Main Points to Understand:

1. Understand the difference between Continuous Review and Periodic Review.

Continuous Review: We monitor inventory continuously and determine when to re-order based on our inventory position. An optimal replenishment strategy is a (Q, r) strategy: when inventory reaches a level r, the re-order point, place an order for quantity Q. The re-order point r should be determined by the appropriate level of safety stock and the expected lead time demand. The order quantity Q should be the EOQ. Note that in this setting we are assuming demand is relatively constant and we are balancing the costs of ordering that depend only on how many orders we place against the costs of inventory.

Periodic Review: We can only order at fixed intervals, e.g., every week or every month and determine how much to order based on our inventory position. An optimal replenishment strategy in this setting is an (s, S) strategy in which the quantity we order each time is either 0, if the inventory level is greater than s, or the amount required to bring total inventory up to level S. We did not discuss the "s" part of (s, S) inventory policies. Nor did we discuss strategies for finding optimal values of "s" and "S". In this setting, order costs are fixed and we are trying to control inventory costs.

Note that in the continuous review setting the order quantity is fixed, but the time between orders varies. In the Periodic Review setting, the time between orders is fixed and the order quantity varies.

2. Understand the impact of increasing order frequency in each setting.

In the continuous review setting, to increase order frequency we must reduce the order quantity. This will increase ordering costs and reduce inventory costs. Since safety stock in this setting depends on variability in lead time demand and increasing frequency of orders has no impact on lead time, increasing frequency of orders in this setting has no impact on the chances that we will stock out in an order cycle. It does however increase the number of order cycles in a fixed period, e.g., a year and so to ensure the same fill rate we need to carry more safety stock if we increase frequency.

In the periodic review setting also reduces inventory if we ignore safety stock. It is less clear what impact increasing frequency has on safety stock. In the periodic review setting the risk of stocking out in an order cycle depends on the variability of demand in a period determined by the sum of the time between orders and the lead time. So, increasing the frequency of orders reduces the time between orders and so reduces the variability of this demand. So, increasing the frequency of orders reduces the chances that we stock out in an order cycle if we hold safety stock constant. On the other hand, increasing the frequency of orders means we face this reduced risk more often. The final impact of increasing frequency will depend on the balance between these two factors. So, there are two trade-offs to evaluate:

The impact of increased frequency on safety stock: This involves the balance between

• the reduced risk of stock out faced with each order cycle
• the increased frequency with which we face this risk

The impact of increased frequency on total inventory: Even if increasing frequency increases safety stock, this increase may not be as large as the decrease in cycle stock. In our experience, increasing frequency typically reduces both safety stock and (as a natural consequence) total inventory.

3. Recognize that when lead times are long, the risks we face of stocking out from one order cycle to the next are dependent. The intervals in question overlap.

4. Recognize that we can only increase frequency without increasing "ordering costs" in certain settings. For example, if we order 300 ocean containers each week, then increasing the frequency and ordering 100 ocean containers three times each week will only have a minor impact on ordering costs. The bulk of the costs are the rates we pay per container. If, however, we receive 1 truck per week from each of three suppliers, increasing the frequency of deliveries from any one supplier will likely have a significant impact on "ordering costs" -- it will take more trucks to accomplish this. If we can arrange to mix the shipments from our three suppliers so that rather than getting one truckload per week filled with the goods from each supplier, we recieve three trucks per week each containing goods from all three suppliers, then we can increase the frequency with a smaller impact on the "ordering costs". In this case, the impact is reduced to the costs of collecting the three smaller loads into a single truck.