Printer Friendly

Automated assembly - new game, new rules.

Automated assembly-new game, new rules

The past decade has seen a pronounced change--one that is evolutionary and ongoing--in the way American industry thinks about automation. Even the decision-making process has become more complex and sophisticated. Expectations of upper management and shareholders alike are, although not always reasonable, vastly elevated.

Catch-phrases such as factory of the future, universal automation, zero inventory, and zero defects flavor every discussion, and influence every decision relative to automation. While many of these concepts are misunderstood and poorly applied, they collectively represent a widely shared conviction that costs can be significantly reduced while quality is simultaneously improved. This can be accomplished, it is believed, through universal application of process control, coupled with creative management of limited resources.

Stated differently, we are automating in a world of limited capital. The judgment required in allocating scarce resources is still the hallmark of excellent management. Today, that management is informed and influenced by various factors, but still motivated by the age-old desire to survive and prosper in a highly competitive environment.

Automatic assembly and test is perhaps the most generalized field in automation, combining disciplines from nearly all other areas of automation. As a result of this jack-of-all-trades origientation, assembly and test provides a useful vantage point from which to study some of the changes in thinking about automation in general.

The old way

As recently as the early 1960's, there were no companies in the US, much less the world, dedicated to providing discrete product lines and engineering services for automatic assembly and test. Large and small machine-tool and specialty companies offered various combinations of incomplete services, but none could convincingly claim status as an "assembly house."

In fact, back then virtually all sophisticated assembly equipment was built in-house. None of today's successful builders had yet committed the required capital and engineering talent for what we now know as the automatic assembly business.

In comparison with todayhs methods, justification for automatic assembly was simple. During the 1960s and early 1970s, direct labor cost for assembly was the almost exclusive controlling factor. Only occasionally did one hear of other considerations such as reducing insurance costs, cutting inventories, or improving quality. At that time, the equation was far more basic: How long will it take to retrieve the investment in automation from savings in direct labor costs?

Accordingly, two threshold criteria had to be met when considering automatic assembly of a product. First, annual volume had to be high enough to make use of all or nearly all the capacity of assembly equipment, at least on a one-shift basis. Second, assurance of product life exceeding the equipment's payback period was required.

Most automotive components, for example, satisfied both criteria. This explains why carmakers have historically been heavy users of automatic assembly equipment. Conversely, toymakers frequently enjoy huge annual volumes, but almost never the assurance of long product life.

Vendors such as ourselves ordinarily are not privy to the exact formula a customer uses in determining payback period for a piece of equipment. It seems, however, that the most frequently mentioned period falls between one and two years.

Typically, automatic assembly equipment takes at least one year from the purchase-order date to the date of initial production runs. For a capital expenditure to be justified, thereofre, a mature product must have a reasonably guaranteed product-life exceeding three years from the assembly equipment's order date.

Is it doable?

Two decades ago, the technical feasibility of assembly automation could be considered only if adequate volume and product life were reasonably certain. Few products were designed for automatic assembly. Almost without exception, the designer's main concerns were product function and appearance.

Since production tooling for component parts often was already in place by the time automation of assembly was discussed, changes in part configuration or material selection to aid the assembly process were in most cases impossible to obtain. Usually engineers with the automation vendor were last to affect the process, and were at the mercy of all prior decisions.

The result? Frequently, automatic assembly was either not possible or highly inefficient, because irreversible decisions were made without regard to their impact on assembly viability. More subtly, because human beings are excellent "quality inspection devices," many companies are either unaware of, or unconcerned by, poor piece-part quality. The people manually assembling a product simply discarded bad parts with little or no lost total production. The only real cost was in some scrapped parts.

By contrast, automatic assembly systems do not merely discard bad parts. These systems either stop and wait for human assistance, or make a bad or partially finished product. In worst cases, they become jammed and require repair. Often, manufacturers do not adquately assess the quality of their component parts, and the impact poor part-quality would have on automated assembly.

In short, the assembly engineer historically had to work with the cards he was dealt. This meant that a lot of theoretically excellent applications for automatic assembly were not practical in the real world. A lack of timely input from assembly engineers resulted in many missed opportunities. Further, an unfocused approach to part quality brought many automatic assembly projects to an unsatisfactory end.

A new look at costing

The forces that changed American industry have affected economic justification of automation in general, and of assembly in particular. Where savings in direct labor was once almost the sole accounting measure by which return on investment could be measured, this savings category is now only one of several considerations.

A more broadly based management approach, together with sophisticated accounting methods, now routinely accounts for savings from reductions in inventory, work in process, scrap rates, process downtime and repairs, and warranty/product repairs. Though not as easily quantified the following factors are equally important:

* Heightened consumer confidence in product quality.

* Reductions in workers' chronic injuries caused by repetitive manual operations.

* Reduced exposure to product liability, made possible by better quality assurance.

* Marketing value of independent quality reports and analysis, such as that found in the magazine Consumer Reports.

As vendors, we routinly see automatic assembly and test programs implemented where direct-labor savings alone do not come close to justifying the capital expenditure. This is particularly true where the labor is offshore.

For the end-user of automatic assembly and test equipment, it is imperative that cross-disciplinary teams be created. In these teams, broad perspective can be achieved and maintaine during evaluation of the cost-benefit side of assembly automation.

