Our Sites

The capacity dilemma in custom metal fabrication

Why today’s churning markets require a new viewpoint about capacity

Most companies start the new year armed with a plan, a new budget, and a lot of hope that the year will unfold somewhere close to what has been forecast, along with a prayer that some pleasant upside is lurking.

Of course, the probability that the plan will be exactly right is almost zero, since forecasting the most important and volatile drivers—orders, shipments, and timing—is by nature problematic. But the plan doesn’t have to be exactly right, just right enough: say, within 5 percent of what the actual results will turn out to be.

In a relatively benign, static market environment, that’s usually achievable. In a market that is growing very slowly but is nevertheless churning, planning can be quite prone to significant error. A churning market is one in which customers are restless, demanding more on service and pricing to offset the lack of revenue growth, and are willing to churn the supply base—change suppliers—to get what they want.

Therein lies part of the challenge for suppliers: In this type of environment, which certainly exists as 2015 starts, how does one really plan with any significant degree of confidence? To answer this, we have to understand the nature of the challenge.

Two issues are involved here. The first is the amount, timing, and mix of revenue. The second is capacity, which determines the budget. The two are directly related, but usually in only one direction: Revenue influences capacity. But in a churning market, there is a subtle component mixed in. The subtle part is how capacity relates to revenue surety—that is, how capacity estimate errors affect the risk to revenue.

Capacity planning is normally driven by expected revenue. This sounds pretty easy in a static, slow-growth environment. But what happens if revenue itself is driven, to some degree, by actual capacity? In other words, what if the relationship between revenue and capacity is not one-way but instead bilateral, with each affecting the other? This scenario requires some serious thought.

In “normal” times, managers try to set planned capacity at the level required to accommodate forecast revenue and mix, and then offset and time the requirements to provide some desired level of service as measured by on-time delivery (OTD). In these times, if a shop has more capacity than estimated, that error surfaces as excessive cost and lower profits; if the shop has too little capacity, the error surfaces as extended lead times.

With too much capacity, managers naturally responded by reducing the variable part of capacity with real demand. With too little capacity, managers usually informed customers that the new lead time was now, for example, four weeks instead of two. The economic impact and risk generally were low. Customers weren’t thrilled, but they lived with the change. The economic bias, therefore, was to err on the side of too little actual capacity.

That was then; this is now.

The Risk Has Changed

Now, in the case of excessive capacity, we do the same thing: reduce variable capacity to match demand. But how about in the case of too little capacity in a churning marketplace? Does the old remedy still apply? For some of your customers, the answer is undoubtedly yes, if the lead time stretch isn’t too significant. But for some the answer is probably no.

Unfortunately, the ones that will reject the extended lead times are usually your largest customers—the ones with the biggest impact and most stringent demands for service, and the ones you simply can’t afford to lose. If you persistently extend lead times or provide poor OTD performance because of lack of actual capacity, you risk succumbing to market churn. You can lose business—a lot of it—and you will have one hell of a time replacing it.

The significant loss of business from a major customer because of delivery issues in a churning, low-growth market is a very real possibility because poor service is no longer just an annoyance. It’s costing the customer money. As large and powerful as this customer seems in this type of market, it really has less leverage over its customers. It too is fighting like mad for every order. It must respond very quickly and is probably operating with very low inventories. This means that your customer and its suppliers must perform on short cycles.

If one of those suppliers is you, the only way to accommodate this demand is to have the capacity to do so. So the risk of having insufficient capacity is no longer low. It is significant. For example, the loss of $500,000 in business will translate to a hit in the neighborhood of $200,000 to contribution margin and the bottom line. Ouch!

In markets that are slow-growth but churning, the relationship between capacity and revenue is a two-way street. Revenue still influences capacity, but capacity also influences revenue, and the risk-adjusted economic bias errs on the side of too much capacity—usually termed a capacity buffer.

At this point we run smack into the capacity dilemma. If we have more than is required at any given time to service customers reliably, our profitability is impaired, day in and day out. This is certain. If we have too little, we face the real threat of losing business as a result of poor service. This is not certain, but if it happens, the outcome can be profoundly negative—much more negative than having a thoughtfully planned amount of excess capacity.

One way out of the dilemma is for markets to rebound to a condition of robust growth. At this point, the risk profile would return to “normal.” In the meantime, it sure would be nice to escape this dilemma without increasing ongoing costs or playing the risk game with key customers. Happily, for most fabricators, there is a way out.

Effective Capacity

Unlike absolute capacity, which assumes 80 to 100 percent utilization and efficiency of people and machines at a given product mix, effective (or realizable) capacity is the capacity that you actually have in real life. It’s lower than absolute, often much lower. There are many reasons that absolute capacity degrades to the lower value. Among them are:

  • Unfavorable product mix.
  • Excessive scrap and rework.
  • Machine or people downtime higher than planned or expected.
  • Employees’ variation in skills, performance, attendance.
  • Non-value-adding shop procedures like searching, excessive moving, counting, confusion, resolution.
  • Large-batch processing that creates excessive work-in-process and job cycle times.
  • Scheduling issues, such as multi-point scheduling and uncoordinated sequencing.

When you account for these, you find that your effective capacity—what’s actually realized—often is much less than what you had planned, and to your great frustration, you are behind the eight ball in OTD performance.

Here’s the rub: When you are operating at near planned capacity at any constraint operation, even small degradations in that capacity will have a large impact on actual throughput. But this also implies that even relatively small improvements in capacity at those constraints will produce outsized improvements.

This is where practical lean methodologies come to the rescue. Lean is not just about reducing costs. It’s also about increasing capacity with the resources you already have. The reason is simple: With lean you are removing non-value-adding time (waste) and replacing it with value-adding time. You are increasing effective capacity.

The capacity robbers cited previously are exactly what practical lean methods are designed to minimize. Relatively simple deployments of practical lean elements, such as 5S and the visual workplace, machine uptime monitoring and practices to increase that uptime, scheduling discipline, cross-training, and information standardization, can drive large improvements in the capacity that you actually realize.

This is the capacity buffer that you need. It’s the low-cost, low-risk bias that you create to account for uncertainties in demand (both in volume and customer service), mix, and variation in people skills and other attributes.

In the era of slow-growth, churning markets, where customer pricing and service demands are severe, fabricators must have a bias toward excess capacity. It’s simply too risky otherwise. But this buffer cannot be one that adds to recurring costs. That would be solving one problem (service) while making the other problems (cost/price) worse.

If there was ever a time to begin lean processes in your operations, it is now, in this environment. They are your way out of the capacity dilemma.

About the Author

Dick Kallage

Dick Kallage was a management consultant to the metal fabricating industry. Kallage was the author of The FABRICATOR's "Improvement Insights" column from May 2012 to March 2016.