The role of safety stock (a.k.a., "minimum inventory levels") and its relationship with demand is commonly misunderstood. A popular intuition is to increase safety stocks whenever an increase in demand is anticipated. However, this misses the point of what safety stock levels should be designed to do.  The use of average expected rates of demand to calculate required safety stock levels may lead to unnecessarily high inventory levels. To make matters worse, safety stock parameters are often ratcheted back down again when it is determined that company is now suffering from excessive inventory costs. Management attention then moves cyclically between inventory shortages and overstock due to this reactionary approach to inventory management.

The challenge with planning inventory is to strike the right balance between not having too much and not having too little. This is true whether you are a distributor whose primary objective is stocking product to fulfill customer orders or a manufacturer who must maintain adequate on-hand raw materials.

In all cases, achieving reliably high fill rates with the minimum amount of inventory investment is the goal.

Safety stock can be looked at as insurance that protects the company against demand variability (read as "fluctuations" or "spikes"). The key characteristic of demand variability is uncertainty.  Basing the calculation of safety stocks on average demand does not do an adequate job of characterizing the uncertainty in demand.  Statistical approaches are therefore needed that seek to measure demand uncertainty and variability to effectively calculate safety stock levels.

To illustrate the role of demand in setting safety stock levels, consider the following two examples in which we contrast two different scenarios.

In the first case, we imagine planning inventory levels for bar soap. This is an example of a product that is characterized by high demand volume and low variability. We say low variability because the sales of soap tends to remain fairly constant from month to month with little or no seasonal variation. We can then afford to set safety stocks relatively low because our statistical analysis will reveal low uncertainty in the level of demand. This is the opposite of what many material managers expect. We will therefore see inventory oscillating widely between minimum and maximum planned stock levels as shown in the graphic below to support the high levels of demand.

The second case is the converse: It may, for example, be a medical product that is characterized by low sales volume and high variability. Our statistical approach will in this case yield a relatively high safety stock level to guard against demand spikes. Inventory, in this case, should oscillate within a narrower Min/Max control band.

Don't automatically assume that future demand spikes can't be effectively predicted.  To do so is to allow over-conservatism to creep into setting safety stock levels. We've seen numerous situations where inventory reductions as high as 80% are possible while still maintaining fill rates of 95% or more. What's more is that projected inventory savings and fill rate performance can often be reliably estimated using actual sales or usage history to simulate future performance. This allows the operating manager to confidently re-plan his inventory with a lower cost and higher performance.

Altemir Consulting provides inventory and safety stock management consulting to optimize your material cost performance.