Logistics & TransportPopular

Safety Stock Calculator

Calculate safety stock using the statistical service-level method with demand variability, lead time variability, and Z-score for service levels from 90% to 99.9%. Shows reorder point and the stockout risk for each service level target.

Target Service Level

Standard items, most manufacturing.

Avg Daily Demand (units)

Demand Std Dev (units/day)

Standard deviation of daily demand.

Lead Time (days)

Lead Time Std Dev (days)

Standard deviation of supplier lead time.

Unit Cost ($/unit, optional)

Used to show inventory value and carrying cost.

Your Results

Enter demand and lead time data, then click calculate.

Embed This Calculator on Your Website

Add this free calculator to your blog, website, or CMS with a simple copy-paste embed code.

Introduction

Safety stock is the buffer inventory that absorbs demand spikes and supplier delays so you do not stock out. The statistical method, taught in every APICS and ASCM course and documented in the MIT King safety stock paper, sizes the buffer from the standard deviation of demand, the standard deviation of lead time, and a Z-score tied to your target service level. The math is unforgiving: moving from a 95% to a 99% service level raises required safety stock by about 41%, and from 99% to 99.9% adds another 33%. Most buyers set safety stock by gut or by a fixed days-of-supply rule, which is why the U.S. retail food industry loses $15 to $20 billion a year to stockouts. This calculator applies the full statistical formula, including both demand variability and lead time variability, so the buffer matches the actual risk profile of the SKU.

What This Calculator Does

This tool calculates safety stock and reorder point using the full statistical service-level method. You enter your target service level (90% to 99.9%), average daily demand, standard deviation of daily demand, average lead time in days, standard deviation of lead time, and optional unit cost. The calculator computes safety stock using the formula that accounts for both demand variability and lead time variability, the reorder point, the relative contribution of demand versus lead time variability, and the inventory value and carrying cost of the safety stock.

The Formula

Safety Stock = Z x sqrt(L x sigmaD^2 + d^2 x sigmaL^2) | Reorder Point = (Avg Daily Demand x Lead Time) + Safety Stock

The Z-score is the standard normal value for the target service level: 1.28 for 90%, 1.65 for 95%, 1.96 for 97.5%, 2.33 for 99%, 2.58 for 99.5%, and 3.09 for 99.9%. The sigmaD is the standard deviation of daily demand, measuring how much day-to-day demand fluctuates. The L is the average lead time in days. The sigmaL is the standard deviation of lead time, measuring how much supplier delivery time fluctuates. The d is the average daily demand. The term under the square root combines demand variance over the lead time (L x sigmaD^2) with lead time variance scaled by demand (d^2 x sigmaL^2). The reorder point is the demand expected during lead time plus the safety stock.

Step-by-Step Example

Select the target service level

A buyer of a critical SKU selects 95% (Z=1.65), the standard level for most manufacturing items.

Enter demand data

The SKU averages 40 units per day with a standard deviation of 10 units per day, calculated from the past 90 days of sales.

Enter lead time data

The supplier averages 14 days lead time with a standard deviation of 3 days, calculated from the past 20 purchase orders.

Review the results

Demand component: 14 x 10^2 = 1,400. Lead time component: 40^2 x 3^2 = 14,400. Under root: 15,800. Safety stock: 1.65 x sqrt(15,800) = 207 units. Reorder point: (40 x 14) + 207 = 767 units. Lead time variability dominates at 91% of the variance, so supplier reliability work will cut safety stock more than forecasting work.

Real-World Use Cases

New SKU Safety Stock Setting

A buyer launching a new product with no sales history uses the parent category demand standard deviation and the supplier known lead time standard deviation to set the initial safety stock before real demand data accumulates.

Service Level Trade-off Analysis

A supply chain manager models the same SKU at 95%, 99%, and 99.9% service levels to quantify the inventory carrying cost of each step up in service, then picks the level where the marginal carrying cost equals the marginal stockout cost.

Supplier Reliability Justification

A buyer comparing two suppliers with the same price but different lead time standard deviations (2 days vs 5 days) calculates the safety stock difference to justify paying a 3% premium to the more reliable supplier.

Import Lane Buffer Sizing

A DTC brand importing from China on a 35-day ocean lane with 7-day lead time standard deviation sizes safety stock for the lead-time-variability-dominated case, where freight discipline beats forecasting accuracy.

