What is Safety Stock?
In inventory management safety stock is an established amount of additional inventory that will be maintained to mitigate the risk of a stock out; it is an inventory buffer in the event sales experience an unplanned increase, or if a supplier takes longer than expected to deliver an order. While carrying this additional inventory comes with a cost, the cost of carrying inventory may be worth it when the alternative is a potential lost sale.
Safety Stock Formula
There are three ways to calculate safety stock.
- Basic This is the easiest calculation and you simply multiply average lead time by average demand. This method doesn't take service levels into consideration.
- Standard The standard approach seeks to address the risks associated with a stock out. This is done by taking into consideration a service level (also referred to as a service factor or the fill-rate).
- Complex This third method uses a factor (which is to represent Norm's Inverse) while also taking into consideration a fill-rate.
SSL = avg LT * avg D
avg LT = average lead time
avg D = average demand
SSL = Z × σLT × avg Dwhere:
Z is the desired service level
σLT is the standard deviation of lead time
D avg is demand average
SSL = [[(LT × σD^2) + (D avg × .92^2)]^1/2] × Zwhere:
LT is the average Lead Time from order placement to receipt
σD is Demand Standard Deviation
avg D is demand average
.92 is squared and allows us to achieve a similar result to what we would get if we used Norm's Inverse
Z is the desired service level (also known as fill rate)
To use fill-rate (or service level) you would refer to z-scores and pending the percentage desired for the fill-rate there is an equivalent z-score, as follows:
99.9% fill-rate = 3.49 z-score
99% fill-rate = 2.33 z-score
97% fill-rate = 1.88 z-score
95% fill-rate = 1.65 z-score
91% fill-rate = 1.34 z-score
How to Calculate Safety Stock
- Determine Average Lead Time
- Determine the Average Demand
- Set Your Service Level
- Put the Pieces Together
Lead time is the time it takes to get a product from when it is ordered to the time it is received. Many businesses will uses 12-36 months to calculate average lead time to ensure there are enough orders to accurately represent the average lead time; if a business only received one order from a supplier that order may not be an accurate representation of the suppliers lead time (you will want multiple orders). If you are using the "standard" safety stock formula, you will want the standard deviation of the lead times to capture variability.
Average demand is the average number of units sold during a period of time; this could be weekly, monthly, quarterly, or annually. For products that have a fast turn over (for example goods that are perishable) you may want to review weekly sales, but for products that sell irregularly you may want to base demand on a monthly basis. Simliar to the lead time, you will want to look at demand over a 12-36 month period to ensure there are enough data points that accurately represent demand. Typically a business will review monthly sales to set their safety stock level, which is only part of the forecasting and inventory decision making process.
While service level is important, keep in mind that there is a cost associated with it. The higher the service level, the more inventory the model is going to tell you to buy. If you're not familiar with how the model works, play with it and make adjustments to your service level expectations to see how the results are impacted. At some point it may be a good idea to do a cost-benefit analysis; analyze the cost of the additional stock compared with the cost of a lost sale.
While this is an arbitrary example, if you set a 99.9% service level because you want to guarantee you will never have a stock-out for the item, the model may say buy 100 more than you need for safety stock. But, if you only average selling 100 units a month for the particular product it may not make sense to carry so much additional inventory.
Once you know the standard deviation of your lead time, the average demand, and have set a service level, you can plug the numbers into the formula to set your stocking levels. Often this is done with multiple products, sometimes thousands or tens-of-thousands.