Safety stock calculator
Last reviewed
Safety stock is the extra stock that covers sales running faster, or deliveries arriving later, than you planned. Here are the two common formulas side by side.
What is safety stock?
Safety stock is the extra inventory you hold in case sales run faster than expected or a delivery arrives late. It is the buffer between what you plan for and what really happens. The safety stock glossary entry has the definition. This page helps you size the buffer.
The two formulas
Simple method (max minus average):
Safety stock = (max daily sales × max lead time) − (avg daily sales × avg lead time)
It assumes your busiest day and your slowest supplier could happen together, and covers that case. It needs no statistics.
Statistical method (service level):
Safety stock = Z × σ(daily sales) × √(lead time in days)
σ is the standard deviation of your daily sales, a measure of how much they vary. Z comes from the service level:
| Service level | Z |
|---|---|
| 90% | 1.28 |
| 95% | 1.65 |
| 98% | 2.05 |
| 99% | 2.33 |
A service level of 95% means you expect to avoid a stockout in about 95 of every 100 replenishment cycles. It is not the share of units you can supply.
Worked example
A SKU sells 12 units a day on average, with a best day of 20 and a standard deviation of 4. The supplier’s lead time is 30 days on average and 36 on the slowest recent order.
- Simple: (20 × 36) − (12 × 30) = 720 − 360 = 360 units
- Statistical, 95%: 1.65 × 4 × √30 = 1.65 × 4 × 5.48 = 36.2, so 37 units rounded up
The results differ by almost a factor of ten, and both are correct for what they assume. The simple method stacks the worst day on the worst lead time for the whole wait. The statistical method covers normal swings in daily sales at your chosen service level, and assumes the lead time doesn’t vary.
Which number to use
- If a stockout is costly and the SKU earns well, lean toward the larger number, or raise the service level.
- If the SKU is slow, cheap to lose or easy to backorder, the smaller number saves cash.
- If your lead time varies a lot, add it to the statistical method. The standard version is Z × √((lead time × σ(daily sales)²) + (avg daily sales × σ(lead time))²), where σ(lead time) is the standard deviation of lead time in days.
Measure your lead time from past orders with the lead time calculator.
Put it to use
Safety stock feeds the reorder point. With 12 units a day, a 30-day lead time and 37 units of safety stock, the reorder point is 12 × 30 + 37 = 397 units. When available stock falls to that level, place the order.
Recalculate when your sales pattern or your supplier changes, and again before a season you know will be busier.
Frequently asked questions
What is the safety stock formula?
There are two common versions. The simple one is (maximum daily sales × maximum lead time) − (average daily sales × average lead time). The statistical one is Z × standard deviation of daily sales × √lead time, where Z comes from the service level you choose.
How do you calculate safety stock?
Pick a method and gather your numbers from recent history. For the simple method you need your average and highest daily sales and your average and longest lead times. For the statistical method you need the standard deviation of daily sales, the average lead time and a service level such as 95%.
How do I choose a service level?
A service level is how often you accept running out during a lead time. At 95% you expect to be covered in about 19 of every 20 replenishment cycles. Higher levels need more stock for each extra point, so aim higher for fast, profitable SKUs and lower for slow or low-margin ones.
Which safety stock method should I use?
Use the simple method when you have little history or want a cautious number quickly. Use the statistical method when you have at least a few months of daily sales and a fairly steady supplier. If your lead time also varies a lot, the statistical formula on its own will understate the stock you need.
What is Z in the safety stock formula?
Z is the number of standard deviations of demand you cover, set by the service level. The common values are 1.28 for 90%, 1.65 for 95%, 2.05 for 98% and 2.33 for 99%.