Printer Friendly

Concurrent engineering in product design/development.

Teamwork that includes early involvement of outsource vendors and 3-D CAD tools has bridged traditional gaps between design, engineering, and manufacturing stages.

In recent years, much has been written in trade and consumer publications about such concepts as "concurrent" or "simultaneous" engineering, "design for manufacturing," "early vendor involvement," "total product development," and "lean manufacturing." Like all buzz words, these terms try to convey a lot with a little, and are often vague and confusing as a result. Their importance, however, is that they represent attempts by different people to describe something real that is happening in the world around them. These terms attempt to capture and label a basic--and in many ways evolutionary--change in the way companies design and develop new products. It's a complicated process, partly based on new technologies, partly occurring in response to competition from overseas, and partly just a reapplication of plain old common sense. But it is not simply a vague concept, someone's idea of what manufacturing will be like twenty years from now. It is a real process that can provide healthy benefits to companies that apply the lessons.

This article examines what we see as the main elements of the process--the features and benefits--and reviews some specific case histories that show how the techniques are being applied.

Future Needs

We began nearly two years ago by looking at what was likely to happen within the plastics industry with respect to product development, and within the design industry overall--how it was going to change, and what we could do to make sure we adapted to it.

What we found happening can be summarized by a quote from the author Tom Peters. In his book Thriving on Chaos, he says that for business to survive in the future, it must meet uncertainty

by emphasizing a set of new basics: world class quality and service; enhanced responsiveness through greatly increased flexibility; and continuous, short-cycle innovation and improvement aimed at creating new markets for both new and apparently mature products and services.

To see what was likely to happen in product development in the plastics industry, we surveyed our own customers, people within the industry (such as material suppliers), and people within our own companies. As a result, we came up with a list of what we believe will be the future needs of original equipment manufacturers (OEMs). The needs can be divided into three categories.

First, we must find a way to develop the right products. In too many cases, time and resources are wasted in developing the wrong products. The most frequent cause of new product failure is that the developers simply haven't identified what the customer wants. We need to develop niche products. Much as the automotive industry has done with cars, and as resin suppliers have done with blends and alloys, we must develop and manufacture products that fit increasingly tight segments of the market.

We must develop value-added products. This means improved functionality, features, and performance. Customers tend to gravitate toward the product in which they see more value.

Second, we need to bring the new products to market faster; CEOs of major OEMs have indicated this to be one of their highest priorities for the '90s. We can accomplish it through better market research. By identifying the product the customer wants, we won't waste time developing the wrong product; instead, we make the whole process faster and more efficient.

Once we've identified the right product, we need to be able to shorten the development cycle. For example, we cannot continue to take five years to develop cars in the U.S. if the Japanese do it in three. One way to shorten the cycle is to make better, more efficient use of technology and human resources--through effective use of outsourcing, for example.

Lessening the Risks

The third need of OEMs--sort of the flip side of increasing product value--is to reduce the risk involved in developing new products. Try getting a second round of venture capital if your development program is not on track, or if your product's initial market acceptance is not favorable. New products are a risky business, and we need to lessen the risk if we are to be more competitive overall. We see four ways to lessen the risk.

First, we need to work with professionals. There's too much at stake to put our trust in inexperience. Product development specialists can provide insight into the design and manufacturing process that can help manufacturers avoid costly mistakes or, better yet, help them find value-enhancing improvements. This will generate a higher success rate and lower product risk.

Second, we need to achieve a better return on our product development dollars. This speaks generally to all areas, but to two things specifically. Venture capitalists--even venture groups within existing companies--work on averages; they know they can't expect winners every time, and that some losses will occur. So, they hedge their bets. But as more and more products are developed for smaller, niche markets, room to maneuver becomes increasingly scarce. To improve our return, we need to improve our batting average and lower the overall costs of developing new products.

Third, we have to get it right the first time. The industry can no longer absorb the cost of re-engineering the product, or of retooling or fixing a mistake once a product enters the manufacturing arena.

Last, but not least, is the increasing movement to achieve higher and higher quality levels (6-Sigma quality), not only in the manufacturing cycle but in product development as well. We, as an industry, need to design quality in, not add it on.

A Changing Model of Development

How does all of this apply to the process of product development? What's changing, how is it changing, and what will it all mean?

