Reorder point: the formula, and how to set it properly

The reorder point is the single number that decides whether you ship or apologise. It is the inventory level at which a replenishment order has to be raised so that stock arrives before the shelf empties. Get the reorder point formula wrong by a few hundred units and you either sit on cash you did not need to spend, or you explain a backorder to a customer who has options.

Reorder point (ROP)

The inventory level that triggers a replenishment order. It covers expected demand over the replenishment lead time plus a buffer for the variability in that demand and that lead time. It answers when to order; the order quantity answers how much.

The reorder point formula

ROP = (Average daily demand × Lead time in days) + Safety stock

• Average daily demand — units consumed per day over a recent, representative window. Take it from issues or shipments, not from forecast, and exclude periods when you were stocked out (those days understate real demand and shrink your buffer).
• Lead time — the measured gap between raising the purchase order and the stock being available to pick. That includes supplier production, transit, receiving and put-away. It is almost never the number the supplier quoted.
• Safety stock — the buffer that absorbs variability. Without it, the reorder point protects you against the average outcome only, which by definition fails half the time.

A worked example, start to finish

One purchased item at a mid-market distributor. Everything below is derived from these five inputs, and every step is shown so you can check it against your own numbers.

Step one — demand during lead time. 160 units × 15 days = 2,400 units. That is what you will consume while the order is in flight, if nothing surprises you.

Step two — safety stock, treating lead time as fixed. Safety stock = z × σ demand × √lead time = 1.6449 × 38 × √15. The square root of 15 is 3.873, so 38 × 3.873 = 147.2, and 147.2 × 1.6449 = 242 units.

Step three — add them. 2,400 + 242 = 2,642 units. When inventory position falls to 2,642, raise the order. At a 5,000-unit order quantity rounded to a pallet, the order-up-to level is 2,642 + 5,000 = 7,642 units.

Lead-time variability changes the answer more than demand does

The 242-unit buffer above assumes the supplier always arrives on day 15. Real suppliers arrive on day 12 or day 21. When both demand and lead time vary, the two variances add and you take the square root of the total:

Safety stock = z × √( LT × σdemand² + demand² × σleadtime² )

Substituting: 15 × 38² = 15 × 1,444 = 21,660 for the demand term, and 160² × 3² = 25,600 × 9 = 230,400 for the lead-time term. Total variance is 252,060, so the combined deviation is √252,060 = 502 units. Safety stock becomes 1.6449 × 502 = 826 units, and the reorder point rises from 2,642 to 3,226 units.

Read that ratio before you spend anything. The cheapest way to cut this reorder point is not a better forecast — it is a supplier conversation. Getting lead-time deviation from 3 days to 1 day drops the combined deviation to √(21,660 + 25,600) = √47,260 = 217 units, safety stock to 357 units, and the reorder point to 2,757. That is 469 fewer units of permanent buffer, worth $3,827 a year in carrying cost at $8.16 a unit, for the price of a scheduled call and an honest lead time in the item master.

Service level is a spending decision, not a target

The z value is the only input in the formula that is a pure policy choice, and it is where the money goes. Using the combined deviation of 502 units, here is what each service level costs on this one item.

Periodic review stretches the exposure window

The formula above assumes continuous review: the moment inventory position touches 2,642, an order goes out. If a planner runs the replenishment report every Monday instead, demand can cross the trigger on Tuesday and sit there unnoticed for the rest of the week. The exposure window is lead time plus the review period, not lead time alone.

With a 5-working-day review cycle, the protection interval becomes 15 + 5 = 20 days. Demand over that window is 160 × 20 = 3,200 units. Safety stock on demand variability alone becomes 1.6449 × 38 × √20 = 1.6449 × 38 × 4.472 = 280 units, so the order-up-to level is 3,480 units. Keep the continuous-review figure of 2,642 while reviewing weekly and you are late by 2.5 days of demand on average — 400 units — and by 800 units in the worst case.

How to set a reorder point in your ERP

1. 01

   Pull 90 days of actual issues by item

   Daily quantities, not monthly totals — you need the deviation as well as the mean. Strip out intercompany transfers, returns and any day the item was at zero on hand. A stockout period drags the average down and makes the buffer you calculate too small, which caused the stockout.

2. 02

   Measure lead time from receipts, not from the item master

   Take the last 15 to 20 receipt lines for the item or supplier, compute order date to put-away date for each, then take the mean and the standard deviation. The quoted lead time is a sales input; the received lead time is a fact. If the two differ by more than about 20%, fix the master data before you touch anything else.

3. 03

   Choose a service level per ABC class, not per item

   Rank the catalogue with an ABC analysis, then pick three z values and attach them to classes: for example 2.3263 for A, 1.6449 for B, 1.2816 for C. Per-item service levels feel precise and become unmaintainable within two quarters.

4. 04

   Calculate safety stock with the combined-variability formula

   z × √(LT × σdemand² + demand² × σleadtime²). Use the demand-only version only when you have measured lead-time deviation and found it genuinely near zero — blanket orders with fixed weekly releases are the honest case for that.

5. 05

   Add the review period if replenishment is not continuous

   Replace lead time with lead time plus review interval in both the demand term and the safety-stock term. Most ERPs with a live replenishment engine give you continuous review; a Monday-morning spreadsheet does not.

6. 06

   Key the level and the buffer into separate fields, once — then recheck quarterly

   Enter demand during lead time plus safety stock as the reorder point or minimum, and the order quantity separately. If your ERP also has a safety stock field that it adds to the reorder point, put only lead-time demand in the reorder point field — double-counting the buffer is the most common single error on an item record. Demand drifts gradually but lead times move in steps, so a change of supplier, freight mode or port routing should trigger a recalculation that week rather than at the next annual review.

Where reorder points quietly fail

• The buffer is counted twice — once inside the reorder point and again in a separate safety stock field. You end up holding 484 units where 242 was the policy, on every item.
• Demand was measured during a shortage. Shipped units are censored demand: you cannot ship what you do not have. Use ordered quantity where you can capture it.
• Lead time is stale. The master says 7 days, the last 20 receipts average 15. The reorder point inherits the fiction and no formula will save it.
• One reorder point across multiple warehouses. Demand rate, lead time and deviation are all location-specific. A single national ROP over-stocks the small branch and starves the big one.

The hard part is the inputs, not the arithmetic

Everything above is four multiplications and a square root. The week goes on assembling the inputs: 90 days of daily issues by item and location, order-to-receipt dates for the last 20 receipts by supplier, and current inventory position net of allocations, all on one time basis. In most ERPs that is a saved search, an export, a pivot table and someone's Tuesday — repeated every quarter.

This is the class of question ERPray was built for. Ask "average daily issues and standard deviation by item for the last 90 days, top 100 items by usage value" and get the answer computed live from your own account, with the query printed underneath so you can check what counted as an issue before you set policy on it. Then size the order quantity with the EOQ calculator and check the buffer against your coverage using the days of supply calculator. For the buffer itself, safety stock formulas compared puts four methods side by side; EOQ explained covers the quantity decision; and the ABC analysis guide decides which items deserve a calculated reorder point at all.