One plan, built from your own numbers.
plnnd connects to your orders, stock levels, lead times and bill of materials. Everything it shows, from a forecast to a criticality score, is calculated on that data.
Six models. Switch until it fits.
Every material behaves differently. Pick a model per material, compare how stock and risk change, and keep the one that matches reality. Each forecast comes with a range, so you can see how sure it is.
- Baseline (90-day mean)A simple reference point. Every other model has to beat it.
- Moving Average (12w)Steady demand that shifts slowly.
- Croston TSBIntermittent demand, such as spare parts with many zero weeks.
- Holt-Winters SeasonalDemand with a clear yearly or quarterly pattern.
- AutoETS (Multiplicative)Trend and seasonality that grow with volume.
- Global LightGBM MLMany materials at once, learning patterns across your whole range.
Ask "what if" before you commit.
A supplier slips three weeks. Demand jumps 20%. A line goes down for maintenance. Run the scenario and see which orders, materials and lines it touches, and when.
Know what to act on first.
plnnd flags the week each material drops below safety stock. It then ranks every material by criticality, based on lead time, stock cover and what depends on it, so the most urgent work sits at the top.
Connect your own AI model. Ask it why.
Link the model you already use with your own API key. plnnd only sends it encrypted data. In return you get plain-language answers about your plan: what drives a forecast, why a material is critical, and which forecast model suits it best.