Volume 3 Mechanical Engineers’ Handbook Third Edition Manufacturing and Management doc

Additionally, DFM&A design programspromote team cooperation and supplier strategy and business considerations at an early stage in the product development process.. It minimizes the tota

Volume 3

Mechanical Engineers’ Handbook Third Edition

Manufacturing

and Management

Edited by Myer Kutz

JOHN WILEY & SONS, INC.

Copyright 䉷 2006 by John Wiley & Sons, Inc All rights reserved.

Published by John Wiley & Sons, Inc., Hoboken, New Jersey.

Published simultaneously in Canada.

No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form

or by any means, electronic, mechanical, photocopying, recording, scanning, or otherwise, except as permitted under Section 107 or 108 of the 1976 United States Copyright Act, without either the prior written permission of the Publisher, or authorization through payment of the appropriate per-copy fee

to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, (978) 750-8400, fax (978) 750-4470, or on the web at www.copyright.com Requests to the Publisher for permission should be addressed to the Permissions Department, John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ 07030, (201) 748-6011, fax (201) 748-6008, or online at http: / / www.wiley.com / go / permission.

Limit of Liability / Disclaimer of Warranty: While the publisher and author have used their best efforts

in preparing this book, they make no representations or warranties with respect to the accuracy or completeness of the contents of this book and specifically disclaim any implied warranties of

merchantability or fitness for a particular purpose No warranty may be created or extended by sales representatives or written sales materials The advice and strategies contained herein may not be suitable for your situation The publisher is not engaged in rendering professional services, and you should consult a professional where appropriate Neither the publisher nor author shall be liable for any loss of profit or any other commercial damages, including but not limited to special, incidental, consequential, or other damages.

For general information on our other products and services, please contact our Customer Care Department within the United States at (800) 762-2974, outside the United States at (317) 572-3993

or fax (317) 572-4002.

Wiley also publishes its books in a variety of electronic formats Some content that appears in print may not be available in electronic books For more information about Wiley products, visit our web site at www.wiley.com.

Library of Congress Cataloging-in-Publication Data:

Mechanical engineers’ handbook / edited by Myer Kutz.—3rd ed.

10 9 8 7 6 5 4 3 2 1

To Alan and Nancy, now and forever

Contents

Vision Statement xiContributors xiii

Bhaba R Sarker, Dennis B Webster, and Thomas G Ray

5 Production Processes and Equipment 173

Magd E Zohdi, William E Biles, and Dennis B Webster

6 Metal Forming, Shaping, and Casting 245

Magd E Zohdi and William E Biles

Murray J Roblin, updated by Anthony Luscher

8 Statistical Quality Control 315

Magd E Zohdi

9 Computer-Integrated Manufacturing 328

William E Biles and Magd E Zohdi

William E Biles, John S Usher, and Magd E Zohdi

11 Coatings and Surface Engineering: Physical Vapor Deposition 396

Allan Matthews and Suzanne L Rohde

12 Product Design and Manufacturing Processes for Sustainability 414

I S Jawahir, P C Wanigarathne, and X Wang

PART 2 MANAGEMENT, FINANCE, QUALITY, LAW,

18 Total Quality Management, Six Sigma, and Continuous Improvement 583

Jack B ReVelle and Robert Alan Kemerling

19 Registrations, Certifications, and Awards 616

Jack B ReVelle and Cynthia M Sabelhaus

Jack B ReVelle

21 What the Law Requires of the Engineer 701

Alvin S Weinstein, and Martin S Chizek

David A Burge and Benjamin D Burge

23 Electronic Information Resources: Your Online Survival Guide 758

Robert N Schwarzwalder, Jr.

