Printer Friendly

MEs in the '90s: breaching the design wall.

MEs in the '90s:

The walls are coming down! No, not the trade-barrier walls (maybe another time), but the walls of specialization we erected between design and manufacturing. The walls that helped create the inferiority feelings MEs have suffered for decades. The walls that helped engender the perception among engineering students that design engineering was glamorous and shop-floor engineering was drudgery - punishment for those who couldn't cut it in high-tech math and technology courses.

What's bringing down these walls? Time-to-market is the key motivator - it will be the '90s competitive edge. It's driving the auto industry to reduce the time it takes to get a radically new car model from the drawing board to the showroom. Japan has shown it can be done in as little as four years. Detroit has traditionally taken five or six years. By 1995, industry experts tell us, that auto bogey will shrink to three years, and to a mere two years by the turn of the century. Car companies who can't turn around this fast will be trying to sell Model Ts in a world market that expects last month's space-age technology. Clearly, they won't survive, and the message's the same for those of you in other industries.

It is becoming clear that a major competitive problem for US manufacturing is not that we're being outproduced so much as outdesigned. It will take time to get our act together, but when the full integration of our design and manufacturing functions is accomplished, we will once again be unbeatable in the marketplace.

The surge into SE

Thus, time to market - whether autos or other discrete products - is the motivation for leveling the artificial barrier between design and manufacturing, and the method is simultaneous engineering (SE). Surprisingly, the SE movement has already taken hold in most of the companies we contacted (programs they initiated anywhere from five years ago to just last week), but none have it fully perfected yet.

MEs have know this for years: Half your time is being spent on engineering changes, the majority either unnecessary or avoidable with a little foresight and cooperation. If these headaches can be anticipated and worked out ahead of time, a new design project could sail through production in record time. Also, if designers really knew what manufacturing was capable of producing reliably, quality would improve.

In the past, the design task was simple: design the best products. Today, it's much more complicated. A winning design must be manufacturable - at minimum cost with maximum control of its quality. Designers can't do this all alone. They need your help in detailing the part, establishing its tolerances, and deciding whether it's really best to make it in-house or outsource it.

Some SE obstacles

But between today and the future payoff of full implementation of SE lie a few problems: * Role changes. MEs - traditionally scientists, technicians, and mathematicians - will function in the future as operations integrators, more responsible for coordinating people, information, and technology within manufacturing. To do this, a significant number of you will need to improve management and business skills, develop new areas of knowledge, and achieve a much broader managerial perspective.

Also, decades of designers playing the role of prima donnas and MEs that of shop-floor grunts must now change. To work effectively in teams, your status and theirs must equalize, and it will over time, but meanwhile, there will be a few crushed egos as each of you learns the importance of the other's expertise. * Time loss. Getting manufacturing involved earlier means that it will take longer to get designs finalized. Yet, once the production process starts, this time loss is made up with interest: products move faster, the production cycle is shorter, inventory can be reduced, and most important, the deluge of transitional design changes is reduced to a trickle. * Organizational shifts. The team approach means pulling people away from their immediate supervisors for long periods of time, giving them new autonomy, while asking them to serve several masters. This can strain the organization and blur the manager's ability to rate his engineers' performance. * Statistical training. The new emphasis on process efficiencies will eventually require training for both ME and DE in analytical methodologies to measure and then optimize process characteristics; i.e., statistics courses and instruction in Design of Experiments (DOE) or Taguchi methods.

Getting management approval to spend time and money on this may be difficult with their short-term orientation. Also, they may be unwilling to commit resources for process testing and development. * Productive meetings. All engineers who know how to conduct a good, productive meeting, please raise your hands. Hmmm, not many hands, right? Well, meeting efficiency will be vital to productive teamwork. In addition to the clashing-ego problem mentioned already, engineers have a tendency to provide too much useless input and waste valuable time. Also, if a team has a weak person in a leadership role, its performance will suffer. * Finding time. A company's people resources may be stretched too thin to make this transition. It may be a challenge to spare time for teamwork when many departments have been cut back in personnel. * Database development. The benefits to SE teamwork of a common electronic database are obvious, but establishing one may be difficult and slow the time it takes for SE methods to produce tangible benefits (although some feel the database issue is secondary to establishing SE methods). * Motivation. As engineers rotate from one team to another, they will find themselves leading on some projects and following on others. With reduced layers of management, motivation will be via peer pressure, engineer to engineer, department to department. But, some middle and upper management people tell us their MEs are too timid to take risks. To be a hero, you must be willing to risk getting fired. Will you be willing to take such risks when you get the chance? Will you be confident that your risks will be rewarded?

