Replenishment Calculator

Behind the number

Replenishment quantity restores stock to a target position: target level − stock on hand − stock on order + backorders. The last two terms are the ones most often dropped, and each drops in a different direction.

The target level covers demand over the exposure period, lead time plus review interval, plus safety stock. Everything else is bookkeeping about what is already committed.

A real example

Weekly review, three-week lead time, so a four-week exposure period. Demand 40 a week gives 160 units, plus safety stock for four weeks of variability: 1.65 × 12 × √4 = 40. Target level 200.

Stock on hand 88, stock on order 60, backorders 12. Replenishment = 200 − 88 − 60 + 12 = 64 units.

Drop the on-order term and the answer becomes 124, nearly double, and 60 units of it already in transit. Drop the backorder term instead and it becomes 52, leaving twelve customers waiting through another cycle.

The usual mistakes

It assumes stock on order will arrive within the lead time. Where a shipment is already late, counting it as available understates what is needed, and the replenishment run will look correct while the shelf empties.

Allocated stock is the mirror problem. Units on hand but committed to specific orders are not available to meet new demand, and including them in the on-hand figure produces a replenishment quantity that is systematically too small.

Using the result

Use available-to-promise rather than on-hand in the calculation: on hand, less allocated, less anything unsellable. That is the quantity that can actually meet the next order, and it is frequently well below the headline stock figure.

Then run replenishment on a fixed schedule rather than when someone notices. The whole calculation assumes a review interval, and an irregular one makes the target level wrong in an unpredictable direction.

Consolidating replenishment across products

Running replenishment per product produces many small orders; running it per supplier on a schedule produces one larger order and spreads a single ordering cost across everything in it.

The technique is to set a review day per supplier and replenish every product from that supplier on it, including items not yet at their trigger if they are close. Adding an item a week early costs a week of carrying cost on a few units; leaving it out costs a whole extra order.

That is the joint replenishment problem, and the practical rule is simple enough to use by hand: include anything that would hit its reorder point before the next scheduled review. It typically cuts order count substantially with almost no increase in stock, and the freight savings alone often justify it.

Replenishment should be calculated per location and sourced centrally, not calculated centrally and split. A total requirement divided by site allocates stock in the wrong proportions whenever demand is uneven, which it always is.

The practical arrangement is a target level per location and a single consolidated purchase against the sum of the gaps. That captures the ordering economics of one large order while placing the stock where the demand actually is.

It also makes transfers visible. Where one site is above target and another below, the replenishment run should propose a transfer before it proposes a purchase, and building that priority into the calculation prevents a great deal of unnecessary buying.