The Real Economic Order Quantity
Traditional Economic Order Quantity
Open any Operations Management book or Operational Accounting book and you will see a figure, much like the one below (Fig. 1), used to describe what we have come to know as the Economic Order Quantity or EOQ. This is the amount of material (a part, component, sub-assembly) that should be ordered to balance the cost of material and the cost of holding the material necessary to meet the factory’s production schedule.
All major MRP or ERP software solutions have algorithms designed to estimate this quantity depending on your manufacturing schedule needs, corporate policies, lead times, and amount of risk you are willing to accept.
The problem is that a major factor in determining the “Ordering Costs” is usually ignored. This then means that the curve is wrong, and the equilibrium point (The EOQ) is also wrong.
The parameters commonly used to determine the Ordering Cost are the transactional costs of placing the order, receiving it, paying, etc. The higher the quantity you order, the less frequently you incur these costs. It also includes the cost of the component itself which may have some variation (usually small) based on the annual quantity that you buy.
For commodities like, bolts, nuts, springs, etc. this is a fine approximation. However, when you are buying engineered components, made specifically to fit your product’s design, this approximation can be far off the true cost.
How Batch Size Effects the EOQ of Custom Parts
One of the most significant factors in the actual cost of a custom made component is the batch size, which is directly tied to the order quantity itself.
Have you ever looked at catalogs that sell corporate apparel, like T-shirts or hats? You will notice that if you buy 100, the price each may be $12 each, but if you buy 250 then it is only $8 each, and if you buy 500 is $6.50, and then if you buy 1000 it is $5.75. Notice that there is a large difference at the lower quantities, but a small difference at the higher quantities. That is exactly what happens when you buy custom castings, plastic components, or even sheet metal stampings.
The problem is that your MRP or ERP system does not have a definition of how the cost of each component varies with its batch size, therefore the EOQ is erroneous.
The figure below (Fig. 2), shows how the cost of a real sheet metal part changes with batch size – regardless of the annual quantity. So, if you were buying 6,000 parts per year in weekly batches of 120, you would pay about $3.50 for each, but if you were buying in monthly quantities of 500, you would pay closer to $2.50 – almost 30% less – well worth the cost of carrying the inventory.
If you know which parts are being ordered in quantities that are inflating your costs, you have options to reduce those costs. For example, you can simply order a higher quantity less often or you can come to an arrangement with your supplier to manufacture the item in higher quantities and sell them to you in smaller ones (if you have inventory turns restrictions).
In the latter example, suppliers usually charge for their cost of carrying the inventory and will want you to take the risk of obsolescence in case the part is suddenly no longer needed. That risk is usually limited to the agreed upon batch size.
If you are already ordering in large quantities, where your cost per part has stabilized, then the behavior of EOQ is very much like commodities and your ERP or MRP system is producing good EOQ estimates.
But how do you know which parts are being ordered at sub-optimal quantities and which ones are not?
Finding the Real Economic Order Quantity
That is where a modern product cost estimating tool becomes very useful.
Using such a tool’s bulk costing capability, you can cost hundreds of similar parts at one time. When the costing process is complete, you can quickly plot the cost of setting-up the part for the various processes involved in manufacturing it, versus the cost of the part.
If the cost of set-up is more than 20 to 25% of the total cost of the part, you should address those “outlier” parts – see sample chart below (Fig. 3).
Continuing with your costing tool, estimate each outlier part at various batch sizes to create a plot similar to that of figure 2 (above), and determine the real curve describing Ordering Cost. With the most advanced product cost estimating software, you can run each outlier part, at multiple batch sizes in a single action.
Want to learn more?
Check out this short intro to aPriori video.