While evaluating the fit and benefits for an IBP-Inventory (IBP-I) implementation at customers I am often asked about Economic Order Quantity (EOQ) and whether IBP-I can support it. In general the obvious answer is YES, IBP-I can certainly calculate an EOQ and consider it while generating invenory targets. But the real life implications of using both are a bit more complicated. The purpose of this blog is to explore this concept further and hopefully open a discussion on the nuances of using EOQ within IBP-I.
APICS defines EOQ as “a type of fixed order quantity model that determines the amount of an item to be purchased or manufactured at one time.” The basic premise is to balance costs on a a production or ordering schedule where small frequent batches incur a fixed order or changeover cost and large batches have excess inventory carry costs. Borrowing from APICS again we see the basic EOQ formula to be:
Q* = Economic Order Quantity
D = Annual Demand
K = Ordering Cost (Or changeover / setup costs)
h = inventory holding cost per unit
This formula is great due to its simplicity and ease of use. It does however make some simplifying assumptions about a known static demand that make it less than optimal for real world situations. There are methods for expanding upon this basic EOQ formula to account for these assumptions but that is not in scope of this discussion.
We can easily configure this formula in a calculated key figure in the standard IBP configuration capabilities. The discussion on the right sources for ordering costs and holding rate will vary by industry and company, but that should be part of any standard initial blueprinting project. ***Side comment here: In my experiences it is critical to ensure that the business side of the company is involved and makes the final decisions for inputs. IT can support this process, but getting the planners to adopt the final outputs require buy-in on the input side.
Once we have an EOQ calculated, the question is what impact will it have on IBP-I? The goal of IBP-I in an end to end multi-stage optimization is to minimize the inventory cost across the entire supply chain. Won’t this immediately override all the work that the EOQ has done in the balancing act? The answer is no. EOQ is considering inventory at a single location with a single cost. However, we know that in the real world the cost of inventory increases as it flows through the supply chain. This is due to reasons such as value add, transportation costs and less flexibility in demand it can fulfill (i.e. opposite of postponement). Therefore, IBP-I has a much more comprehensive view of the inventory challenge. Things like differing costs, supply uncertainy and forecast error are just a couple of the examples of complexity that IBP-I was designed to handle.
Does that mean an EOQ is worthless within IBP-I? The answer again is no. IBP-I certainly needs to take into account the manufacturing or ordering costs as it does the inventory cost minimization. To account for this, I’ve recommended to customers that EOQ should be input to IBP-I as a minimum lot size. IBP-I will then take these minimum orders as a constraint and make sure to recommend an inventory target that takes the ordering / setup into consideration.
To summarize, EOQ and IBP-I certainly have some overlapping capabilities and functionalitites but that doesn’t mean one is right and the other is wrong. It is definitely possible to use both concepts in IBP-I and get value from each of them.
What are your thoughts on using EOQ in IBP-I and/or inventory planning in general? Is there enough room in the inventory planning process for competing algorithms?