployment Manager shows how many OPC UA clients are connected and the latest tag values.

What teams build with Process Historians

  • Trend defect rates by shift

    Store defect and output counts as time series and compare them by shift, line or product.

  • Investigate a quality event

    Look back at what the vision system reported at the moment a problem started, next to the rest of your process data.

  • Correlate vision results with process conditions

    Compare defect rates with temperature, line speed or other tags your historian already collects.

Works with

ProductHow Roboflow connects
AVEVA PI SystemPI Connector or Adapter for OPC UA
Canary HistorianOPC UA Collector
FactoryTalk Historian SEOPC UA server points
Proficy HistorianOPC UA DA Collector
Ignition Tag HistorianIgnition OPC UA connection

Questions

Any historian that can collect data from an OPC UA server. AVEVA PI, Canary, FactoryTalk Historian SE, Proficy Historian and Ignition document this in their own guides.

No. Roboflow's OPC UA Server service runs on the edge device, and the historian's collector connects to it directly.

By default the service accepts anonymous connections with security policy None. Turn on Require authentication so collectors log in with a user name and password.

No. The OPC UA Server service and the OPC UA Writer Sink run on your own edge device, on the plant network.

The OPC UA Server service and the OPC UA Writer Sink are part of the Enterprise plan. Talk to our team to add them to your workspace.

Why are companies choosing Roboflow?

The end-to-end platform for vision AI: create datasets, train models and deploy to production.

  • Flexible deployment modes

    Move from a scalable cloud API to edge hardware like NVIDIA Jetson without managing complex dependencies.

  • Integrated workflow builder

    Combine multiple AI models, custom business logic and third-party integrations into one production-ready pipeline in a low-code editor.

  • Model monitoring and alerts

    Dashboards track confidence, inference time and request volume for your deployed models, and email alerts tell your team when confidence suddenly drops.

View all
Get started

Build on the Platform

For developers, engineers, and technical founders who want to get hands on. Try the free tier; the docs are open.

Bring it into your operation

For heads of AI, operations leaders, and enterprise teams. Bring a known problem, or work with us to find the first one worth taking on.