Common Mistakes to Avoid

Using the basic formula and ignoring lead time variability. The basic formula SS = Z x sigmaD x sqrt(L) assumes lead time is constant. For most real suppliers it is not. The full formula SS = Z x sqrt(L x sigmaD^2 + d^2 x sigmaL^2) accounts for both. On long import lanes, the lead time term dominates and the basic formula understates safety stock by 30% to 50%.
Setting service level by gut. A 95% service level means 1 stockout per 20 replenishment cycles. A 99% level means 1 per 100. The difference in safety stock is about 41%. Picking the level without quantifying the stockout cost leads to either overstock or lost sales. Calculate the stockout cost first, then pick the level where marginal carrying cost equals marginal stockout cost.
Using a fixed days-of-supply rule. A rule like keep 14 days of safety stock ignores both demand variability and lead time variability. Two SKUs with the same 14-day rule but different demand volatility and supplier reliability need very different buffers. The statistical method sizes each SKU to its actual risk profile.
Estimating standard deviation from too few data points. Standard deviation from 5 data points is unreliable. Use at least 30 days of demand data and 20 purchase orders for lead time. With fewer points, widen the standard deviation or use the parent category standard deviation as a proxy.
Not recalculating when supplier or demand changes. Safety stock is not a set-and-forget number. When a supplier improves or degrades reliability, or when demand volatility shifts with seasonality or a new channel, recalculate. A safety stock set 18 months ago is almost certainly wrong today.

Frequently Asked Questions

What is the formula for safety stock with lead time variability?

The full formula is Safety Stock = Z x sqrt(L x sigmaD^2 + d^2 x sigmaL^2), where Z is the service level Z-score, L is average lead time, sigmaD is the standard deviation of daily demand, d is average daily demand, and sigmaL is the standard deviation of lead time. The basic formula SS = Z x sigmaD x sqrt(L) ignores lead time variability and understates the buffer when suppliers are unreliable.

What Z-score should I use for a 95% service level?

Use Z = 1.65 for a 95% service level. Other common values: 1.28 for 90%, 1.96 for 97.5%, 2.33 for 99%, 2.58 for 99.5%, and 3.09 for 99.9%. Moving from 95% to 99% raises required safety stock by about 41%, and from 99% to 99.9% adds another 33%.

How do I calculate the standard deviation of demand and lead time?

Use at least 30 days of daily demand data and 20 historical purchase orders for lead time. In Excel, use STDEV.S on the daily demand column for sigmaD and on the lead time column for sigmaL. With fewer than 30 data points the estimate is unreliable, so widen the standard deviation or use the parent category standard deviation as a proxy.

What is the reorder point formula?

Reorder Point = (Average Daily Demand x Lead Time) + Safety Stock. This is the on-hand inventory level at which you place a new order. The first term is the demand expected to arrive during the lead time, and the second term is the safety stock buffer for variability.

How much does lead time variability affect safety stock?

On long import lanes, lead time variability often dominates the safety stock calculation. For a China-importing DTC brand on a 35-day ocean lane with 7-day lead time standard deviation, the lead-time term can be 30 to 40 times the demand term. In that case, improving supplier reliability or switching to a faster lane cuts safety stock far more than improving forecast accuracy.

Accuracy and Disclaimer

This calculator applies the standard statistical safety stock formula using the normal distribution. The formula assumes demand and lead time are approximately normally distributed, which holds for most steady-state SKUs but not for highly seasonal or intermittent demand. Z-scores and service level relationships reflect standard APICS and ASCM methodology. Actual stockout risk depends on demand distribution, lead time distribution, and order cycle frequency. This is not inventory management advice. Consult a supply chain professional for guidance on a specific SKU or network.

Conclusion

Safety stock sized by the statistical method protects against the two risks that actually cause stockouts: demand spikes and supplier delays. Run the calculation here, then check which variability component dominates, since that tells you where to focus improvement. If lead time variability dominates, supplier reliability work beats forecasting accuracy work. Pair this with our Supplier Lead Time Cost Calculator to quantify the dollar cost of that lead time variability, or the Total Cost of Ownership Calculator to weigh the carrying cost of safety stock against the cost of a stockout.