Report from the front

Peter C Van Hull is a leading engineering consultant, a partner in Anderson Consulting, Detroit, MI. As Director of their North American auto-industry practice, he provides auto companies with strategic planning, competitive assessment, quality and productivity improvement, and most important for our discussion here, product-development-process improvement.

We asked him what success the auto industry is having implementing simultaneous engineering? "I'm not familiar with Chrysler," he admits, "but Ford and GM are well into it. They have been working on it for at least four years, yet both have a long way to go to fully implement it. The Japanese were doing this long before US industry got serious about it, but there's no question US industry is very serious about it today.

"Toyota may have been the leader, and Honda has also done a very good job of exploiting the idea of short product life cycles and short-cycle manufacturing. Honda has that down to a new car developed in four years now, and is driving US automakers to match that capability."

Have the US automakers finally destroyed the wall between design and manufacturing? "Conceptually, yes, but the real issue is implementation, and that takes time. There are many issues involved: new ways of doing things, new knowledge, functional barriers, turf battles, and the idea of putting more resources up front."

Managing change

"Think about it," he continues, "you're starting a whole new way of doing things - new business processes - and there will be a transition period with some programs done under the old process and some under the new. The old process still requires a huge pool of resources - people fixing problems at the start of production or thereafter. Implementing SE programs means shifting huge numbers of them to the front end of the process. Most companies can't do both simultaneously - finish the old and start the new."

So, he sees a five or six-year transition period before full SE benefits emerge. "Perhaps, they need more resources during this period, but this demand will peak and then go down. There's also a skills issue that limits how fast they can make this transition. Bringing in new people would not really contribute much if they don't have the right experience or skills."

What's the consultant's role in this process? "Integration," Van Hull replies. "We get the parties together, provide information to enable them to decide what to do differently, and give visibility to the problems. We help make the process work by helping them deal with the `management of change' issues, but management itself must make the decisions. Our role is facilitating this process. These changes don't happen easily - it's a very different way of doing business - and this can't be done from the outside."

Pyramids of teams

We asked Van Hull if there's such a thing as a typical automotive SE team? "After a vehicle is subdivided into its major systems and components, teams are established on a system by system basis. The number of people on a team will depend on the complexity of the system addressed. Transmissions or drive axles are very different from seats or instrument panels, for example. So a series of teams is required, and then a hierarchy of teams to coordinate team activities. At the top is a program-management vehicle team.

"A typical team will include design people (including design management), manufacturing engineers, purchasing and materials-management people, line manufacturing people,(*) manufacturing management, financial people (to deal with cost pressures, target costs, etc), and quality people (depending on how that function is set up within the company). Essentially, these are all the functions required to design and produce the product - including all the supplier issues - and they must be involved from the start."

Do design people always lead the team? "Somebody must lead, but it doesn't have to be design. Ideally, it should be the people with the best leadership skills, and if not, that can be a problem."

How about a situation where a team member is not up to the task? "If the skills aren't there, you will have a screwed-up program. Staff it wrong, and you can predict the team will fail. The key to correcting a staffing problem lies with the team leadership and the program-management team. That's their job. It's the responsibility of the leader to know what the right skills are and where to get them."

Is this a problem, based on your experience? "Yes. Where there is weak leadership, the project fails, and the blame belongs to that leadership."

Team-member motivation

What are the rewards or motivation for team members? Does their performance get back to the right boss for review purposes? "That depends on who's in charge of salary review, the program manager or the boss of the team member's functional department. As you look at the evolution of the SE process, you can build a case that performance review should be based on the success of the program. The real reason you have a functional organization is the development of skills and capabilities, and their execution is through the program or product teams. So evaluation should be based on that execution.

