Challenge
An international manufacturing organisation was undertaking a global ERP roll-out. However, the programme team had neglected to consider the requirements for a data platform to make use of newly digitised processes and generated data.
The existing reporting platform ran on legacy technologies, making it slow to refresh data, difficult to access, and restricted to basic paginated reporting.
The organisation sought to build a future proof data platform which would allow them to leverage modern data processing tools and new technologies like AI.
Solution
An Azure hosted modern data platform was designed to store and process data ready for reporting using Power BI.
Data was ingested directly from the ERP system into Azure Data Lake storage in near-real time, before being converted to an efficient Parquet data format using Azure Data Factory.
Azure Synapse Analytics Serverless was used to process data through a medallion architecture. The bronze layer being used to provide raw data, the silver layer to provide cleansed and validated data and the gold layer to provide data ready for visualisation.
Azure DevOps was used to provide CI/CD capabilities, allowing for development, testing and production to be provided in separate environments.
With the core solution implemented over 20 historic reports and 40 datasets were migrated to the new data platform.
Project Details
Project Duration - 2 week design, 8 week build, 12 week platform migration
Project Team - 1 delivery lead, 1 data architect, 3 data engineers
Benefits
Full migration from legacy platform completed without disrupting existing programme timelines.
Access to full Power BI Data visualisation capabilities for interactive reporting.
Available access to full suite of Azure data, IoT and AI tooling.
Report refresh time down from 24 hours to 2 minutes.
Additional data exploration, workshops and data roadmap provided at no cost.