24 Sources of Mechanical Engineering Information 777

Fritz Dusold and Myer Kutz

Preface

The third volume of the Third Edition of the Mechanical Engineers’ Handbook comprises

two parts: Manufacturing and Management Each part contains 12 chapters Contributorsinclude business owners, consultants, lawyers, librarians, and academics from all around theUnited States

Part 1 opens with a chapter from the second edition on Product Design for turing and Assembly (DFM&A) The centerpiece of Part 1 includes the chapters that inearlier editions of the handbook have been called ‘‘the handbook within the handbook.’’Developed by a team at Louisiana State University and the University of Louisville, thesesix chapters, which have been updated, span manufacturing topics from production planning,production processes and equipment, metal forming, shaping, and casting, statistical qualitycontrol, computer-integrated manufacturing, to material handling The chapter on classifi-cation systems remains unchanged from earlier editions; the chapter on mechanical fastenershas been revised extensively Part 1 has three chapters entirely new to the handbook: a chapter

Manufac-on physical vapor depositiManufac-on, Manufac-one Manufac-on envirManufac-onmentally cManufac-onscious manufacturing, and Manufac-one Manufac-on

a new approach to dealing with process technology in the context of design, tooling, ufacturing, and quality engineering The latter chapter is indicative of how much contributorscan give of themselves Its content is the lifeblood of its author’s consulting practice.Part 2 covers a broad array of topics The 12 chapters can be broken down into fourgroups The first two chapters cover project and people management The first of thesechapters, on project management, deals with a subject that has appeared in previous editions,but the chapter is entirely new, to reflect advances in this field The people managementchapter has been revised The following three chapters deal with fundamentals of financialmanagement and are unchanged The next three chapters, contributed by a team led by JackReVelle, treat a set of management issues, including Total Quality Management; registrations,certifications, and awards; and safety engineering Two chapters cover legal issues of interest

man-to engineers, including patents The final two chapters cover online and print informationsources useful to mechanical engineers in their daily work The chapter on online sources

is a new version of the chapter that appeared originally in 1998

Vision for the Third Edition

Basic engineering disciplines are not static, no matter how old and well established they are.The field of mechanical engineering is no exception Movement within this broadly baseddiscipline is multidimensional Even the classic subjects on which the discipline was founded,such as mechanics of materials and heat transfer, continue to evolve Mechanical engineerscontinue to be heavily involved with disciplines allied to mechanical engineering, such asindustrial and manufacturing engineering, which are also constantly evolving Advances inother major disciplines, such as electrical and electronics engineering, have significant impact

on the work of mechanical engineers New subject areas, such as neural networks, suddenlybecome all the rage

In response to this exciting, dynamic atmosphere, the Mechanical Engineers’ Handbook

is expanding dramatically, from one volume to four volumes The third edition not only isincorporating updates and revisions to chapters in the second edition, which was published

in 1998, but also is adding 24 chapters on entirely new subjects as well, incorporating updates

and revisions to chapters in the Handbook of Materials Selection, which was published in

2002, as well as to chapters in Instrumentation and Control, edited by Chester Nachtigal

and published in 1990

The four volumes of the third edition are arranged as follows:

Volume I: Materials and Mechanical Design—36 chapters

Part 1 Materials—14 chaptersPart 2 Mechanical Design—22 chapters

Volume II: Instrumentation, Systems, Controls, and MEMS—21 chapters

Part 1 Instrumentation—8 chaptersPart 2 Systems, Controls, and MEMS—13 chapters

Volume III: Manufacturing and Management—24 chapters

Part 1 Manufacturing—12 chaptersPart 2 Management, Finance, Quality, Law, and Research—12 chapters

Volume IV: Energy and Power—31 chapters

Part 1: Energy—15 chaptersPart 2: Power—16 chaptersThe mechanical engineering literature is extensive and has been so for a considerableperiod of time Many textbooks, reference works, and manuals as well as a substantialnumber of journals exist Numerous commercial publishers and professional societies, par-ticularly in the United States and Europe, distribute these materials The literature growscontinuously, as applied mechanical engineering research finds new ways of designing, con-trolling, measuring, making and maintaining things, and monitoring and evaluating technol-ogies, infrastructures, and systems

