Safety stock and reorder points: the formulas and when to use them

Safety stock is the inventory you hold to absorb variation — demand that runs hotter than expected, or a delivery that arrives late. Too little and you stock out; too much and cash sits on shelves. Here's how to size it.

The reorder point

The reorder point is the stock level at which you should place an order so it arrives before you run out:

reorder point = average daily demand × lead time (days) + safety stock

The first term covers expected demand during the lead time. Safety stock covers the variation.

Safety stock when demand varies

If lead time is reliable but daily demand fluctuates, a widely used formula is:

safety stock = z × σ(daily demand) × √(lead time days)

• σ(daily demand) is the standard deviation of daily unit sales.
• z is the service-level factor: about 1.28 for 90%, 1.65 for 95% and 2.33 for 99% of cycles without a stockout.

Worked example

A product sells an average of 20 units a day with a standard deviation of 6. The supplier's lead time is 16 days, and you want a 95% service level.

• Safety stock = 1.65 × 6 × √16 = 1.65 × 6 × 4 = 39.6, so 40 units
• Reorder point = 20 × 16 + 40 = 360 units

When stock on hand plus stock on order falls to 360, place the next order.

When lead time varies too

If deliveries are unpredictable, lead-time variation often matters more than demand variation. One common extension is:

safety stock = z × √( L × σ(d)² + d² × σ(L)² )

where d is average daily demand, L the average lead time, and σ(L) the standard deviation of lead time in days.

A simpler approach: buffer days

Many teams don't have reliable variability data, especially for new products. A practical alternative is to hold a fixed number of days of extra cover:

safety stock = buffer days × average daily demand

A 14-day buffer on a product selling 20 a day is 280 units. It's less precise than the statistical formula, but easy to explain and adjust: increase the buffer for unreliable suppliers or critical products, and reduce it for slow movers that tie up cash.

Choosing service levels by product

Not every SKU deserves the same protection. A common pattern is to give your highest-revenue products (A items in an ABC analysis) a higher service level, and accept more risk on the long tail.

In Stokrio

Stokrio uses the statistical method: safety stock covers both forecast error and lead-time variability (measured from your received purchase orders) at a target service level, which is higher for A-class products and lower for C-class. See replenishment.