Skip to main content
Solved

ML Workbench Examples / GitHub Repositories

  • September 22, 2026
  • 5 replies
  • 74 views
Ayse Nurperi Ozgen
Level 9
Forum|alt.badge.img+14

I’m currently exploring the Celonis Machine Learning Workbench and looking for some practical examples to learn from.

I found the official PyCelonis Examples repository on GitHub, which is very helpful for understanding PyCelonis and interacting with Celonis resources. However, I was wondering whether there are also repositories, sample notebooks, or templates focused more on end-to-end ML Workbench use cases.

I’m particularly interested in Supply Chain examples such as:

  • Late/on-time delivery prediction
  • Supplier or procurement risk prediction
  • Inventory-related predictions
  • Lead time prediction
  • Bottleneck or process outcome prediction

I also came across the older O2C – On-Time Delivery Prediction example, but it seems to rely on the legacy celonis_ml package.

Does anyone know of any up-to-date PyCelonis 2.x / ML Workbench GitHub repositories, Jupyter notebooks, or sample projects that could be used as a starting point?

And if you have built something similar yourself and can share the general approach or architecture, I’d love to learn from it. 😊

Thanks in advance!

Best answer by abhishek.chatu14

This is the approach that I would follow for the same
 

  1. Define prediction point — only use information available at that point in time.
  2. Feature engineering — supplier performance, lead time, PO changes, quantity, material, plant, etc.
  3. Export data to ML Workbench using PyCelonis 
    • PyCelonis Documentation
  4. Train and validate ML model — start with Logistic Regression / Random Forest / XGBoost.
  5. Generate predictions for current/open POs.
  6. Bring predictions back into Celonis as risk scores/predicted outcomes.
  7. Use predictions operationally — dashboards, Action Flows, alerts/interventions.
  8. Measure impact — compare predicted-risk cases, interventions, and actual outcomes.
  9. Supply-chain ML examples to use as modelling references:
    Supply Chain Analytics — Late Delivery / Shipment Duration Prediction
    Supply Chain Late Delivery Prediction
    Late Delivery Prediction — ML/API Example
View original

    5 replies

    abhishek.chatu14
    Level 12
    Forum|alt.badge.img+8

    This is the approach that I would follow for the same
     

    1. Define prediction point — only use information available at that point in time.
    2. Feature engineering — supplier performance, lead time, PO changes, quantity, material, plant, etc.
    3. Export data to ML Workbench using PyCelonis 
      • PyCelonis Documentation
    4. Train and validate ML model — start with Logistic Regression / Random Forest / XGBoost.
    5. Generate predictions for current/open POs.
    6. Bring predictions back into Celonis as risk scores/predicted outcomes.
    7. Use predictions operationally — dashboards, Action Flows, alerts/interventions.
    8. Measure impact — compare predicted-risk cases, interventions, and actual outcomes.
    9. Supply-chain ML examples to use as modelling references:
      Supply Chain Analytics — Late Delivery / Shipment Duration Prediction
      Supply Chain Late Delivery Prediction
      Late Delivery Prediction — ML/API Example
      Ayse Nurperi Ozgen
      Level 9
      Forum|alt.badge.img+14

      Thank you so much for the detailed explanation! This is exactly what I was looking for. 

      Really appreciate you taking the time to share this!

        Chitra
        Celonaut
        Forum|alt.badge.img+2
        • Chitra
        • Celonaut
        • September 28, 2026

        Thank you ​@abhishek.chatu14 for the detailed answer👏🏻

        ​@Ayse Nurperi Ozgen 

        I’m glad to hear that the solution provided in the community helped resolve your issue!

        If you have a moment, could you please mark the answer as "Best answer"?

        This helps other community members who might encounter the same issue find the right answer more quickly.

        Thank you for being a part of our community! 

          Ayse Nurperi Ozgen
          Level 9
          Forum|alt.badge.img+14

          Thank you ​@Chitra ! 😊 It was indeed the best answer and helped me resolve my issue. I’ve marked it as “Best answer”. Thanks again, ​@abhishek.chatu14 for your support! 👏🏻

            Chitra
            Celonaut
            Forum|alt.badge.img+2
            • Chitra
            • Celonaut
            • September 28, 2026

            Thank you for the quick response ​@Ayse Nurperi Ozgen 

            Appreciate it 🙂

              Community Code of ConductAccessibility statement