🧮 Forecast → MPS → MRP → Capacity → Cost
Five linked planning steps, the way they actually chain together in an operation. Build a seasonal forecast, net it into a production schedule, explode that through the bill of materials with time-phased MRP, check it against line capacity, and roll the whole thing up into cost and working capital. Everything recalculates on every keystroke. Nothing leaves your browser.
This is the C-Class trap in miniature: set every index to 1.00 and the forecast becomes the annual average spread evenly — which is wrong in every single month for an event-driven product. The article explains why →
Projected available = opening stock + production − forecast demand. When it would fall below the safety stock floor, a production order is raised and rounded up to the minimum batch and batch multiple. Production must start 21 days before the period it lands in.
| Component | Qty/unit | Scrap % | On hand | On order | Lead (d) | MOQ | Unit cost | Source |
|---|
CSV columns: Component,Qty per unit,Scrap %,On hand,On order,Lead time days,MOQ,Unit cost,Source. A header row is detected and skipped.
Each component runs its own MRP record. Gross requirement comes from the production plan, projected available carries forward, and any planned order is offset backwards by the component lead time to give the release period. PAST means the release period is already behind us — that requirement cannot be met on the current plan.
Enter available output per period. Load above 100% cannot be produced in that period — the plan has to be smoothed by building earlier, which pulls the material requirement earlier too. Change a capacity figure and step 3 does not move on its own; smoothing is a decision, not an automatic recalculation.
This is the cash profile of the plan — when purchase orders are released, not when goods arrive or when invoices are paid. Payment terms shift it later; the commitment is made here.
Steps 2 and 3 are the “Raw Material Availability Check” gate in the demand-to-delivery flow — the decision point that routes a plan either straight into production or into a three-to-six month procurement cycle.
An index of 1.00 is an average period. The reason this step exists separately is that a single annual average — the flat index — is wrong in both directions for anything event-driven: it over-forecasts the quiet periods and badly under-forecasts the peaks. Press “Set flat index” to see the damage on the chart.
Lot sizing is what turns a smooth requirement into a lumpy production plan. That lumpiness is then inherited by every component below it, which is why material demand is almost always spikier than customer demand — the bullwhip starts here, inside your own plan.
The offset is the whole point. A component with a 95-day lead time consumed in period 6 has to be released in period 2 — and if that period has already passed, the plan is not achievable no matter how well everything downstream is executed.
Material feasibility and capacity feasibility are different questions and a plan has to pass both. A period above 100% load has to be smoothed by building earlier, which moves the material requirement earlier too. This tool flags the overload rather than silently levelling it, because how you smooth is a commercial judgement — build early and carry inventory, or move the promise date.
Two numbers matter more than the total. The first is MOQ excess — value bought above what the plan needs, which becomes carried inventory by construction rather than by mistake. The second is the spend profile: the same annual cost is a very different business problem if two-thirds of it commits in one quarter.