Safety Stock Calculator

How to use the calculator

1. Choose the period your demand figures are in: day, week or month.
2. Enter the average demand and its standard deviation for that period. If you do not know the standard deviation, open Calculate from my usage history and paste 8 to 52 recent usage figures.
3. Enter the average lead time in days, and the lead-time standard deviation if your deliveries arrive early or late.
4. Choose a service level. 95% is a common starting point.
5. Read the safety stock and the reorder point, then check the working below the result.

The calculator works for one item at a time. Use the figures for that item only, not the total for a product group.

What safety stock is

Safety stock is extra stock kept to cover demand that is higher than expected, or a delivery that arrives later than expected, while you wait for a replenishment order. Without it, any day of high usage during the lead time can cause a stockout. Too much of it ties up money and space, and for food and other dated products it increases the risk of stock expiring.

The reorder point is the stock level at which you place the next order. It is the demand you expect during the lead time plus the safety stock. This calculator gives both, so you do not need a separate reorder point calculator.

Safety stock formula

The statistical formula used by this calculator, when both demand and lead time vary, is:

When the lead time does not vary (σL = 0), the formula becomes:

All figures must use the same time period. The calculator asks for lead time in days and converts it for you: days ÷ 1 for daily demand, ÷ 7 for weekly demand, and ÷ 30 for monthly demand. The conversion is shown in the working, so the units cannot be mixed by accident.

The formula assumes demand and lead time vary independently of each other and roughly follow a normal distribution.

Reorder point formula

When the stock of the item falls to the reorder point, you place the next order. The safety stock is the part of the reorder point that is there to cover variation.

Service level and Z values

The service level here is the cycle service level: the probability of not running out of stock during one replenishment cycle, from placing an order until it arrives. It is not the same as the fill rate, which is the share of demand supplied from stock. A 95% cycle service level means that, on average, about 1 replenishment cycle in 20 is expected to run short.

Z grows faster as the service level approaches 100%, so each extra step costs more stock than the one before. For a custom service level between 50% and 99.9%, the calculator works out Z from the standard normal distribution.

Worked examples

An item is used at an average of 40 units a day, with a standard deviation of 12 units a day. The average lead time is 10 days, and the target service level is 95% (Z = 1.6449).

1. Lead time does not vary. Safety stock = 1.6449 × 12 × √10 = 62.4, rounded up to 63 units. Demand during lead time = 40 × 10 = 400 units. Reorder point = 400 + 63 = 463 units.
2. Lead time varies with a standard deviation of 2 days. Combined standard deviation = √(10 × 12² + 40² × 2²) = √(1,440 + 6,400) = 88.54 units. Safety stock = 1.6449 × 88.54 = 145.6, rounded up to 146 units. Reorder point = 400 + 146 = 546 units.
3. Simple max − average method, with a highest daily usage of 60 and a longest lead time of 14 days. Safety stock = (60 × 14) − (40 × 10) = 840 − 400 = 440 units. Reorder point = 400 + 440 = 840 units.

The second example shows how much an unreliable lead time adds: a variation of 2 days more than doubles the safety stock. The third shows why the simple method is only a rough guide. It uses the single worst day and the single latest delivery together, which is often far more cover than the target service level needs.

Simple max − average method

This method needs no standard deviation, which makes it easy to use. It has no service level, and one extreme day of usage or one very late delivery in your history raises the result for every order after it. Use it as a rough check, and prefer the statistical method when you have usage history.

Getting the inputs right

• Use consumption, not orders. Demand should be the quantity actually issued or used in each period, from your stock records.
• Use enough history. 8 to 52 periods is a practical range. Leave out periods that were not normal, such as a shutdown or a one-off project, unless you expect them to happen again.
• Measure the full lead time. Count from placing the order until the stock is available to use. If goods are held for inspection or quality release after arrival, decide whether your process counts that time and apply it consistently.
• Keep units consistent. Enter demand in the same unit of measure you count in, such as cartons, kg or bags.
• Recalculate regularly. Demand and supplier performance change. Review safety stock when usage patterns or suppliers change.

Assumptions and limits

• Demand and lead time are assumed to vary independently and to follow a roughly normal distribution. For slow-moving or very irregular items, the result is less reliable.
• Stock is assumed to be checked continuously, with an order placed as soon as it reaches the reorder point. If you only review stock on fixed days, more cover may be needed.
• The calculator does not work out how much to order, costs, or minimum order quantities.
• For perishable products, calculated safety stock should also be checked against shelf life, consumption rate, storage conditions, and your company’s inventory policy.

The result is a planning estimate. Check it against your own stock policy before you change reorder levels in your system.