Most professional-level mechanical engineering publications tend to be specialized, rected to the specific needs of particular groups of practitioners Overall, however, the me-chanical engineering audience is broad and multidisciplinary Practitioners work in a variety

di-of organizations, including institutions di-of higher learning, design, manufacturing, and

con-sulting firms as well as federal, state, and local government agencies A rationale for anexpanded general mechanical engineering handbook is that every practitioner, researcher,and bureaucrat cannot be an expert on every topic, especially in so broad and multidiscipli-nary a field, and may need an authoritative professional summary of a subject with which

he or she is not intimately familiar

Starting with the first edition, which was published in 1986, our intention has always

been that the Mechanical Engineers’ Handbook stand at the intersection of textbooks,

re-search papers, and design manuals For example, we want the handbook to help youngengineers move from the college classroom to the professional office and laboratory wherethey may have to deal with issues and problems in areas they have not studied extensively

in school

With this expanded third edition, we have produced a practical reference for the chanical engineer who is seeking to answer a question, solve a problem, reduce a cost, orimprove a system or facility The handbook is not a research monograph The chapters offerdesign techniques, illustrate successful applications, or provide guidelines to improving theperformance, the life expectancy, the effectiveness, or the usefulness of parts, assemblies,and systems The purpose is to show readers what options are available in a particularsituation and which option they might choose to solve problems at hand

me-The aim of this expanded handbook is to serve as a source of practical advice to readers

We hope that the handbook will be the first information resource a practicing engineerconsults when faced with a new problem or opportunity—even before turning to other printsources, even officially sanctioned ones, or to sites on the Internet (The second edition hasbeen available online on knovel.com.) In each chapter, the reader should feel that he or she

is in the hands of an experienced consultant who is providing sensible advice that can lead

to beneficial action and results

Can a single handbook, even spread out over four volumes, cover this broad,

interdis-ciplinary field? We have designed the third edition of the Mechanical Engineers’ Handbook

as if it were serving as a core for an Internet-based information source Many chapters inthe handbook point readers to information sources on the Web dealing with the subjectsaddressed Furthermore, where appropriate, enough analytical techniques and data are pro-vided to allow the reader to employ a preliminary approach to solving problems

The contributors have written, to the extent their backgrounds and capabilities makepossible, in a style that reflects practical discussion informed by real-world experience Wewould like readers to feel that they are in the presence of experienced teachers and con-sultants who know about the multiplicity of technical issues that impinge on any topic withinmechanical engineering At the same time, the level is such that students and recent graduatescan find the handbook as accessible as experienced engineers

Contributors

Dell K AllenBrigham Young UniversityProvo, Utah

William E BilesUniversity of LouisvilleLouisville, KentuckyWilliam BrettNew York, New YorkBenjamin D BurgeIntel Americas, Inc

Chantilly, VirginiaDavid A BurgeDavid A Burge Co., L.P.A

Cleveland, OhioMartin S ChizekWeinstein Associates InternationalDelray Beach, Florida

Fritz DusoldNew York, New YorkKaren L HigginsNAVAIR Weapons DivisionChina Lake, California

I S JawahirUniversity of KentuckyLexington, KentuckyByron W JonesKansas State UniversityManhattan, KansasRobert Alan KemerlingEthicon Endo-Surgery, Inc

Cincinnati, Ohio

Myer KutzMyer Kutz Associates, Inc

Delmar, New YorkGordon LewisDigital Equipment CorporationMaynard, MassachusettsAnthony LuscherThe Ohio State UniversityColumbus, Ohio

Joseph A MaciarielloClaremont Graduate UniversityClaremont, California

Allan MatthewsSheffield UniversitySheffield, United KingdomThomas G Ray

Louisiana State UniversityBaton Rouge, LouisianaJack B RevelleReVelle Solutions, LLCSanta Ana, CaliforniaMurray J RoblinCalifornia State Polytechnic UniversityPomona, California

