Back to portfolio
Data EngineeringAnalyticsAI & Automation

Supply Chain Stock Analytics & Alerting Platform

An event-driven stock analytics platform that ingests inbound email files, calculates allocation and anomaly signals, and delivers automated reports and failure alerts.

75%

Better query performance

65%

Faster dashboard refreshes

Automated

Output delivery & failure alerts

Problem

Stock position files arrived by email and were processed manually, which made allocation calculations slow, inconsistent and hard to trust. Analysts lacked a dependable signal for anomalies, and failures in the chain were only discovered after reports were already late.

Architecture

Stage 01 · Inbound Email Files

Stock files delivered by email from operational teams.

Simplified flow · No confidential implementation details

Solution

  • Built event-driven ingestion with Power Automate capturing inbound files into Azure Blob Storage, then Azure Data Factory loading them into Azure SQL Database.
  • Implemented complex T-SQL stored procedures for stock allocation, anomaly detection and calculated output generation.
  • Optimised fact tables with columnstore indexes and composite indexing to keep heavy analytical queries interactive.
  • Added automated observability: delivery of calculated output files plus email notifications when a step fails.

Key engineering contributions

  • Developed event-driven ingestion workflows from inbound email to blob storage.
  • Implemented complex T-SQL stored procedures for allocation and anomaly logic.
  • Built inventory stock allocation and anomaly detection logic.
  • Optimized fact tables using columnstore and composite indexing.
  • Configured automated observability and email notifications.

Technologies

  • Azure Data Factory
  • Azure Blob Storage
  • Azure SQL Database
  • T-SQL
  • Power Automate
  • Power BI

Want the full story behind this platform?

Happy to walk through the engineering decisions and trade-offs.

Mukesh Varma Gangaraju · Senior Data Engineer at Stryker