A reorder point is the stock level at which you place the next order, and it has exactly two parts: what you will consume while the order is in transit, plus a buffer for the weeks that go wrong. Written out, that is demand during lead time, plus safety stock. Every other way of setting it — a round number somebody liked, two pallets because two pallets fit — is a guess wearing a number's clothing.
Small plants usually get the first part roughly right and the second part badly wrong, in both directions at once: months of cover on cheap, stable materials, and none at all on the one import that stops the line.
Demand during lead time
Start with the easy half. If you use 400 kg of a fatty alcohol a week and your supplier takes four weeks door to door, you will consume 1,600 kg between placing an order and receiving it. That is the floor. Order at 1,600 kg and, if nothing varies, stock reaches zero exactly as the truck arrives.
Two things people get wrong here. The first is using the supplier's quoted lead time rather than the one you actually experience — quoted four weeks, actually six with customs, is a two-week hole in your calculation. Use your own receipt history. The second is forgetting that lead time includes your own internal steps: approval, payment terms, the day the container sits before it is unloaded.
Safety stock is about variability, not caution
The buffer exists because neither usage nor lead time is constant. The relevant question is not "how careful do we want to be" but "how much do these two things actually move".
For a material where usage is steady and the supplier is reliable, a small buffer is genuinely enough. For a material where either varies a lot, you need more cover for the same service level — and the cover scales with the variability, not with the price or the importance of the material.
The practical version, for a plant without a statistician:
- Pull the last twelve months of consumption for the item, by month.
- Note the highest month, and your average month.
- The difference is what a bad month looks like. Cover a bad month's worth of the lead-time period, and you have a defensible buffer.
This is cruder than the textbook formula and lands in roughly the same place for most SME materials. Where it lands badly is items with real seasonality, where the right answer is a reorder point that changes by season rather than a bigger buffer all year.
Not every item deserves the same attention
Classifying items by annual spend — the familiar pattern where a small share of items accounts for most of the money — tells you where to spend your effort. The logic is straightforward:
- High-value, few items. Tight reorder points, short cycles, reviewed often. Carrying cost here is real money.
- Mid-value. Set a reorder point, review quarterly, leave it alone in between.
- Low-value, many items. Order in bulk, hold generous cover, stop thinking about it. The stationery of your bill of materials. Optimising these costs more in attention than it saves.
The failure mode is treating every line the same: either agonising over pallet wrap or, more commonly, applying the same casual approach to the one imported active with a twelve-week lead time.
What a stockout actually costs
Buffers get cut when someone looks at working capital in isolation. The comparison that matters is against the cost of running out, which for a manufacturer is rarely just a late delivery. A stopped line means idle labour, a changeover to something else, a campaign split in two, and often a customer who finds a second supplier. Against that, four weeks of cover on a critical raw material is cheap.
Say it in those terms when you set the number, and write the reasoning down next to it. A reorder point with a sentence explaining it survives the next cost review. A bare number does not.
Where the system should help
Reorder points are only useful if something watches stock against them without a person remembering to look. In Aourix, procurement runs as plan, source, order, receive — requirements are consolidated into a plan rather than each shortage becoming its own separate scramble, so the buying decision sees all the demand for a material at once instead of one batch at a time.
That consolidation is where the saving is. Three production batches each short of the same material is one order, not three, and one lead time rather than three.
This week
Take the five materials that have stopped a batch in the last year. For each, write down your real lead time from receipt history, your highest consumption month, and the current reorder point if one exists. If the reorder point is below lead-time demand alone, you have found why the line stopped — and the fix is arithmetic, not a supplier conversation.