"So the issues are: Who's doing the evaluation? and Whose evaluation counts most? If this is not straightened out from an organizational standpoint, and you're trying to deal with teams when all their incentives, measurements, and rewards are based on their functional roles, you have a system that will self destruct. You better have the incentives and rewards aligned with how you're trying to do business.

"It's amazing how many organizations don't consider these things. If your reward systems are not reinforcing the action and behavior you want, you're asking for trouble."

How does the company, in turn, evaluate the benefits it's getting from the team? "There are several key things: time to market; meeting the cost, quality, and performance targets for the product; and the absence of problems. (The more problems, the worse the team has performed.)

"The basic issue, though, is time to market - being able to steal a competitor's market share by coming out with a better product before it can respond to that same opportunity."

Can you quantify that? "Our Delphi survey says that the product leadtime target for whole vehicles in 1995 will be three years, and two years by the end of the decade. Historically, lead time has been five and six years. If you can't learn to play the time game in automotive and our other highly competitive global industries, you're dead. The price of not being able to play is extinction. Fear of failure is a great motivator."

Trading places?

Does the idea of rotating roles - ME and DE swapping jobs temporarily - make sense? "Although you cannot rotate people in the middle of a product-development effort, I've seen some of this development of cross-functional skills being done increasingly, but it takes a long time to do.

"What happened in the auto industry, as happened in other major industries, is that they got carried away with specialization. They originally took this approach to labor on the assembly line, and then applied it to the salaried and technical organization. They created people with very narrow technical specialties - people who by design can't see the big picture. It took them 20 years to accomplish this, and it will take them nearly as long to uncreate. It is really an issue of continuing education and continual broadening of people."

Is there a need for formal cross training? "Yes, but the training that will stick is on-the-job training, much more than classroom training. Supervisors and upper-level people who know all the necessary things must train the people working for them. The key role for leaders of these teams and departments will be to train people. Some formal education or classroom training is important, but the organization must figure out how to continue the educational process with on-the-job training.

"As far as cross fertilization goes, the product design people should go through training in design-for-manufacturing and just-in-time production. MEs also need to know these things, but I'm not so sure that they do, even today. Obviously, MEs need to know about statistics and be able to deal with design for experiments."

Designing those experiments

Van Hull reports that DOE methodology has caught on fast in the automotive industry. Doesn't it require extensive statistical training? "Yes," he replies, "but the key is keeping it simple. If these things get too complex, they break down. I'm not so sure that Taguchi methodology, for example, will be applied on a broad basis.

"US industries are using DOE philosophy, but they're keeping it as simple as possible. On the other hand, oversimplification doesn't work either. An understanding of basic statistics is the key, but the type of math that is the domain of those with masters degrees is not the way the world works. Using statistics effectively is the way the world is going.

"For example, one auto-company casting division is using DOE tools and techniques very effectively, and that's not an exception. These techniques are spreading and becoming a way of life for the industry. Of course, with both DOE and simultaneous engineering, people expect these things to happen overnight, when the reality is that they will take five to ten years to implement."

How about the problem that these experiments create scrap? "The question is whether to do these tests in the middle of production processes or invest in duplicate capacity and do them off-line. Some things are better done on-line and others off-line. Anything can be made simple by breaking it into its basic elements. You can afford to schedule extra production, make bad parts, and throw them away when the payback is major process improvement.

"I've been in many plants - glass, plastics, steel, etc - where a basic process is still considered an art, not a science, and what happens happens almost by `black magic.' I translate that to mean: `We don't know what the hell the process variables are.' Well, guess what - there's probably somebody somewhere in the world who, for any one of those processes, does understand it and is probably the world-class producer.

"So, the price of gaining that understanding will be well repaid - particularly when you're looking at huge production volumes. Although the results are typically long term, that payback will be very large."

Role of the computer

What's the computer's role in the SE team concept? "It's vital, but it better be focused. The idea of creating a great database in the sky with no focus - where nobody can figure out what information is really required - can waste a lot of money. My thought is effective use of automation and computer support is very important, but just throwing money at it as an end all - feeling that it will solve the problem all by itself - is something many companies have found doesn't work.