As Fig. 1 shows, the traditional product development process has been essentially linear and segmented, with some very hard breaks between phases of the process. The two most critical breaks are the gap between design and engineering, and again between engineering and manufacturing. Basically, this "throw it over the fence" model shows that we didn't talk very much with each other. As designers, our concerns involved aesthetic appearance and packaging. We'd develop drawings and "throw them over the fence" to engineering. Engineering's concerns were not in aesthetics, but in making the technology work. They would utilize our drawings when they were suitable, and ignore them when not. Then they would throw the drawings over the fence again, and it would be up to manufacturing to somehow weld the two sets of drawings together and actually make something with them. Sometimes it worked, and sometimes it didn't.

Obviously, this is an oversimplified description of the process. Certainly, many exceptional products have been developed, and many product development groups have not followed this rather striated model. However, this has been the general model of development among U.S. companies. If it shows one glaring weakness over all others, it is that we were not working together; rather than working on the whole, we were all working on our separate parts. Not only was there very little teamwork; designers often had little conception of the difficulties a molder might face in executing a concept, and there was no methodology for the two to work together.

Figure 2 shows our view of how the process is changing. The helical structure graphically represents an organic combination of early vendor involvement and concurrent engineering. Essentially, it characterizes the product development process from product planning all the way through the production release of the parts, and provides an alternative to the throw-it-over-the-fence approach. The concept breaks down barriers between the areas involved in the process.

In the center is the core team of the client; they own the basic technology. Peripherally, getting that technology from an idea to production and market requires many services and players who must work together. The resin suppliers should be involved early in the process, not just in specifying materials after the part drawings are done, but in providing finite element and mold flow analysis when part design is still easily changeable. Tooling people also need to get involved early in the process so that products can be designed to fit manufacturing early on. And, certainly, industrial design must be involved early, when a product comes together conceptually. Design is not a cosmetic application, a matter of just making parts; it's a matter of building a product.

The other notion, which reflects the ideas of Tom Peters, is that the process needs to be iterative. One of the problems with the American concept of mass production is that, after putting so much investment into getting the job done, we want to turn the switch and let it run (and run and run), and then get a big payback. However, international competition has proven that this no longer works. In a niche market, you must constantly improve the product to stay in the game. This means that throughout the cycle, you make improvements as you go.

So far, the process appears to be nothing more than basic teamwork. But what's different now--and what's making the teamwork feasible--is the blue band that goes across the top of Fig. 2, indicating the CAD tools that are finally coming on line. The tools help bridge the gaps, and permit us to work together effectively.

This is an important point. CAD and CAM have been buzz words for so long that everybody is bored with them. They're old hat. Everybody has CAD, and everybody does CAM. Right?

Well, not quite. The concepts have been around for a long time, and many people have been working at CAD and CAM. But only recently have the tools become sophisticated enough to be practical, and powerful enough to encompass the entire product development process.

It is the development of 3-D CAD that has put these tools over the top. The ability to develop 3-D solid models on screen is bridging the gaps between design, engineering, and manufacturing. For example, 3-D files can be electronically transferred to molders, and used to cut tools directly; the same file can also be used for rendering and analysis. This has saved time and cost, and has reduced the risk of failure, more than any other element of the process.

Specific Cases

How, specifically, do early vendor involvement and concurrent engineering really work? Following is an excellent example of how early vendor involvement can sometimes make critical differences in product development. In this case, the product is a gel stainer, shown in Fig. 3. The problem was to create a product, for low volume production, that would allow a lab technician to position electrophoresis gels so that they could be stained and then dried. The original design direction was for a multiple-piece unit that would be assembled for the stainer/holder. Because the molder was involved early, he was able to suggest a tooling approach that would permit the same function with a single part. The design was then altered to accommodate the tooling suggestion. There was also significant involvement with the material supplier to identify a material candidate and a process that would provide a combination of very thick and thin sections, chemical resistance to stains and dyes, and the ability of the plastic part to shed the dyes without absorbing the colors. The final result was a structural foam polyester, molded with counterpressure, that worked very well. This sounds simple, and it is, but it happened because of the involvement of outsource vendors at the design stage of product development.

The next two examples involve products developed as part of a "Global Harmonization" program intended to create an overall look and feel for a company's products worldwide.

Figure 4, representing one of the earlier projects in the program, shows the value of bringing in outside vendors and specialists early and viewing the development process as a whole, rather than as a series of segmented steps. In this case, designing the product from the start with an eye toward ultimate manufacturability reduced costs considerably and enhanced success in the market.