Suzanne L RohdeThe University of NebraskaLincoln, Nebraska

Cynthia M SabelhausRaytheon Missile SystemsTucson, Arizona

Bhaba R SarkerLouisiana State UniversityBaton Rouge, Louisiana

Los Angeles, CaliforniaJohn S Usher

University of LouisvilleLouisville, Kentucky

X WangUniversity of KentuckyLexington, Kentucky

P C WanigarathneUniversity of KentuckyLexington, KentuckyDennis B WebsterLouisiana State UniversityBaton Rouge, LouisianaAlvin S WeinsteinWeinstein Associates InternationalDelray Beach, Florida

Magd E ZohdiLouisiana State UniversityBaton Rouge, Louisiana

Mechanical Engineers’ Handbook

PART 1

MANUFACTURING

2.1 What Is DFM&A? 5 2.2 Getting the DFM&A

to control for most organizations There are some experts who openly say that if we have

no new technology for the next five years, corporate America might just start to catch up.The key to achieving benchmark time to market, cost, and quality is in up-front technology,engineering, and design practices that encourage and support a wide latitude of new productdevelopment processes These processes must capture modern manufacturing technologies,piece parts that are designed for ease of assembly, and parts that can be fabricated usinglow-cost manufacturing processes Optimal new product design occurs when the designs ofmachines and of the manufacturing processes that produce those machines are congruent.The obvious goal of any new product development process is to turn a profit by con-verting raw material into finished products This sounds simple, but it has to be done effi-ciently and economically Many companies do not know how much it costs to manufacture

a new product until well after the production introduction Rule #1: The product development

team must be given a cost target at the start of the project We will call this cost the unit manufacturing cost (UMC) target Rule #3: The product development team must be held

accountable for this target cost What happened to rule #2? We’ll discuss that shortly In themeantime, we should understand what UMC is

where BL ⫽burdened assembly labor rate per hour; this is the direct labor cost of labor,

benefits, and all appropriate overhead cost

MC⫽material cost; this is the cost of all materials used in the product

Mechanical Engineers’ Handbook: Manufacturing and Management, Volume 3, Third Edition.

Edited by Myer Kutz Copyright  2006 by John Wiley & Sons, Inc.

TA⫽tooling amortization; this is the cost of fabrication tools, molds and assemblytooling, divided by the forecast volume build of the product

UMC is the direct burdened assembly labor (direct wages, benefits, and overhead) plus thematerial cost Material cost must include the cost of the transformed material plus piece partpackaging plus duty, insurance, and freight (DIF) Tooling amortization should be included

in the UMC target cost calculation, based on the forecast product life volume

Example UMC Calculation BL ⴙMCⴙTA

Burdened assembly labor cost calculation (BL)

Labor

BL⫽($18.75⫹138%)⫽$44.06 / hr

Wages ⫹Benefits overheadBurdened assembly labor is made up of the direct wages and benefits paid to the hourlyworkers, plus a percentage added for direct overhead and indirect overhead The overheadadded percentage will change from month to month based on plant expenses

Material cost calculation (MC )

(Part cost ⫹ Packaging) ⫹ DIF ⫹ Mat Acq cost ⫽

Material FOB assm plantMaterial cost should include the cost of the parts and all necessary packaging This calcu-lation should also include a percent adder for duty, insurance, and freight (DIF) and an adderfor the acquisition of the materials (Mat Acq.) DIF typically is between 4 and 12% andMat Acq typically is in the range of 6 to 16% It is important to understand the MC becausematerial is the largest expense in the UMC target

Tooling amortization cost calculations (TA)

(Tool cost) # of parts

TA⫽$56,000 / 10,000⫽$5.60 per assembly

TC is the cost of tooling and PL is the estimated number of parts expected to be produced

on this tooling Tooling cost is the total cost of dies and mold used to fabricate the componentparts of the new product This also should include the cost of plant assembly fixtures andtest and quality inspection fixtures

