Free tool
Safety stock calculator
Enter average demand and lead time, how much each one varies, and a target service level to get a statistically sized safety stock and reorder point.
Your demand and lead time
The service level below is a cycle service level: the chance a single order cycle does not stock out before the next delivery arrives. It is not fill rate, which measures the share of units shipped on time instead of the share of cycles that stock out.
Estimated safety stock
106 units
- Z (service level)
- 1.64
- Demand-only safety stock
- 37 units
- Reorder point
- 386 units
This combines demand variability and lead-time variability (the King/ASCM formula). The demand-only figure ignores lead-time variability and understates the buffer whenever lead time is not perfectly reliable.
The formula
Z = NORM.S.INV(service level), the inverse standard normal CDF
Safety stock = Z × SQRT(lead time × demand SD^2 + demand^2 × lead-time SD^2)
Reorder point = average demand × average lead time + safety stock
- Z
- How many standard deviations of buffer the target cycle service level needs.
- Demand SD
- How much demand varies from one period to the next around its average.
- Lead time
- Average days between placing an order and receiving it.
- Lead-time SD
- How much the supplier's lead time varies from one order to the next.
- Cycle service level
- The probability a single order cycle does not stock out before the next delivery, not the share of units shipped on time (fill rate).
Worked example
- A store sells an average of 20 units a day (SD 6) from a supplier whose lead time averages 14 days (SD 3), and wants a 95% cycle service level, which gives Z = 1.64.
- Ignoring lead-time variability, demand-only safety stock is 1.64 × 6 × SQRT(14) = 37 units.
- The combined formula also accounts for lead-time variability: 1.64 × SQRT(14 × 6^2 + 20^2 × 3^2) = 106 units, nearly 3× the demand-only figure.
- Reorder point is 20 × 14 + 106 = 386 units.
When it applies
- Reorder-point systems where an order fires whenever stock falls to a trigger, not on a fixed calendar day.
- Items with demand and lead time you can estimate an average and a spread for, not just a single guess.
- Suppliers whose delivery times are not perfectly fixed, where the combined formula's lead-time term matters.
When it breaks down
- Extremely short product lifecycles with no demand history to compute a standard deviation from.
- Demand that is not roughly bell-shaped, such as intermittent or lumpy demand with many zero-demand periods; the normal-distribution assumption behind Z understates the tail risk there.
- A fixed review-cycle system (order every Monday, regardless of stock level), which calls for a par level or a min/max sized to the review period, not a reorder point.
Common mistakes
Reading service level as fill rate
A 95% cycle service level does not mean 95% of units ship on time. Fill rate is volume-weighted and reads higher for the same stock; treating the two as the same number understates how exposed a lower service level actually is.
Mixing periods
A weekly demand standard deviation paired with a lead time in days silently miscalculates the buffer. Convert everything to one period first. This calculator does it for you, but a spreadsheet copy usually will not.
Ignoring lead-time variability
The demand-only formula assumes a lead time that never moves. When a supplier's lead time actually varies, skipping that term understates the safety stock this page's own worked example by roughly three times.
Questions
What is safety stock and how do you calculate it?
Safety stock is the buffer held above expected demand over the lead time, sized to absorb variability in demand and delivery so a normal-range fluctuation does not turn into a stockout. The statistical method multiplies a Z-score for the target service level by the combined variability of demand and lead time: Z × SQRT(lead time × demand SD^2 + demand^2 × lead-time SD^2). A simpler rule of thumb, max demand × max lead time minus average demand × average lead time, needs no distribution but carries no stated service level.
What is a good safety stock level?
There is no universal figure. It depends on the service level the item needs to hit, how variable its demand and lead time actually are, and what a stockout costs versus what holding the extra unit costs. Set it per SKU from that item's own demand and lead-time variability rather than applying one buffer size across a whole catalog.
What is the difference between safety stock and reorder point?
Safety stock is the buffer itself. Reorder point is the trigger to place an order: average demand over the lead time, plus that buffer. When stock reaches the reorder point, it is time to order.
Can safety stock be negative?
Not as this formula computes it: the service level here runs from 50% to 99.9%, which keeps Z at zero or above, so Z × the variability term can only be zero or positive. A Z below zero would need a service level under 50%, which this calculator does not allow. On-hand stock dropping below the safety-stock line is normal, though: that is the buffer being drawn down during a cycle, not the safety-stock figure itself going negative.
Does safety stock affect EOQ?
No. EOQ is a separate trade-off between ordering cost and holding cost for cycle stock, and safety stock is not one of its inputs. Safety stock does raise average inventory and its holding cost, since it sits on top of cycle stock, but it does not change the order quantity EOQ recommends.
Related
Published by Skuvelo. Results are estimates computed from the figures you enter, not a reading of your own sales or stock.
A calculator answers once. Skuvelo keeps answering.
This tool computes one number from what you type. Skuvelo computes it continuously, for every SKU, from your own sales and stock.