Developing, Deploying And Scaling A Global Asset Management Capability
summary
Aggreko builds a cloud pipeline on Azure/Databricks ingesting telematics from 10–15,000 assets to run condition‑based alarms, currently deploying about 30 alarms/models that trigger texts, emails and work orders and deliver millions in avoided failures and parts savings per year. The stack uses Prefect, Kubernetes, Python and Power BI; models run on scheduled Databricks jobs and CICD, they work but create maintenance and observability pain points as models scale and pipelines or schemas drift.
AI-written summary · about
chapters (8)
- 0:00Introduction and team overview
- 1:01Greco’s global asset management
- 5:16Modern data platform journey
- 8:49Use cases and business alignment
- 12:22Prioritizing and quantifying value
- 13:25Scaling models and future challenges
- 18:20Key advice for data teams
- 19:23Q&A session
related talks
18mDatabricks drives predictive maintenance, grid optimization with analytics
41mPredictive Maintenance with Azure Synapse
44mSmart application on Azure at Vattenfall
27mTerraforming Data Analytics Environments
19mLessons from Building a Large-scale, Multi-cloud Data Platform at Databricks
24mAsset based orchestration: the future of data orchestration
29mBuilding data mesh on Databricks and Azure for a large manufacturing company
44mData Mesh: Lessons from the trenches