The question is, ‘‘How can the product development team quickly and accurately sure UMC during the many phases of the project?’’ What is needed is a tool that providesinsight into the product structure and at the same time exposes high-cost areas of the design

Designing for Manufacturing and Assembly (DFM&A) is a technique for reducing the cost

of a product by breaking the product down into its simplest components All members ofthe design team can understand the product’s assembly sequence and material flow early inthe design process

DFM&A tools lead the development team in reducing the number of individual partsthat make up the product and ensure that any additional or remaining parts are easy to handle

2 Design for Manufacturing and Assembly 5

and insert during the assembly process DFM&A encourages the integration of parts andprocesses, which helps reduce the amount of assembly labor and cost DFM&A effortsinclude programs to minimize the time it takes for the total product development cycle,manufacturing cycle, and product life-cycle costs Additionally, DFM&A design programspromote team cooperation and supplier strategy and business considerations at an early stage

in the product development process

The DFM&A process is composed of two major components: design for assembly (DFA) and design for manufacturing (DFM) DFA is the labor side of the product cost This is the

labor needed to transform the new design into a customer-ready product DFM is the materialand tooling side of the new product DFM breaks the parts fabrication process down into itssimplest steps, such as the type of equipment used to produce the part and fabrication cycletime to produce the part, and calculates a cost for each functional step in the process Theprogram team should use the DFM tools to establish the material target cost before the newproduct design effort starts

Manufacturing costs are born in the early design phase of the project Many differentstudies have found that as much as 80% of a new product’s cost is set in concrete at thefirst drawing release phase of the product Many organizations find it difficult to implementchanges to their new product development process The old saying applies: ‘‘only wet babieswant to change, and they do it screaming and crying.’’ Figure 1 is a memo that was actuallycirculated in a company trying to implement a DFM&A process Only the names have beenchanged

It is clear from this memo that neither the engineering program manager nor the ufacturing program manager understood what DFM&A was or how it should be implemented

man-in the new product development process It seems that their defman-inition of concurrent neering is, ‘‘Engineering creates the design and manufacturing is forced to concur with itwith little or no input.’’ This is not what DFM&A is

engi-2.1 What Is DFM&A?

DFM&A is not a magic pill It is a tool that, when used properly, will have a profoundeffect on the design philosophy of any product The main goal of DFM&A is to lowerproduct cost by examining the product design and structure at the early concept stages of anew product DFM&A also leads to improvements in serviceability, reliability, and quality

of the end product It minimizes the total product cost by targeting assembly time, part cost,and the assembly process in the early stages of the product development cycle

The life of a product begins with defining a set of product needs, which are thentranslated into a set of product concepts Design engineering takes these product conceptsand refines them into a detailed product design Considering that from this point the productwill most likely be in production for a number of years, it makes sense to take time outduring the design phase to ask, ‘‘How should this design be put together?’’ Doing so willmake the rest of the product life, when the design is complete and handed off to productionand service, much smoother To be truly successful, the DFM&A process should start at theearly concept development phase of the project True, it will take time during the hecticdesign phase to apply DFM&A, but the benefits easily justify additional time

DFM&A is used as a tool by the development team to drive specific assembly benefitsand identify drawbacks of various design alternatives, as measured by characteristics such

as total number of parts, handling and insertion difficulty, and assembly time DFM&Aconverts time into money, which should be the common metric used to compare alternativedesigns, or redesigns of an existing concept The early DFM&A analysis provides the product

Memorandum: Ajax Bowl Corporation

DATE: January 26, 1997

TO: Manufacturing Program Manager, Auto Valve Project

FROM: Engineering Program Manager, Auto Valve Project

RE: Design for Manufacturing & Assembly support for Auto Valve Project

CC: Director, Flush Valve Division

Due to the intricate design constraints placed on the Auto Valve project engineering feels they will not have the resources

to apply the Design for Manufacturing and Assembly process Additionally, this program is strongly schedule driven The budget for the project is already approved as are other aspects of the program that require it to be on-time in order to achieve the financial goals of upper management.

