Inventory is a highly visible asset and, in many companies, also the largest asset. In today’s highly competitive global economy, inventory has become the focus of improvement for many companies.
Inventory typically includes: cycle stock and safety stock.
Cycle stock provides the buffer between replenishment deliveries and is fairly straightforward.
Safety stock protects against variability in both demand and lead times.
Therefore, setting the correct levels of safety stock is tricky and requires a thorough understanding of its drivers.
In our engagements, we have seen many inventory management professionals develop interesting approaches to managing and reducing their safety stock.
Here, we will discuss some of the common pitfalls in those approaches.
Pitfall 1: Setting safety stock to zero will reduce inventory
In some of our recent engagements, we observed that some supply chain professionals reduce safety stock to zero.
They believe that they can reduce inventory if they set their safety stock to zero and in most cases, their inventory level did go down but so did their service level.
For most companies, this unwanted side-effect will end up costing them much more than the cost of some extra inventory.
Therefore, we must better understand the relationship between safety stock and service level to effectively reduce inventory without reducing the service level.
The most commonly used safety stock calculation is as follows.
The equation shows that the safety stock level is the multiplication of service level and variations in demand and lead time.
In order to have zero safety stock, the z-score of the service level will have to be zero or variations in demand and lead time have to be zero.
Assuming that average lead time, average demand, the standard deviation of the lead time and the standard deviation of the demand stay the same, any safety stock reduction will surely result in service level reduction.
Pitfall 2: A textbook safety stock formula works for my supply chain
One common mistake for many inventory practitioners is to use a textbook formula without fully understanding its range of applicability.
For example, the safety stock formula described above is the most commonly used formula in inventory management today. This model will calculate the necessary safety stock to achieve a target service level provided that both lead time and demand are normally distributed.
This model does not take into account the upstream failure rate, reorder period, or order quantity requirements which are part of cycle stock.
Therefore it is not recommended to use only this model if the upstream failure rate, reorder period, or order quantity requirements are significant business constraints in your supply chain.
Pitfall 3: Safety stock declines as average supplier lead time declines
Safety stock is designed to prevent stock-outs when there is variability in your demand and supply. Changes in your mean lead time and demand affect your cycle stock but not your safety stock. By reducing the variability, you reduce your safety stock.
Let us look at two cases to understand this. Assume that you have to provide a 95% service level for your customer, the demand for your product is 10 per day and forecast error is 2.
- Case 1: Supplier A has an average lead time of 15 days and the standard deviation is 10 days.
- Case 2: Supplier B has an average lead time of 24 days and the standard deviation is 1 day.
Assuming the receipt period equals the total lead time, which supplier will reduce your safety stock?
In both cases, you have to safeguard your inventory for 25 days, but the safety stock values are different for each of them. When you plug the values into the safety stock equation, Supplier B will have the lowest safety stock. Since you know that the lead time is 24 days, your cycle stock will be ordered in such a way to satisfy the demand for 24 days and you only need to safeguard your supply with a safety stock for 1 extra day which is the variability here.
But in case of a 15 day lead time with 10-day variability you are sure that the product will reach you at an average of 15 days, but it may take up to 25 days so you will order your cycle stock so that it satisfies 15 days demand and safety stock to safeguard for the extra 10 days variability.
So the variability of your lead-time increases your safety stock. The same principle also works in the case of demand. The variability of your demand increases your safety stock.
Let us look at an extreme case, if your supplier says he will have no lead time variability and your demand forecaster says he has predicted the demand perfectly, then you will have no use for the safety stock since you know what amount you need and how long it takes for your supplier to ship it to you. You will order the exact amount you need for each review period which is your cycle stock.
The bottom line is that you don’t need safety stock when there is no variability.
Pitfall 4: Safety stock eliminates stock-outs
Safety stock is designed to prevent the majority of the stock-outs, not all of them. You can design your safety stock to satisfy your customer service level but there will always be stock-outs.
One of the main variables when calculating safety stock is the service level. As shown in the graph, safety stock increases with the customer service level. When the service level values reach above 95% the safety stock number increase exponentially. Statistically speaking, the safety stock is infinite for a 100% service level.
The safety stock equation is designed to deal with variability. Variability means you cannot assure the lead-time of your suppliers and cannot forecast your demand perfectly. If you could do that, you wouldn’t need safety stock at all. With safety stock, sometimes you overshoot your inventory level and sometimes you fall below your predictions. The first case leads to stock-outs and the second case leads to excess inventory.
We can argue that in the second case we will never experience stock-outs and have 100% customer service level. But it doesn’t mean the safety stock was calculated for this scenario. The second case occurs because of the variability. It costs you more when you have inventory on hand which has no demand.
To sustain this, we should always have a balance between cost and service level. We can slowly improve our service level by decreasing our variability but with the limited practical resources, we will always have stockouts at some point in time.
The bottom line is stockouts are inevitable but we can keep them to a minimum by reducing variability.
Variability along the supply chain has a large impact on inventory requirements that is often not realized. Reducing inventory often requires understanding which drivers are the most important in your supply chain and alleviating them in order to improve overall performance.
Source: OPS Rules Blog: Insights into Supply Chain and Operations Strategy
Related: Principals of Inventory Management