Electronic Design

Bob's Mailbox

Hey Bob:

Thanks for the skepticism you express in your Pease Porridge column----it's what we need from editors. It's very hard to successfully compete with the deep-pocket boys from "EMPEROR'S MENS WEAR INC.," when editors glom onto every hype campaign with fawning praise.

As you have observed, the feather merchants have usurped and redefined "Corporate Reengineering" to mean fixing everything just by merely attending their seminars, being reborn, and writing checks.

Over the past 30 years, we have faced a fundamental problem in the manufacturing world which nearly defied solution. It was chaos. Chaos caused by a lack of discipline in documentation; that is to say, deplorable record keeping. Oh, the bean counters were doing fine, as always, financial accounts were fine, and they took charge of the data-processing operations and guarded them jealously.

In design and manufacturing engineering, technicians in short sleeves struggled with the production of thousands of parts and assemblies making up the product lines. They wrote many thousands of shop routings, process plans, and tool designs; and managed hundreds of machine tools and assembly operations to keep thousands of products flowing into the marketplace.

But, as we began to carefully examine the systems and procedures used by these engineers and technicians, we saw that they were overwhelmed by the sheer volume of the work and detail----plus the knowledge required to do all of this work. In fact, they couldn't even keep up with the state of the art in their own companies.

The left hand was constantly, diligently duplicating what the right hand had just done, and not too well much of the time. They were constantly reinventing the wheel. And, as the pace of change quickened, they were forced to speed up the technical work until mere confusion developed into chaos. Duplication and unnecessary variety increased exponentially.

Product designers had little generic access to existing designs that would allow them to create better designs with each generation of change. They literally "duplicated" existing designs because they couldn't find existing parts and assemblies they could reuse in new designs. As a result, parts variety proliferated, drastically reducing production-lot sizes. The relative parts-to-workstation ratio grew out of control, and scheduling and production control became even more chaotic.

A plant that had successfully produced 30,000 different parts at previous volume "X" was strangled by complexity trying to produce 60,000 parts at half of that volume. The smaller lot sizes resulted in shorter production runs, and greatly increased downtime for setup and tool changes. Process planners blindly routed new parts through the manufacturing operations as best they could----without integrated plans or guidelines. Duplicate or similar parts were routed through whatever work center seemed appropriate to the planner, who had no ability to learn how other similar parts were routed. Or, what various routing options had yielded in cost or quality. All that information was (gasp!) lost in the files. He had no formal knowledge, or formal procedures, or formal guidelines.

Seeing this, we decided to develop tools to eliminate that chaos of knowledge unavailability. We took copies of the design and manufacturing files home with us and sorted them out. They were truly chaotic. Over 200,000 drawings in nearly random order.

Almost none of the drawing files were organized, therefore a designer could retrieve existing designs without knowing the specific part numbers of designs previously drawn. Nor could a manufacturing engineer locate similar parts or assemblies for reference. It was always cheaper and more expedient to redesign parts than to spend hours searching through the design files for something that might not exist. Consequently, 5% to 10% of the designs on the drawing board at any given time are exact duplicates of existing designs----most "lost in the files."

Similarly, process plans are duplicated or needlessly proliferated because the existing plans for similar parts cannot be located. Thousands of parts flowing uncoordinated through the manufacturing enterprise causes pure chaos. Accountants may know how many are on order, are finished, or in inventory. But, they don't have any idea what the shop is making or how it should be made. Essentially, they don't know a great deal about what business the manufacturing organization is in!!

Manufacturing firms do not make machine tools, or appliances, or aircraft. They manufacture hundreds of thousands of parts and subassemblies that are subsequently assembled into those products. That involves an enormous task of planning and control that is nearly impossible to carry out. Parts making is the major business that the shop is in, but few managers seem to realize that.

And that parts-making process needs to be reengineered. We need to establish new tools to organize product-design knowledge. We need formalized design standards and procedures to obtain more-manufacturable designs. We need formalized process-planning systems with standard routings and rationally engineered processes.

We need to identify the specific nature of the overall production requirements before we buy another machine tool, or a new automated production system. We need to rationalize the plant layout and materials-handling system for organized, flexible manufacturing.

That's reengineering! The real Consultants have been doing it for 30 years. It's not done by generating employee empowerment in meetings of quality circles, or hotel conferences with MBA consultants. They haven't a clue as to the nature of the problem----let alone a solution. Besides, it's too much work. They get paid excessive fees to hang banners, show slides, and mouth professorial platitudes. And, they bask in the comfort of knowing that the vast majority have climbed on their bandwagon. It all reminds me that we are truly the sons and grandsons of the Hadacol generation.

There's a lot of hard work ahead for those companies that want to find out what business they are in, and address their shortcomings. There ain't no magic pills or cheap diet plans. Neither the PC programs nor the software gurus can do the work. It requires engineering----reengineering by those who understand the problem and are able and willing to engineer the solutions. Men like them are scarce. But, look around, you just might get lucky.

GILES E. LOVELACE CMfgE

Productivity Consultant

Industrial Cybernetics

Charleston, S.C.

p.s. I usually don't mail these letters, Bob. But it seems to me that industry is being sold the sizzle to the detriment of the entire society and a lot of people are joining the medicine show.

Giles----here's your problem: It's not easy to provide a good technical DESCRIPTION of your subassemblies. We IC guys have libraries of cells: a minimum NAND, a 2X Inverter, or a 100X Buffer are easy to describe and catalog. But how do you categorize a Library of Brackets, so you can find one with 10-mm holes when you need it? And when is a bracket not a bracket? When it's a flange, or a brace, or a corbel. Man, you're gonna need a good librarian! Now, there's nothing I could possibly say about "redesign" or reengineering. But if I just happened to get some letters, I might be able (in strict confidence) to put together some interesting stories...----RAP

All for now. / Comments invited!  RAP / Robert A. Pease / Engineer

Address:

Mail Stop D2597A

National Semiconductor

P.O. Box 58090

Santa Clara, CA 95052-8090

 

Hide comments

Comments

  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.
Publish