The types of analysis suggested above do not constitute an exhaustive list. Rather, they suggest a thought process involved in a modern approach to cost-justification of automatic assembly and test.

Managers responsible for equipment acquisition, and the implementation of automation, must become involved in the accounting end of cost justification. Today, the big players have begun to do it, and for a good reason: The competition is, or soon will be, thinking this way.

The struggle for "sameness"

Many years ago, the firearms industry pioneered the concept of parts interchangeability on manufactured products. For over a century, it was enough if virtually all Part A's fit into virtually all Part B's as shown on an assembly drawing.

Now, though, in many industries it in understood that indistinguishability is the goal, not interchangeability. Furthermore, parts should be not merely indistinguishable, but optimally dimensioned for product performance and reliability.

This is where SPC--statistical process control--comes into play. This concept may have had a more profound impact on automatic assembly and test than any other in the past decade. SPC makes an impact mainly in two ways:

* Piece parts coming to the assembly machine are more nearly alike. That is, they are manufactured to much tighter tolerance specs than at any time in the past.

* Processes performed on the assembly equipment--e.g., soldering, crimping, fluid dispensing, measuring--are subjected to this statistical analysis.

For decades, vendors of automatic assembly equipment have championed the movement toward sameness in parts. We've done this for the most basic reason: Machines run better when parts are nearly identical. For many types of parts and products, however, controlling the process to obtain sameness of parts is either not practical or not cost-justifiable.

Products such as fuel injectors--by their nature extremely accurate, precise devices made up of precision components--are inherently well-suited to automatic assembly because of parts sameness. In the end, the more a manufacturer can justify putting a product's parts into process control, the more efficient will be assembly of those parts.

Similarly, sameness is important in processes to be performed in assembly machinery. These processes must be well-defined and proven before reliable assembly engineering can be done.

Typically, problems in this areas arise during projects being engineered simultaneously. In these cases, the product is being designed at the same time manufacturing processes and assembly engineering are being defined. Simultaneous engineering can be dangerous, however, if there is not a fully integrated, interdisciplinary team reviewing each step of the project as it progresses.

During analysis of a product's suitability for automatic assembly, processes such as welding, gluing, riveting, staking, sizing, and so on, must be given extremely close attention. The greatest potential dangers are in areas where the process itself is inherently difficult to control--for instance, resistance welding. Another risky situation is that in which materials to be processed may be changed along the way by product designers.

Experienced assembly engineers develop a sixth sense about these potential trouble spots, but cannot be asked to provide all solutions. Again, a collegial, interdisciplinary approach is the only way to minimize risks inherent in simultaneous engineering of product and process.

On mature products

If the product and process are fully developed and proven, but assembly methods are antiquated, different types of stumbling blocks to successful automation may arise. Some common barriers are:

1. Inability to feed or handle a part that is reliably oriented in a known position. Typically, this type of part is symmetrical on the outside but has critical, asymmetrical features on the inside. One example is a switch housing.

2. The fact that a process--for instance, leak testing, curing an adhesive, or running-in a motor--is so relatively time-consuming compared to other assembly or test processes required.

3. Governmental or private regulatory impediments, such as UL approval, which would have to be obtained from scratch if even a small change were made in the product or process.

4. Cosmetic concerns that limit the way a given part can be manipulated or processed.

5. Marketing concerns over customer recognition of, and comfort with, an existing product. Such concerns may generate resistance to changes that would enhance automated handling.

Today, these problem areas are addressed by assembly engineers far more readily than, say, five years ago. Steady progress in parts-handling technology, new sensing and positioning devices, new materials and adhesives, and advances in other fields--all provide approaches and solutions to what once were seemingly insolvable problems. In fact, many applications that were evaluated and rejected in the past can today be implemented efficiently, and should be reconsidered.

Start the team early

No single concept is more important to successful implementation of automatic assembly than that of using an interdisciplinary team approach to problem-solving from the very inception of the project. This holds true without exception in either assembly of a mature product, or introduction of a new one. Ideally, the team should include experienced professionals from product design, manufacturing, assembly, finance, and marketing.

Even if a user has assembly engineers on staff, the broader experience of a vendor's assembly engineers is usually helpful. In any case, the earlier the assembly engineering individual or group is included in the process, the better the end result.

Reputable vendors of automation are willing to review projects, without obligation, at the early stages. Managers responsible for implementing automatic assembly and test would do well to involve outside experts earlier in the decision-making process than has been common in the past.

Worldwide competitive pressures continue to transform the historic vendor/buyer relationship into more of a partnership. Successful automation managers will be those who take full, early advantage of available expertise, and work closely with vendors throughout the life of the project.
COPYRIGHT 1989 Nelson Publishing
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1989 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Bodine, David
Publication:Tooling & Production
Date:Sep 1, 1989
Previous Article:CAD/CAM optimizes punch-press tool performance.
Next Article:Monitoring quality welds at GM.

Related Articles
Finding flexibility in automated assembly.
Video makes it better.
Ten years after: CIM leaders revisited.
Two faces of Australia.
EDITORIAL\An imperial era ends\New rules reduce powers of the Assembly's speaker.
An alternative to cheap labor: low-cost regions play a major role, but they do not have all the answers.
Custom workcells for IML.
Perking up production: advances in technology are making custom automation and assembly more popular than ever, even for smaller product runs.

Terms of use | Copyright © 2016 Farlex, Inc. | Feedback | For webmasters