Volume, volume, volume!
I recently participated in an online discussion on LinkedIn as to the impact of volume on cost. To be honest, this is one of my favorites, although controversial topics.
A LinkedIn user was listing all the typical cost drivers for a product, material, overhead, burden profit and of course, volume. I took and still take exception to volume being a cost driver. It often becomes a price driver, but I will get into that shortly.
The first thing that needs to be understood is, are you looking at cost from the perspective of a seller or buyer? There is a significant difference.
Assume your perspective is that of a seller. You must make every effort to capture your true cost, including all inefficiencies. You probably are attempting to cover your costs based on booked or proposed business with some level of confidence that it will come to fruition. In essence, spreading your cost over known business. For example, if your capital is 90% utilized it still needs to cover 100% of the costs. The same can be said if it’s 75% utilized. Still further 50% utilized and so on. A business needs to cover its expenses or it runs the risk of going out of business, correct? Well, there is a flipside to this philosophy. You may be pricing yourself out of business. If your competition has the philosophy that their sales department can and should fill the remaining under-utilized capacity, they can price the product based on time used. So, if the customer is utilizing 50% of the capital, that customer only pays for 50% of it versus the 100%. This in turn will reduce the selling price accordingly. There are pitfalls to this philosophy as well. What happens if you never sell the excess capacity? You’ll go out of business, or at least fire the sales department.
Now let’s assume the perspective of the buyer. How do you expect or model your products to be priced if your volume is only enough volume to fill a portion of a capita? The answer, it depends. If you are asking a supplier to purchase capital that for one reason or another cannot be used for any other product, ever, then you probably should pay for it all. We define this as dedicated capital. I would argue that if this is the case, then you should write your contract to reflect that you own that capital. This could provide you tax benefits and you could recoup the residual cost at the end of life. If you as a buyer are not asking for dedicated capital, I would argue that you only pay for the time you are utilizing said capital, up to a realistic point. If you are using 10% of the capital, you should only pay 10% of the costs. If you are using 50% of the capital, you should only pay for 50% of the capital. As you approach 100% you may end up paying for 100%. It would be unrealistic to expect a supplier to sell a small amount of time remaining.
If you understand and agree with this argument thus far, what additional cost is there for a lower volume from a buyer’s perspective? Let’s leave material out of the discussion for the time being. The only other buckets that could be affected by volume would be changeover or downtime, and overhead and profit. Changeovers should be minimized. Lean suggests they should be completed in a matter of minutes. Even if they trend towards hours, how much does it actually contribute to the cost?
To understand how much it actually contributes to the cost, we must first understand how to calculate it. A simple version would be it’s the lost time for the machine and the labor to perform the necessary setup. This cost will then be spread over the total number of parts produced in the next batch. Obviously, if you produce 1 part, it has to cover all the cost. If you produce 10 parts, each part receives 10% of the changeover cost. If you produce 100 parts each will see a contribution of just 1% of the cost. Likewise, if the batch is 10,000 parts the contribution is 0.01%. You can see the contribution quickly becomes relatively small.
As for overhead and profit, the argument can and should be made that each part produced should bare an equal part of those costs. Part 1, part 10,000 or part 100,000 should all bare the same percentage. If not, it could be argued that the lower volume parts are subsidizing the higher volume. One customer paying higher rates than other for no apparent reason.
Now to address material. If you agree with the above arguments, then material is just an extrapolation. One enterprise’s product is another’s material. For example, a steel mill produces steel in batches. If you buy the entire batch or a portion of it, it still costs the same to produce it. The buyer, a stamper in this case, should attempt to negotiate the same price for a portion as does the buyer for the entire batch. If the stamping house is successful in this negotiation, using the above logic, volume of material had no bearing on the costs. In very simple term, why would you pay more for less?
In the LinkedIn discussion, I received pushback because manufactures often and maybe correctly only buy enough capital to cover the requirement. I attribute this to LEAN principles. Lean manufacture figures out a TAKT time which is basically time available dived by number of products required. “The dumb beat of production”. This gives a “time per part” required. What should happen when cycle time is less than TAKT time? Should the process run as slow as needed? Here is an example problem:
(Using only a 1 shift operation just for simplicity)
1 Shift = 2000 hours per year.
Yearly volume is 120,000 pcs.
TAKT time is 60 Seconds.
What if this part is a plastic molded part for example and has a cycle time of 30 second with a single cavity mold?
- Should you run a part every 60 seconds (TAKT) time and dedicate this line to one customer? Charge them the full cost, 100% capital, burden? Charge the labor at 60 seconds?
- Should you run at 30 seconds, leaving 50% of the line unutilized, available to sell to another customer? Charge it at a rate of 50% of the capital, burden and now the labor is 50% as well.
- Assume you increase cavitation to a two out mold thereby reducing the cycle time to 15 seconds. This would in turn free up 75% of the line for additional business and lower the cost to your customer yet again. The cost would now be 25% of the original option. (Obviously the cost of the second cavity needs to be included in the analysis as well as any increase in press size but usually these are not as large as expected.)
Note: If your process is close to cycle time, in this case 60 seconds, then you are actually really paying for the time used.
Would you answer depend on your perspective? If you are a seller, which one is safest? Which one could lead to more business? If you are a buyer, which one do you prefer? Which one offers you the best price?
To very quickly address the cycle time vs TAKT and to a larger extent, shifts, I’d like to propose a very simple example. Imagine you own a stamping press. A very easy way to envision these arguments is to imagine every time the press cycles, it prints money. If you own this press, wouldn’t you want it to be running as fast as possible and for as long as possible. If you are running only one shift, the press isn’t printing money on the two remaining unused shifts. Similarly in the initial example, assume it is printing $1.00 per cycle, it if ran at TAKT time it would only produce $2000. If it ran at 30 seconds, it has the potential to generate $4000 and double the cavitation and you are at $8000. (This is one shift) Which would you choose?
Founder and CEO
Society of Cost Engineers
Owner and CEO
Center for Strategic Cost Management