By Dick Morley, CONTROL Contributor
RECENTLY, I HAD major surgery. This surgery replaced both my knees, and additional reconstruction will be needed over several years. A great lesson was forced upon me. I must change my life. No more serious skiing, running or fast Harleys. The lesson appeared during occupational therapy (OP).
I thought that occupational therapy helped me go back to work. Wrong. This is a minor part of OP. This therapy trains the patient to deal with the world the way the world is. Low toilets, fast traffic lights and restaurant buffets are real challenges for even the moderately handicapped. We, the challenged, must deal with the mismatch between the environment and our resources.
Engineers, and especially process engineers, are viewed as the uber-nerds of the world. Yes, but we do make cheese and build bridges. We know that we perform the most useful of functions. We make the world’s food, but what if the world changes? How can we adapt and still perform the essentials for the society we live in? What is our occupational therapy? Can companies and technologies adapt in a positive manner to change?
A venture capitalist I know gave this advice to a new company, which was seeking startup capital:
Don't show a process solution, show a business solution
The cost of making stuff is insignificant
The enterprise does not want manufacturing utilization, it wants market response and time to market
The supply chain and changes in design are key to business productivity
The cost of a manufacturing error is insignificant
A big issue is the availability of raw material
American management thinks that reducing cost is the key to selling in the marketplace
Do I agree with these statements? I’m not sure. But it sure rang a Quasimodo bell with me.
Every project I do seems to be missing stuff, not missing tools. For example, we’re refurbishing a horse trailer. It uses lots of aluminum sheet metal for side and top covers. These are attached with pop rivets. In my shop, we have several pop rivet tools and a wide variety of pop rivets. The tools were optimized for quick and portable insertion of the rivets, but we did not have enough of the correct rivets to finish the job. Efficiency of insertion was not the issue. Availability of parts was the issue. It takes an hour for me to replenish the supply of the special rivets.
Another event happened during PLC manufacturing in the early days. We designed the stacking factor of the early PLC circuit cards with little extra room. Capacitors were chosen for function and fit. Suddenly we could not get the correct physical size. Production was held up for several days. The efficiency of part insertion is unimportant if you don't have the parts.
Let us imagine a control equipment manufacturer happily humming along, building process analyzers. Suddenly a European cell phone manufacture changes its cell phone design. The critical semiconductor we need has another demand. The supply sags for both of us. We cannot anticipate nor control such events.
As control engineers, we have been after faster cycle time with little regard for non-lean manufacturing. Tight cycle times and rigid design requirements reduce time-to-market flexibility. Many Detroit assemblers have a 30% return on net assets (RONA). Assets such as factories and process plants have excellent payback when they make stuff (and things) 80% of the time. When the number is 30%, the process plant is only making sellable cheese one third of the time. The 80% level is seldom attained, except in a “production-limited” market. A “consumption-limited” market forces the plant to only make the needs of the market. That means three times as many factories are needed to supply a market.
We engineers need to design processes that serve a broad segment of the market with flexibility. A consumption-limited market is here to stay. When we optimize a given local design and process, flexibility is lost. We make the business process worse. We should implement systems that can absorb supply chain sag and market changes. In the three examples given, we should allow a broad spectrum of rivets to be used on the horse trailer. The PLC card stack needs to accommodate several types of components, and one facility should be able to assemble several analyzer designs.
Another major issue is: Keep your customer happy. For example, many of the home supply superstores have a paint department that uses fundamental colors and a high tech mixer to custom make an exact color match for the customer. Earlier this year, a major paint provider changed formulation to reduce its costs. Most of the store’s mixers jammed because of the
viscosity difference. Fixing this problem will be more expensive than the chemicals saved, and the company lost its paint manufacturing market share—probably forever. Focusing on local manufacturing issues with little regard for the end user hurts. Reducing the cost of this paint product saved pennies, but sacrificed market share.
Raw plant costs are typically just 5% of the total cost of a product. Marketing, engineering, and other items make up the rest of it. Are we concentrating on the wrong item? What can we, the engineers, do? These days we don't have much say in strategic decisions. However, if the occasion allows, engineers should design processes to absorb suppler sag, and design processes with the market in mind. Product configuration changes should not materially affect the business. The factory needs to keep making product, no matter what. The cost of assembly is insignificant compared to supplier sag. With some occupational therapy, you can survive.
Credited with creating both the modern PLC and building automation control system, Dick Morley quit working so he could toil in a barn in New Hampshire. He lives in his wife Shirley’s house and can be contacted at firstname.lastname@example.org