"The right approach is to understand the process, simplify it, and then use computer support to take it the rest of the way. For example, I suspect that a lot of companies threw a lot of money at CAD/CAM-type systems just to use the CAD portion of the system for automated drafting and have received very little return on their investment.

"My thought is that a wise and well thought out use of computer technology is certainly required to make a success of SE, but not as a prerequisite. You should be doing simultaneous engineering with pencil and paper, if you have to, and then automate the things that really make sense to automate. I think it will be a case of simultaneously development of the electronic database with simultaneous engineering."

Do simulation techniques have a role here? "There's no question simulation techniques can contribute to the simultaneous engineering process. But, you better understand cause and effect and how the variables work if you want to be able to simulate!

"A basic understanding of your process comes first. You need the DOE level of understanding before you can benefit from simulation of the process. The real key to simulation is not the computer tool, it's the people with the knowledge who can define the rules of process interrelationships. The computer just helps you try out more alternatives, faster, without a lot of investment in tooling, etc - a much more economical way to experiment. Clearly, simulation is very important."

Quality by design

Is final product quality really a function of design? "Yes, but the design better incorporate thinking about manufacturability, the rules and guidelines of the manufacturing process. Designers better understand these processes, or they will be creating all sorts of problems. Product and process design must be concurrent; they cannot be isolated. Product design not only determines final product quality, but also cost.

"We're recommending that management have a process in place to continuously improve the SE process. How well business processes are executed by Company A versus Company B is what will separate the winners from the losers.

"Simultaneous engineering is much broader than just engineering. It's simultaneous product development - a process that must be continuously improved. This means management must pay attention to it, and spend time and money to assure this improvement. There's often a manager on these SE teams, and it's management's job to improve the overall team process. The financial processes, purchasing processes, etc, had better match the product-development process, or they will get in the way.

"Integration of these various business processes - that's management's responsibility. Another key overriding issue is sticking with it. This is a long-term change process. This whole simultaneous-engineering process is probably one of the biggest management-of-change problems most organizations will ever face. They must learn how to deal with the management of change. That's the big challenge."

(*)Because the ME never knows as much about his manufacturing processes as those who run them. Says Van Hull, "Some people have assumed that if you put all your engineers together, you've got all the process problems fixed. They fail to recognize the huge gulf that typically exist between the manufacturing line organization and the ME. That may be as wide as the gulf between ME and DE. The line person probably knows more about the realities of the process than the ME."

PHOTO : In 200 BC, in a great feat of engineering, the Chinese build a Great Wall to protect themselves from invaders. In 1200 AD, the wall fails to deter the invading Genghis Khan Mongol hordes. In the 1800s, the US invades from the west.

PHOTO : In the 1960s, in a great feat of engineering, America's auto industry builds a Great Wall to protect their elite designers from the great horde of unwashed manufacturing engineers.

PHOTO : In the 1980s, in a great feat of engineering, the Japanese blast through the wall between their designers and MEs, and invade from the west, capturing US auto-market share with an amazing time-to-manufacture of only four years.

PHOTO : In the 1990s, in a great feat of engineering, America's auto industry dismantles its engineering wall and battles back with simultaneous engineering to meet the shrinking time-to-manufacture bogie: three years by 1995, two by 2000. Moral: Walls are barriers to both sides.
COPYRIGHT 1990 Nelson Publishing
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1990 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:manufacturing engineers; includes related articles
Author:Sprow, Eugene E.
Publication:Tooling & Production
Date:May 1, 1990
Previous Article:NCMS study reveals savings with ductile iron.
Next Article:Milling: pushing carbide to the edge.

Related Articles
Challenging roles MEs will play.
MEs speak out on their changing role.
Reading the road signs.
IBM and Camstar provide ASAT with integrated enterprise resource planning and manufacturing execution solution to enhance business processes.
Online exclusives.
Clearly accountable: FMA clarify the benefits of embracing MES technology.
Mill technology utilization--a holistic approach.
New loaders for central and stand-alone use.
Money available for manufacturing enterprise research.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters