AIPredictive analytics, now inside PI Vision

Leverage your PI System withArtificial Intelligence

Convert process data into actionable insights.

SypinIQ brings artificial intelligence, predictive analytics and real-time excursion capture to the AVEVA PI System, so operators, engineers and managers act on insights instead of raw tags. All of it runs on your own servers.

Conversational AI Predictive analytics Excursion capture Timestamp comparison MCP for AI agents
PI Vision · Unit 21 · Reactor overviewLIVE
Reactor outlet87.4°CH in ~72 min
Header pressure8.42bar▲ within band
Feed flow412m³/h▲ 2.1% vs yday
H crossings 8h132 min above H
SOL · 21TI1103.PV 8 h + 2 h forecast
RecordedForecastH / LHH / LL
PI DataTable recorded
TimeTI1103 °CPI2101 bar
06:12:0487.418.43
06:10:3187.028.40
06:08:5586.778.44
06:07:1286.318.41
06:05:4086.058.39
Events 42Time-weighted avg
SypinIQ assistantConnected to PI
SypinIQ-Pro
Which tags crossed their H limit last shift?
2 tags. 21TI1103.PV crossed H once and stayed above for 32 min. 21PI2101.PV crossed once for 4 min while P-101 was running.
Ask about any tag…
Forecast · 21TI1103.PV

Reaches H (95 °C) in ~72 min at the current rate.

Illustrative data on a sample display. Tag names follow a typical plant convention.

Works with the systems you already run
AVEVA PI SystemPI Vision 2023PI Web APIPI AFMicrosoft 365 CopilotMicrosoft TeamsMicrosoft Entra IDAzure OpenAIOpenAIAnthropic ClaudeModel Context ProtocolWindows ServerAVEVA PI SystemPI Vision 2023PI Web APIPI AFMicrosoft 365 CopilotMicrosoft TeamsMicrosoft Entra IDAzure OpenAIOpenAIAnthropic ClaudeModel Context ProtocolWindows Server
Artificial intelligence on PI

Convert process data into actionable insights.

Your PI System already records every value. SypinIQ adds the AI layer that analyses them and tells people what to do next.

01 · Process data

Real-time PI tags

Live and archived values straight from PI Web API, read-only.

21TI1103.PV = 87.4 °C
02 · AI analytics

Predictive models

Excursion capture, regression forecasts and natural-language analysis.

Reaches H (95 °C) in ~72 min
03 · Actionable insight

Act before the alarm

The right answer, on the display or in Teams, before the limit is crossed.

Likely driver: feed flow +6% since 05:32
A typical request today

One data question.
Four hours.

At most plants a handful of PI experts answer everyone else's data questions. Every request waits in their queue, while the meeting that needed the number goes ahead without it.

Without SypinIQ4 h 07 min
With SypinIQ~2 s
  1. A process engineer needs yesterday's average reactor outlet temperature for the 10:00 operations meeting.

  2. She messages the PI team. The PI expert already has six requests waiting.

  3. The export starts. The first file comes out in the wrong time zone, so it has to be run again.

  4. An 80,000-row CSV lands in Excel and gets cleaned and pivoted by hand.

  5. The answer is ready. The meeting took its decision three hours earlier, without the data.

With SypinIQ: at 09:02 she asks Copilot “What was the average outlet temperature on 21TI1103 yesterday?” and has the number, a chart and an Excel file before 09:03.

See it live

Six products, running live.

Every screen below is a working demo. Values change, excursions are captured and forecasts redraw as you watch.

Ask · Conversational AI in Copilot

Conversational AI for PI, inside Microsoft Copilot.

An AI agent for Microsoft 365 Copilot. Engineers and managers ask about any PI tag in plain language from Teams, Outlook or the Copilot app, and get values, statistics, trends and Excel files back in seconds.

  • No extra AI subscription. It runs on the Microsoft 365 Copilot licenses you already have.
  • Natural-language analytics on live and historical data: time-weighted averages, min, max, totals and counts over any time range.
  • Inline charts and Excel exports, the shift report you would otherwise build by hand.
Microsoft TeamsCopilot · SypinIQ-XLIVE
@SypinIQ-X average reactor outlet temperature per shift yesterday
SypinIQ-X
ShiftAvg °CMax °CAbove H
Night86.992.40 min
Day89.796.118 min
Evening88.394.80 min
The day shift ran hottest and spent 18 min above H.
Message Copilot
Answered in1.8 s
Microsoft 365 CopilotNo extra AI licence

Animated demo · illustrative data

How it works

From PI tag to plain-language answer.

SypinIQ-Plus sits between PI Web API and the places your people already work. It reads, it never writes, and it logs every question.

PI TAGSWHERE PEOPLE ASK21PI2101.PVHeader pressure8.42 bar21TI1103.PVReactor outlet91.6 °C21FI1001.PVFeed flow412 m³/h21LI1204.PVDrum level63.1 %21XI0105.PVP-101 statusRunningSypinIQ-PlusPI Web API · read-onlyEntra ID · audit logMicrosoft 365 CopilotSypinIQ-X in Teams and OutlookPI VisionSypinIQ-Pro and the symbolsYour AI agentsClaude, Copilot Studio via MCP
  1. Install

    One Windows installer sets up the service, admin console and PI Vision files.

  2. Connect

    Point it at PI Web API with a read-only account. PI Data Servers are found automatically.

  3. License

    Send us the machine signature file and apply the key. It validates offline.

  4. Ask

    Drop symbols on displays, open the assistant, or publish the Copilot agent.

Built for OT networks

Your PI data stays where it is.

Designed by a PI administrator for plants behind firewalls, data diodes and strict IT policies.

Read the security details
  • Runs on your own server

    A Windows service inside your network. No SypinIQ cloud.

  • Read-only to PI

    A read-only PI Web API account over HTTPS. Never touches the DCS.

  • Entra ID sign-in, every query audited

    Corporate accounts, optional group restriction, full audit log.

  • Offline licensing

    Signed, machine-bound keys. Works on isolated networks.

See it on your own PI data.

A 30-minute call. We connect to a test PI Web API, ask real questions of your tags and drop the symbols onto one of your displays.