The product is a "portable environmental incubator," designed to be back-packed into remote regions for collection of water samples for testing. Because the primary market comprised public health organizations in developing nations, the manufacturer knew at the outset that the unit would have to be produced economically if it was to have any chance of success. The initial designs and traditional processes (metal fabrication and injection molding) were either too heavy or too expensive. But because industrial design personnel were involved early enough and knew the processing options, they were able to suggest rotational molding of the case from polyethylene. This method offered low-cost tooling and a low piece price for short-run production; it also provided an exceptionally rugged case that could withstand the abuse to which it would be subjected, and that would look appropriate in a lab setting. The product also met the unusual design goal of being capable of flotation, in the event that it accidentally fell into a pond or river. As the manufacturer developed the product concept into a line of several incubators, we were able to design a master pattern from which all the molds could be easily produced.

Again, the process did not rely on smoke and mirrors; its success was the result of effective teamwork by professionals who were looking at the entire picture. From our perspective, you can't just throw a design at production and expect them to be able to build it cost-effectively. You have to consider design objectives concurrently with manufacturing methods and costs.

The next project was one of the first in which we were able to reap the benefits of 3-D solid modeling. It involved a new, low-cost liquid chromatography system, shown in Fig. 5. By using a team approach and advanced CAD tools, the manufacturer was able to launch the system in 18 months--half the usual development time for a new instrument--and at a fraction of the usual development cost.

A key to the development process was a dedicated team, working under a "product champion" or, in this case, a "benevolent dictator." The group included experts in electrical systems, software development, marketing, manufacturing, fluids/optics, and design. They refined the product concept by building a series of component boards that were tested by selected customers to determine optimum function and configuration. In short, they used the process to figure out what the customer wanted.

Next, outside design and tooling vendors were brought in to begin what the team leader described as "the process of making something out of nothing." Working with this team, we were able to go from soft mock-ups and appearance sketches to a working prototype in three months. This initial process established the system's functionality and manufacturability.

At this point, we were able to begin designing and engineering the coverset on 3-D CAD. This was the earliest that we had been able to use CAD in a program of this complexity, and it provided several advantages. For example, we were able to establish volumes for all the internal components--pumps, valves, fluids, and PC boards--and then design the bulkheads and coverset around them. As a result, we could constantly monitor interferences and tolerances of the assembly as we proceeded through the engineering phases.

The process required the manufacturer to make a very early commitment to vendors for tooling and molding. A number of advantages resulted: Not only were we able to electronically send the solid model to the molder to cut the tools, saving time and money, we were also able to use the data files to conduct mold flow and finite element analyses. The manufacturer also took our same 3-D model and downloaded it to create all their manufacturing assembly drawings, saving tens of thousands of dollars and a significant amount of time. This is important; once you've invested in creating the electronic model, you start to get multiple paybacks in manufacturing.

Next is an even more aggressive case of early vendor involvement, where both of our groups were involved very early in the program. Whereas the previous example involved a several-hundred-million-dollar-a-year organization, this case involved a start-up company, demonstrating that the tools of concurrent engineering apply to a wide range of applications and circumstances.

The application, shown in Fig. 6, is a wireless local area network (LAN) product. Initially, we used traditional industrial design tools, because they are still the best way to present or explore the widest variety of ideas in the shortest amount of time. The product uses a box, about 10 inches high, that sits with each group or cluster of PCs and can communicate to a hub location by means of a particular band of microwaves. The benefit is that if you want to rearrange your office, no rewiring is necessary.

Manufacturing volumes were extremely important to the design of the system. The transceivers--or boxes that sit by the computers--will sell in the tens of thousands, the hubs in the thousands. The hubs are all designed for plastics extrusion and RIM molded parts, whereas the transceivers are designed for injection molding. Again, the 3-D model included all the internal components as well as the detail for all external pieces.

The injection molding company became involved with this application early in the program. They were able to work with us to evaluate the overall design with respect to requirements for plastic resins, tooling, and moldability. Their involvement encompassed several areas. First, they suggested and implemented (with the resin supplier) mold flow analysis. This was done to confirm that the part would fill properly, using a center sprue gate configuration, and fill evenly through the vent louver area--a major concern because the gating approach to the part dictated the tooling. They also recommended a design using one part (rather than two different parts) that turned on itself to create the two halves of the housing. The design represented savings of at least $50,000 in tooling, a significant amount for a start-up company.

A couple of issues are apparent. By doing the model in 3-D solids, we were able to check all the fits and potential interferences in plastic to plastic. Thus, when the first sample parts came out of the tool, they were 100% correct. Normally, you need to allow tool margins on all areas where the snap-fits come together, and go through at least two or three rounds of tweaking the tool to get the fit just right. The 3-D solids tool helps us accomplish that with lower risk, lower cost, and a shorter lead time. The industrial design benefit of the computer link is that it permitted us to design shapes that would otherwise have been too complex for the application. By cutting the detailed shapes directly from our electronic file, we have more industrial design latitude. We can give our clients a more state-of-the-art product without negatively affecting the step that goes to the factory. That's a big advantage.

The beauty of 3-D CAD is that it enables you to create constantly changing graphic surfaces--compounded curves and the intersection of those compounded curvatures--to form a more uniformly blended surface. By electronically generating those surfaces, it permits the toolmaker to go in and cut immediately. In this case, we simply developed the tool path from the data. Although we weren't working with a very complicated set of surfaces, 3-D CAD permitted us to generate the tool path directly from the data files, rather than redevelop the geometry on our CAM system.

The other way we used the 3-D database is in the tool design itself. By taking that geometry into our CAD system, we do not have to redevelop the geometry of the part. Instead, we can easily go in and design a mold around the part, literally packaging the outside of the mold (with all its features) around the geometry of the plastic part. That shortens our tool design cycle. First, we use the database to do mold filling analysis, then to assist us in the tool design process, and, finally, to actually cut the mold by using the surfaces from which we develop the tool path.

Our success is attributable to two important facts. Because we were allowed to participate early in the program, we had the opportunity to give direction to the tooling and establish a plastic resin that would fill properly, with help from the material supplier. Second, this combination approach permitted us to build a tool quickly and meet an aggressive delivery schedule of eight weeks for tooling. More important, the first parts out of the press were perfect, with no revision, fixes, or corrections required to meet the exact end product.

That is what we are hearing universally from people who are using 3-D CAD--they get better quotes, shorter lead times, and fewer errors. They also are no longer told "this is how much your part costs are going to be, until we weigh one," because now they know, before the quote, what the part is going to weigh. With these tools, there is not much room for misinterpretation.

Change for the Better

From the cases we've discussed, we can begin to draw some general conclusions about the process. First, the idea of concurrent engineering or design for manufacturing is not pie in the sky; it's real, and based on solid methodology. Second, early vendor involvement is an absolute key to the process. As we have indicated, getting a team of professionals involved early pays enormous dividends. Third, advanced 3-D CAD tools are the catalyst for change in the process, because they permit the various players to work together effectively. Fourth, the use of the tools is also effecting additional changes in the process methodology. We are witnessing an evolution: Every step opens new possibilities and induces additional requirements.

There are a couple of ways in which we are seeing these changes at the molder's end of the process. One is an increased use of aluminum tooling. We find that aluminum permits additional flexibility with respect to mold complexity and enables us to achieve very rapid turnaround. Moreover, with the use of hard coating on these molds, we can still support large quantities (up to 100,000 pieces in some cases).

We are also seeing changes in the design end. We believe that higher and higher demands will be placed on industrial design as a discipline, and technologies are starting to compel such demands. More and more engineering work is being--and will continue to be--pushed upstream onto the designer.

Industrial design consultants can no longer be "pencil and wood shops." Instead, they will need to become "technically compatible vendors." That will become the expectation of our profession, and we will need to deliver on that level of performance. So, the burden falls on industrial designers to be much more skilled and capable. The benefit is that the concept work we do amounts to more than just pretty pictures; it ends up as manufactured products.

The process requires new tools and teamwork, and the tools and teams are, in turn, changing the process, making it more efficient and effective. Going back to our helix of Fig. 2, the process is becoming more organic.

There's one more thing we need to mention. People often ask, "Is this stuff faster and cheaper?" The answer is that it's better, but it's not faster or cheaper unless you look at the whole process. Seen incrementally, the costs of getting vendors involved early and the costs of the new technology can be substantial. But when you view the entire process of bringing products to market (and consider the effect of lessening the risks, optimizing the product, speeding up the process, and perhaps uncovering economies along the way), you see design for manufacturing and 3-D CAD as an investment that pays back many times over when the product goes to market.
COPYRIGHT 1993 Society of Plastics Engineers, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1993 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Concurrent Engineering
Author:Harkins, Jack R.; Dubreuil, Marc P.
Publication:Plastics Engineering
Date:Aug 1, 1993
Previous Article:Selecting materials for optimum performance.
Next Article:Medical plastics.

Terms of use | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters