SKU-Level Demand Forecasting and Anomaly Detection for Retail Supply Chains

[#1]

Team Name	CatSup
Track	Retail Track
Use Case	Deliver 28-day SKU-level probabilistic demand forecasts combined with residual-based anomaly detection and automated procurement alerts so procurement teams can proactively mitigate supply interruptions and optimize inventory spend.
Technology	SAS Visual Forecasting (for demand modeling) and SAS Visual Analytics (for dashboards and anomaly alerts), combined with lightweight Python experiments (Prophet/Isolation Forest) for benchmarking.
Region	AP
Team lead	Dr. Chiao Ya Chang
Team members	Venkata Sai Praneeth Uppala - @iamadatanerd, Dr. Chiao Ya Chang - @Chiaoya, Siddharth Bagale - @bagalesid123
Social media handles	(1) Dr. Chiao-Ya Chang | LinkedIn, (2) Venkata Sai Praneeth Uppala | LinkedIn, (3) Siddharth Bagale | LinkedIn
Is your team interested in participating in an interview?	Y
Data strategy	M5 dataset + synthetic anomalies
Optional: Expand on your technology expertise	Our team specializes in bridging advanced forecasting techniques with practical supply chain impact. We use SAS Visual Forecasting for scalable hierarchical time series models and uncertainty quantification, and SAS Visual Analytics to operationalize results through dashboards and alerts. Python libraries (Prophet, ETS, and Isolation Forest) are used for benchmarking, anomaly detection, and rapid prototyping. Collectively, this expertise ensures the solution is both technically rigorous and production-ready, delivering measurable business value in retail supply chains.