RetailUK4-8 weeks
AI-Driven Inventory Optimisation
6 weeks (PoC) + 8 weeks (rollout)
ML Lead + 2 Data Scientists
1The Challenge
Overstocking and stockouts costing £10M+ annually. Demand forecasting relied on spreadsheets and intuition.
2Our Solution
Built ML-powered demand forecasting integrated with inventory planning systems.
Approach
- Unified historical sales, promotions, and external data
- Trained demand forecasting models by product category
- Integrated predictions with replenishment workflows
- Built monitoring dashboards for forecast accuracy
3Results & Impact
35% reduction in overstock
50% reduction in stockouts
£3M+ annual savings
85% forecast accuracy achieved
Key Takeaways
- PoC approach de-risked the investment decision
- Simple models outperformed complex ones initially
- Change management was as important as model accuracy
Technology Stack
DatabricksAzure MLdbtPower BI
Ready to Achieve Similar Results?
Let's discuss how we can help transform your data capabilities.