In most of inventory management training programs, a lot of time is spent explaining how to guarantee a certain availability of a product. Standard inventory management theory can help us calculate the safety stock for a product taking into account the accepted stock out probability. But do these formulas also apply to short shelf life products as well? In those situations the safety stock can cause obsoletes if the stock level is too high compared to the accepted shelf life.
Using similar inventory management theory techniques as to ensure the product availability, it is possible to calculate stock levels that balance both availability and shelf life.
There is an obvious contradiction between inventory and product freshness. Manufacturers of fresh products like fresh produce, dairy etc. determine the production batch size usually based on tomorrow’s expected demand. Make to stock for these products is not an option since the customer will not experience the product as fresh if it has been on stock for several days. However, when calculating the production batch size, implicitly an inventory management model is used. Assuming a daily production, every day a production batch size is determined to ensure product availability until the next production moment. Because tomorrow’s expected demand cannot be predicted 100% accurate, some safety margin will be added to ensure availability. Unsold products will remain in stock for sales on the next day.
The production batch size is calculated as the difference between the expected demand plus a safety margin, minus the balance on hand, if any. In fact this safety margin is the same as the safety stock in a make to stock situation and can be calculated using the standard safety stock formulas.
The formulas for calculating the safety stock are based on restricting the probability of an out of stock situation and thus missed sales to a predefined maximum. For instance assume that a 2% stock out probability is accepted. This implies that 98% of time stock must be available. In other words, even if market demand is higher than expected, there must be enough products on stock. If market demand is lower than expected, there is no problem in availability. In this situation no safety stock was needed.
However, lower than expected demand is an issue when selling (daily) fresh products because in that situation more products will be left over than expected. These products will be sold the next day and therefore are one day less fresh. What can be done if customers require a really fresh product and only accept a freshness of the day before for a certain volume of the products?
The following analogy can be used: if we can add a safety margin to the expected demand during lead time to guarantee availability, is it then also possible to maximize the probability on obsoletes by subtracting a safety margin from the expected demand during shelf life? To calculate this safety margin we can use the same formula as we use to calculate the safety stock.
In conclusion, there are two different ways to calculate the desired stock level:
Rarely both calculated stock levels will be exactly the same. As long as the maximum stock level is higher than the minimum stock level, there is no problem. We can choose a stock level somewhere between the minimum stock and the maximum stock, without any risk of too little availability or too many obsoletes. In this situation we can aim for a stock level that serves both targets on availability and on freshness.
But what if this is not the case? What if this maximum stock level is lower than this minimum stock level? Then a study is needed to determine whether it is possible to move both stock levels towards each other, for instance by reducing the uncertainty in expected demand, by reducing the lead time or by extending the shelf life. If none of these options are possible, or its effects are too small, then a choice has to be made: do we make a concession on availability or do we accept more obsoletes? By calculating both stock levels as described above, a real fact based decision can be taken.