Power BI solution for turning Milking Robot Data into Farm Intelligence

GEA manufactures and installs milking robots in dairy farms to automate milk extraction from cows. Each robot continuously logs operational data like milking time, cleaning time, and machine status and stores on a local server at the farm. As the deployment footprint grew, GEA required a scalable mechanism to collect this data across all farm sites and put it in front of the right people. This engagement delivered an end-to-end data harvesting, processing, and analytics platform to do exactly that.

Customer Profile

GEA is a global leader in processing technology, supplying advanced equipment and components to the food, beverage, pharmaceutical, and allied industries. The group specialises in sophisticated production processes, offering a comprehensive portfolio that spans machinery, industrial plants, process technology, and end-to-end services, including their connected milking robot systems deployed across dairy farms.

Customer Requirement

Milking robots at each farm continuously generate performance data, but that data was siloed on local servers with no way to extract, consolidate, or analyse it centrally. GEA had no process to harvest this data at scale, which meant machine performance could only be assessed manually and on-site.

The core requirement was a mechanism to collect performance data from every connected farm, process it reliably, and surface it in a way that allowed GEA and their farm customers to evaluate machine health, track operational metrics, and identify malfunctions or underperformance before they escalated.

Our Solution

  • Data from each farm's local server is pushed to the cloud automatically via Azure Functions, eliminating the need for manual exports or on-site data handling.
  • ETL pipelines built with Azure Data Factory and Databricks clean, transform, and consolidate data from all farm sources into a unified, queryable structure.
  • Processed data flows into Azure SQL Database and Azure Data Lake Storage (ADLS), forming a reliable storage layer for both operational reporting and historical analysis.
  • Power BI reports are embedded directly into GEA's customer-facing web application, giving farm operators a live view of their robot performance metrics.
    • SSAS cubes provide multidimensional data models for ad hoc analysis and deeper reporting needs.
    • Power BI dashboards surface milking duration, cleaning cycle data, and deviation indicators in a format farm managers can read without technical training.

Benefits

  • Milking robot performance is now visible across every farm site from a single dashboard, without anyone needing to visit or manually pull data.
  • Malfunctions and operational anomalies in the milking system surface in the dashboard, enabling GEA to act before a fault disrupts farm operations.
  • The data pipeline from farm server to reporting layer is fully automated, removing manual handling and making the system scalable as new farms are onboarded.
  • Farm customers can access and interpret their own performance data directly through the embedded application, without needing to contact GEA support.