Demand variation: Challenge or opportunity in custom metal fabrication

Demand variability is a daily reality in custom fabrication. It creates tension, questions, conflicts, and more. So what are manufacturing leaders and lean practitioners to do?

For starters, understand the source and type of demand variation affecting your company. If the variation is cyclical and repeatable (say you make parts for snow blowers and lawn mowers), then you can develop a strategy that is methodical and reasoned. If the variation is unpredictable and swings wildly (think repair fabrication for customers’ breakdowns and equipment failures), then your approach may need to be creative and fast. In either extreme, you can use some common approaches to improve your ability to be responsive and maintain customer satisfaction.

What Are the Options?

The options presented here cover a wide range. The intent is to help paint a picture that allows you to see possibilities and how one option compares to others.

Let’s start with the weakest and least improvement-oriented option: You accept the demand variation and ramp resources up and down. This does not focus on improvement but instead creates business disruptions. It’s a feast-or-famine environment full of workforce interruptions and economic stress. In essence, this “do nothing” approach will have you in a constant state of firefighting.

Alternatively, you can pursue a level-load strategy to minimize the peaks and raise the valleys. This provides some relief—you no longer operate in a feast-or-famine environment with workforce interruptions—but also induces other economic stresses. You have working capital tied up in inventory. This is especially tricky if you assume all the risk for inventory held during the valley that might become obsolete for lots of different reasons.

The most proactive and progressive option is to increase flexibility to absorb the demand variation and to be able to respond quickly to increases or decreases. The lean body of knowledge offers techniques that result in greater flexibility.

The improvement approach you craft will most likely be different from the company’s down the road because your company is unique. You have a different “current state” in performance, resources, customers, and demand variation. However, some combination of the following fundamentals should help. These stand-alone options are most effective when woven together as a comprehensive improvement approach.

Shorten changeover time. Find the constraint operation and look for ways to shorten the changeover time. Changeover time is non-value-added and consumes capacity. Focus on changing internal elements of the changeover (work done while the machine/operation must be stopped) into external elements (work done while the machine/operation is still producing the previous run of product). Identify opportunities to convert fixtures from single-use into universal-use, eliminating most or all of the changeover. Finally, make sure tools and materials are ready when the changeover starts. The operator shouldn’t spend time looking around for the tools and materials he needs to do the job.

Reduce batch sizes. In many operations, decreasing batch sizes seems counterintuitive. Don’t larger batch sizes drive greater efficiencies? What might be gained in efficiency is lost in effectiveness. Large batch sizes tie up operations, stretch out lead times, and create accountability problems. How do you get all the right parts at the right time at assembly? What if a design specification changes? Who ran those defective parts?

The ideal batch size is one, of course. But even if a batch size of one is not realistic for you, consider a quantity significantly smaller than what you currently have. This will lead to more ideas that will help make operations more robust.

Reduce throughput times. Here, throughput time refers to the time from when the product starts until the product is complete. If you don’t currently measure it, develop the metric to drive the desired behavior. Measurement is a catalyst that provides the baseline for improvement. Once you start measuring, you can determine what an acceptable throughput time needs to be. The gap between current and desired throughput times identifies the magnitude of change required.

Reduce defects and rework. Defects and rework adversely affect flexibility by disrupting flow, creating “hot-item” runs, and inducing major and minor stops—major if you scrap a piece and start over; minor if you stop a process to tweak, adjust, and repair. This can happen from the front end (getting the order accurate); through the making of the parts (physical product quality); to the packaging and shipping of finished goods (right product, right quantity, and right information). As defects and rework reduce flexibility, the demand variability effect grows.

Your unique approach will be a composite of specific improvement methods mentioned here, plus a few others not addressed. When you implement these in a systematic way, you’ll find that these methods complement each other well.

Consider again two extremes of demand variation: the very cyclical but somewhat repeatable contract work of lawn mowers and snow blowers, and the highly unpredictable nature of certain industrial fabrications and repair work.

In either case, lean fundamentals dampen the effect of demand variation. But at the extremes, certain challenges start affecting inventory, working capital, and the workforce.

Cyclical and Repeatable

Say your shop has a customer mix focused on, among other industries, the lawn and garden sector. People who buy lawn mowers may also buy snow blowers. Companies generally know when people buy them, and they also know they usually don’t buy a lawn mower and snow blower at the same time. It’s a classic example of a seasonal product mix: Demand peaks during one part of the year and troughs during the other part of the year.

If you have a capacity constraint on key pieces of equipment that would require significant capital investment, the level loading option may be a way to manage through the peak periods. You would execute this by spreading some of the peak demand backward to the lower-demand periods.

This approach has many positives. It allows you to maintain customer lead-time expectations by having finished-goods inventory and a managed backlog. You can also delay buying capital equipment until you can prove the demand is sustainable and finances are in shape to make investment. It helps avoid layoffs and provide greater workforce stability, which is particularly important for positions requiring highly skilled people.

Still, the approach does have significant downsides. You need to work off of forecast demand; the longer the forecast period, the greater the risk for forecast error. This also puts your company at great risk if the customer does not absorb some risk for demand changes, engineering changes, and obsolescence. It also strains cash flow as you pay for materials and labor that go into product that will sit in inventory for some period of time.

In short, level loading has its weaknesses and challenges, but sometimes the benefit is greater than the cost.

Wild Swings in Demand

So what does the manufacturer do with unpredictable demand that may swing wildly from one day or week to the next? Again, think of a fabricator that supplies new or repaired parts for nearby processing plants. Level loading probably does not make sense, considering the demand unknowns.

In this extreme example, the key is to relentlessly drive down throughput times and maximize flexibility. Your inventory exposure is most likely in work-in-process (WIP) rather than finished goods. However, you can minimize WIP by streamlining flow.

The more you streamline the flow, develop a “once it starts the first operation, don’t put it down until you finish the last operation” attitude, and simplify processes by having tooling and equipment where they need to be and ready for use, the more effectively you can shorten throughput times. This is how these techniques begin to operate as a system—and the power of continuous improvement really comes through implementing lean as a system.

Turn That Challenge Into a Competitive Advantage

Demand variation does create challenges that require focused responses. Wise manufacturing leaders understand the sources of their demand variation and respond appropriately.

As painful and disruptive as demand variation can be, fabricators that minimize its effect also have a competitive advantage. There is no one-size-fits-all solution, so be creative and aggressive when developing and executing your unique improvement plan.