Skip to content
Calc.

Safety Stock Calculator

Safety stock two ways: the max-minus-average method (worst-case demand and lead time versus the averages) and the statistical formula z × √(LT × σ_D² + D² × σ_LT²) for a chosen service level. Shows both, the lead-time demand and the reorder point they imply.

Last updated: October 2026

Safety stock

380 units

Formula below · 2 sources (ascm.org, ocw.mit.edu) · Updated Oct 2026

Compare with similar

About this calculator

Safety stock is the inventory held above the expected demand during lead time to absorb variability in demand and in supplier lead time; without it every above-average day or late delivery becomes a stockout. The calculator offers the two methods in common use. Max minus average: safety stock = (maximum daily demand × maximum lead time) − (average daily demand × average lead time). The first term is the worst case you want to survive, the second the normal pipeline, and the difference is the buffer. It needs little data and is easy to explain, but it covers the compound worst case (peak demand through the longest lead time), which rarely happens, so it tends to over-stock; using the 95th-percentile demand and lead time instead of the all-time maxima tempers it. Statistical: safety stock = z × √(LT × σ_D² + D² × σ_LT²), where LT and D are the average lead time (days) and average daily demand, σ_D is the standard deviation of daily demand, σ_LT the standard deviation of lead time in days, and z the standard normal value for the target cycle service level (1.645 for 95%, 2.326 for 99%). The square root combines the two sources of variability into the standard deviation of demand during lead time; if lead time is constant, set σ_LT to 0 and the formula reduces to z × σ_D × √LT, and if demand is constant it reduces to z × D × σ_LT. The service level here is the probability of not running out during a replenishment cycle, not the fill rate, which is the share of demand met from stock and is higher for the same z. Both methods assume demand and lead time are independent and reasonably stable; seasonal items need the figures recalculated by season. The reorder point is the lead-time demand plus the safety stock, shown as a tile, and safety stock should be set per item: higher service levels for A-class items where a stockout is expensive, lower for C-class items where the carrying cost outweighs the risk.

How to use

Example 1 — max minus average (the defaults): average demand 60 units/day, average lead time 7 days, maximum demand 80 units/day, maximum lead time 10 days. Worst case = 80 × 10 = 800 units; normal pipeline = 60 × 7 = 420 units; safety stock = 800 − 420 = 380 units, covering 6.3 days of average demand. Reorder point = 420 + 380 = 800 units. Example 2 — statistical at a 95% service level with the same item: σ_D = 12 units/day, σ_LT = 1.5 days, z = 1.645. Demand variance over the lead time = 7 × 12² = 1,008; lead-time variance effect = 60² × 1.5² = 8,100; √(1,008 + 8,100) = √9,108 = 95.44 units; safety stock = 1.645 × 95.44 = 157.0 units, and the reorder point 420 + 157 = 577 units. Lead-time variability contributes 89% of the variance here, so a more reliable supplier would cut the buffer far more than smoother demand would. At 99% (z = 2.326) the figure rises to 222 units; with a constant lead time (σ_LT = 0) it falls to 1.645 × 12 × √7 = 52 units.

Frequently asked questions

What is the safety stock formula?

The statistical formula is safety stock = z × √(LT × σ_D² + D² × σ_LT²): z for the service level (1.645 at 95%), LT and D the average lead time and daily demand, σ_D and σ_LT their standard deviations. The simpler max-minus-average formula is (max daily demand × max lead time) − (average daily demand × average lead time).

When should I use max-minus-average rather than the statistical method?

Max-minus-average when you have little history (under about three months), lumpy or non-normal demand, or need a figure that is easy to explain and deliberately conservative. The statistical method when you have at least a few months of daily demand and lead-time records; it gives a smaller buffer for the same protection and lets you choose the service level per item. Many teams start with the first and move to the second as data builds up.

What service level should I choose?

The service level is the probability of getting through a replenishment cycle without a stockout. Common choices are 95–99% for A-class items (high value or high stockout cost), 90–95% for B-class and 85–90% for C-class. Each step up costs more: going from 95% to 99% raises z from 1.645 to 2.326, 41% more safety stock, for 4 points of service.

What is the difference between safety stock and reorder point?

Safety stock is the buffer quantity. The reorder point is the inventory level that triggers a new order: expected demand during lead time plus the safety stock. With 420 units of lead-time demand and 380 units of safety stock the reorder point is 800 units. The reorder point calculator works through that step.

Why does lead-time variability matter so much?

In the statistical formula lead-time variability is multiplied by the average demand squared. In the example, a 1.5-day lead-time standard deviation contributes 8,100 to the variance while demand variability contributes 1,008, so the supplier's unreliability drives 89% of the buffer. Shortening or stabilising lead times often reduces safety stock more than better forecasting does.

Where do the maximum demand and lead time come from?

From history: the highest daily demand and the longest lead time you are prepared to cover. Using the single worst value from years of data gives a very conservative buffer; the 95th percentile of daily demand and of lead time is a common compromise that drops the one-off outliers while still covering most bad weeks.

Related calculators

Sources & references

How we check our calculators

Think a result is wrong? Report a problem with this calculator.