Make central-to-regional warehouse replenishment simple: auto-compute safety stock, give daily suggestions, confirm with one click, and consolidate truckloads. Lightweight, fast and explainable — currently at the pilot (POC) stage.
Replenishment relies on Excel and gut feel, setting safety stock SKU by SKU; net demand recomputed by flipping through tables every day; consolidating full truckloads to save freight is guesswork.
Turn the work of “setting safety stock SKU by SKU and recomputing net demand daily” into a closed loop: compute in one click, suggest daily, and confirm by a person.
| Capability | Maturity | Notes |
|---|---|---|
| Safety-stock & replenishment-parameter engine | Demo | Demand pattern (6 types) + ABC-XYZ + formula method; single-customer scope |
| Replenishment simulation & fill-rate analysis | Demo | Simulated; evaluates how a replenishment strategy affects fill rate |
| Daily replenishment + editable release + ETA | Demo | Daily MRP suggestions, editable, confirm to create a transfer order, auto ETA |
| Daily replenishment loop (archive / live inventory / rolling forecast) | Demo · bridged | Decision archive, inventory rolls with documents, rolling forecast (bridged) |
| Truck-load consolidation | Demo | Full-truckload by pallet (First-Fit-Decreasing); trips ↔ load-rate trade-off |
It is currently at the pilot (POC) stage; capabilities are all POC / demo. It is positioned as lightweight, fast, explainable replenishment decisions — not competing on the multi-echelon planning depth of o9 / SAP IBP.
No. Here the AI (LLM) only generates explanatory text; it does not decide how much to replenish or write decision data. Replenishment quantities are computed by rule-based formulas and confirmed by a person.
No. The sales forecast uses best-of statistical / ML time-series models (not an LLM); safety stock and replenishment quantities use rule-based formulas.
