Safety stock and reorder point calculator
Choose how much stockout risk you accept and see the safety stock and reorder point that account for variability in both demand and lead time. The formula and its assumptions are explained below.
Economic order quantity (optional)
If you fill in all three fields, the EOQ, orders per year and an approximate fill rate are calculated.
The calculation runs in your browser; the values you enter are not sent anywhere.
- Reorder point
- 680.2
- Safety stock
- 180.2
- Safety stock (in days of average demand)
- 1.8 days
- Demand during lead time (mean)
- 500
- Standard deviation of demand during lead time
- 109.54
- z (for the service level)
- 1.645
Where does the safety stock come from?
- Demand variability only73.6
- Lead-time variability only164.5
- Safety stock180.2
The combined safety stock is found from the square root of the sum of squares, not the sum of the two parts, so it is smaller than adding them.
If the service level changes
| Service level | z | Safety stock | Reorder point |
|---|---|---|---|
| 90% | 1.28 | 140.4 | 640.4 |
| 95% | 1.64 | 180.2 | 680.2 |
| 97.5% | 1.96 | 214.7 | 714.7 |
| 99% | 2.33 | 254.8 | 754.8 |
| 99.9% | 3.09 | 338.5 | 838.5 |
σ_DL = √(L·σ_d² + d²·σ_L²) · SS = z·σ_DL · ROP = d·L + SS · EOQ = √(2·D·S / H)
The results assume demand is independent from day to day and normally distributed, lead time is independent of demand, and a continuous-review policy. With intermittent or skewed demand, seasonality and tightly constrained supply, the real need may differ; the results alone are not sufficient for a decision.
Shall we review your stock policy together with your actual consumption data, by item and service level?
Request a call01
How to use it
A
Enter the mean and standard deviation of daily demand and of lead time (enter 0 for the deviation if lead time is fixed).
B
Choose the cycle service level; safety stock and the reorder point are calculated instantly, and the table compares other service levels.
C
Optionally enter order cost, unit cost and holding rate to see the EOQ, orders per year and an approximate fill rate.
02
Formula and assumptions
Safety stock is the buffer held against uncertainty in demand during the lead time. When demand and lead time both fluctuate, the standard deviation of demand over the lead time is σ_DL = √(L·σ_d² + d²·σ_L²). Here d is average daily demand, σ_d the standard deviation of daily demand, L the average lead time (days) and σ_L the standard deviation of lead time. Safety stock is SS = z·σ_DL and the reorder point is ROP = d·L + SS; z is the standard-normal value for the chosen service level (1.645 for 95%).
The formula rests on four assumptions: daily demands are independent and normally distributed; lead time is independent of demand; lead time is normally distributed around a fixed mean; and stock is monitored continuously, with an order placed when it falls to the ROP. With intermittent demand (zero on most days), strong seasonality or lead time that correlates with demand, these assumptions break down.
The service level here is the cycle service level: the probability of no stockout during an order cycle. It is not the same as the fill rate, which shows what share of demand is met from the shelf. At the same cycle service level the fill rate rises as the order quantity grows; when an EOQ is entered, the tool also gives an approximate fill rate (1 − σ_DL·G(z)/Q, where G is the standard normal loss function).
03
Is demand or lead time the driver?
Look at which component the safety stock comes from. Demand variability alone needs z·σ_d·√L, lead-time variability alone z·d·σ_L. For an item with high average demand and an erratic lead time, the second term dominates; fixing the supplier's lead time can then be far cheaper than enlarging the safety stock.
As lead time grows, the demand-driven part of safety stock grows with its square root, whereas variation in lead time acts linearly, multiplied by average demand. That is why measuring the standard deviation of lead time is more valuable than knowing only the mean.
Raising the service level increases safety stock at a growing rate, not linearly: going from 95% to 99% moves z from 1.645 to 2.326. The table sets the stock cost of different service levels side by side to help you decide which level is reasonable for which item; most businesses assign higher levels to important items and lower ones to low-value items (ABC classification is used for this).
04
What does the EOQ do, and what does it not do?
The economic order quantity EOQ = √(2·D·S / H) is the order size that minimises the sum of ordering and holding costs. D is annual demand, S the fixed cost per order and H the annual holding cost of one unit (unit cost × holding rate). Order quantity and reorder point complement each other: ROP says when to order, EOQ says how much.
The EOQ assumes constant demand, fixed costs and no price breaks; the resulting cost curve is quite flat around Q*, so deviating slightly from the EOQ does not raise cost much. Rounding to practical constraints such as pallet or carton size or the supplier's minimum order quantity is usually fine.
FAQ
- Are the numbers I enter sent anywhere?
- No. All calculations run in code inside your browser; no value is transmitted to a server or stored.
- How do I find the standard deviation of demand?
- Calculate the standard deviation from recent daily demand data (at least a few weeks). If you keep demand weekly or monthly, convert it to daily or use square-root time scaling (weekly deviation ÷ √7). This tool does not read raw data; it asks for the summary numbers.
- What do I enter if the lead time is fixed?
- Enter 0 for the standard deviation of lead time; the formula reduces to demand variability only (SS = z·σ_d·√L). Lead time is rarely truly fixed, though, so measure it from past deliveries.
- Which service level should I choose?
- It depends on the cost of a stockout. 98–99% for critical, high-value items and 90–95% for ordinary items are commonly used examples, but the right value varies by organisation. Compare the stock-cost difference in the table with the cost of a stockout.
- Why is safety stock not the sum of the components?
- Demand and lead-time fluctuations are assumed independent, so variances add, not standard deviations. The combined safety stock is therefore smaller than the sum of the two parts.
- Is this calculation suitable for intermittent demand (zero on most days)?
- Usually not. The normality assumption breaks down and safety stock comes out too high or too low. Croston-type methods or empirical approaches based on the consumption distribution suit intermittent demand better.
Let's build your stock policy with your data
We derive demand and lead-time distributions by item and prepare safety stock and reorder point recommendations consistent with your service level targets. Let's talk about your data in a free discovery call.