Skip to content

Analyze Data Agent

You can leverage Production Copilot, adapted to Insights Hub, to query and instantly analyze shopfloor data from assets and operations available in Insights Hub Monitor, thereby reducing the time you spend to analyze production performance and related issues​.

To do this, users can select the production data which is in scope​ using Analyze > Time Series.
Using the Production Copilot chat, you can ask questions by selecting a predefined one or enter your own question and get responses.

Analyze time series data of asset variables, including statistical KPIs (min, max, avg, std), detecting trends and patterns, identifying outliers, correlate events with timeseries values. Advanced machine learning algorithms are not yet supported by this agent.

If required, users can refine their query based on the results.

Production Copilot - ask questions regarding data in chart

How To

  1. Navigate to "Analyze > Time Series".
  2. Open the Production Copilot by clicking the top button in the OS Bar.
    Copilot - Analyze sidebar
  3. Select assets, time series or events data or open an existing analyze profile.
  4. Select the time range.
    • Use small time ranges so that the data context for the request does not become too large. Otherwise, an error message might be displayed.
    • Each time the time range is adapted, the Copilot will use the newly received data and clean the chat history.
  5. Ask questions.
    • Use one of the predefined questions.
    • Enter your own questions.
    • From time to time, you might need to clean the history, when the data context is getting too big.

Limitations

  • You will get the best results for questions asked in English
  • Avoid selecting larger time ranges, because the data context will get too big
    If an error message is displayed, zoom in and try again
    Token error
    The token limit for the context window in total is 128.000. But, this is consumed by different sources like chat history, skill output, agent instructions, llm output etc.

Last update: October 7, 2025