Manufacturing
A secondary steel producer that cut logistics cost 10% and forecast demand to 98.77% accuracy
Logistics cost down 10%, a live spread P&L for the CFO and MD, and demand forecast to 98.77% accuracy.
In numbers
−10%logistics cost, a direct EBITDA gain
98.77%monthly forecasting accuracy, in validation
6 wksfrom kick-off to a live platform
The situation
A secondary steel producer needed better sales prediction, production planning and raw material management. Procurement, yield tracking and dispatch scheduling ran through disconnected ERP modules and spreadsheets. Logistics cost ran at 16–18% of revenue, nearly double the global benchmark, with no live view of spread economics.
What we did
- Built AI-driven demand forecasting models.
- Built inventory optimisation models.
- Set dynamic inventory buffers linked to demand.
- Set up a data-driven framework to optimise production throughput, control inventory levels and minimise waste.
- Built a live spread P&L for the CFO and MD.
The result
- Forecasts held up in validation, and now set the inventory buffers.
- Inventory buffers move with demand.
- Sales, production and raw material are planned from one data-driven framework.
- Logistics cost fell 10%, a direct EBITDA gain.
- The CFO and MD now see spread economics live. The platform went live six weeks after kick-off.
More on this theme: Manufacturing, or what we do.
You built the business. We help it scale.
Start a conversation