Saint-Gobain Sekurit improves SKU-level forecast accuracy 15% and cuts inventory 9.25% with Flowlity

Saint-Gobain Sekurit's automotive glass replacement division planned demand for more than 10,000 SKUs across 30 distribution centers using fragmented ERP systems, manual spreadsheets and macro-level sales forecasts. This led to systematic forecast errors at SKU level, excess inventory and service level gaps, and the team had no ability to anticipate seasonal peaks. Stockouts occurred more often than acceptable.

Sekurit chose Flowlity's AI-powered demand planning platform as a single tool to replace the patchwork of ERPs and spreadsheets. It was deployed progressively across the supply chain, covering sales forecasting, inventory management and purchase forecasting at each distribution center, collaborative planning at the central warehouse, and production forecasting and stock management at the plants. Planners also use a tactical simulation module to test buffer-level changes and automated alerts that flag outliers and shortages.

Forecast accuracy at SKU level improved by 15%, inventory fell by 9.25% and product availability rose from 95.8% to 97.2%. The supply chain director cites an average availability of 97% across the 30 distribution centers, up two points on the previous year, and the availability curve became more stable through the year, removing the deep seasonal dips that previously disrupted service to garages and fitting centers. The team has since built operational KPIs on Flowlity data, including OTIF, planning adherence and stock health indicators.

• Consolidate data from fragmented ERPs into one reliable pipeline before expecting an AI model to add value.
• Roll out progressively across distribution centers, the central warehouse and plants so each phase delivers value and builds the foundation for the next.
• Pair forecasts with simulation, alerts and collaborative planning, since the gains came from how planners act on forecasts as well as the forecasts themselves.