In the meeting on Tuesday, engineering set down the guidelines for manufacturing involvement on the Auto Valve project This was agreed to by several parties (not manufacturing) at this meeting.

The manufacturing folks wish to be tied early into the Auto Valve design effort:

1 This will allow manufacturing to be familiar with what is coming.

2 Add any ideas or changes that would reduce overall cost or help schedule.

3 Work vendor interface early, manufacturing owns the vendor issues when the product comes to the plant, anyways Engineering folks like the concept of new ideas, but fear:

1 Inputs that get pushed without understanding of all properly weighted constraints.

2 Drag on schedule due to too many people asking to change things.

3 Spending time defending and arguing the design.

PROPOSAL—Turns out this is the way we will do it.

Engineering shall on a few planned occasions address manufacturing inputs through one manufacturing person Most correspondence will be written and meeting time will be minimal It is understood that this program is strongly driven by schedule, and many cost reduction efforts are already built into the design so that the published budget can be met The plan for Engineering:

● When drawings are ready, Engineering Program Manager (EPM) will submit them to Manufacturing Program Manager

(MPM).

● MPM gathers inputs from manufacturing people and submits them back in writting to EPM MPM works questions

through EPM to minimize any attention units that Engineering would have to spend.

● EPM submits suggestions to Engineering, for one quick hour of discussion/acceptance/veto.

● EPM submits written response back to MPM and works any Design continues under ENG direction.

● When a prototype parts arrives, the EPM will allow the MPM to use it in manufacturing discussions.

● MPM will submit written document back to EPM to describe issues and recommendations.

● Engineering will incorporate any changes that they can handle within the schedule that they see fit.

Figure 1

development team with a baseline to which comparisons can be made This early analysiswill help the designer to understand the specific parts or concepts in the product that requirefurther improvement, by keeping an itemized tally of each part’s effect on the whole assem-bly Once a user becomes proficient with a DFM&A tool and the concepts become secondnature, the tool is still an excellent means of solidifying what is by now second nature toDFA veterans, and helps them present their ideas to the rest of the team in a commonlanguage: cost

DFM&A is an interactive learning process It evolves from applying a specific method

to a change in attitude Analysis is tedious at first, but as the ideas become more familiarand eventually ingrained, the tool becomes easier to use and leads to questions: questionsabout the assembly process and about established methods that have been accepted or ex-isting design solutions that have been adopted In the team’s quest for optimal design so-lutions, the DFM&A process will lead to uncharted ways of doing things Naturally, then,

2 Design for Manufacturing and Assembly 7

Figure 2 Key components of the DFM&A process.

the environment in which DFA is implemented must be ripe for challenging pat solutionsand making suggestions for new approaches This environment must evolve from the topdown, from upper management to the engineer Unfortunately, this is where the process toooften fails

Figure 2 illustrates the ideal process for applying DFM&A The development of anynew product must go through four major phases before it reaches the marketplace: concept,design, development, and production In the concept phase, product specifications are createdand the design team creates a design layout of the new product At this point, the first designfor assembly analysis should be completed This analysis will provide the design team with

a theoretical minimum parts count and pinpoint high-assembly areas in the design

At this point, the design team needs to review the DFA results and adjust the designlayout to reflect the feedback of this preliminary analysis The next step is to complete adesign for manufacturing analysis on each unique part in the product This will consist ofdeveloping a part cost and tooling cost for each part It should also include doing a produc-ibility study of each part Based on the DFM analysis, the design team needs to make some

2 end plate screws

motor

2 stand-offs

2 bushings

2 motor screws

sensor base set screw

cover

4 cover screws

Figure 3 Proposed motor drive assembly (From Ref 1.)

additional adjustments in the design layout At this point, the design team is now ready tostart the design phase of the project The DFM&A input at this point has developed apreliminary bill of material (BOM) and established a target cost for all the unique new parts

in the design It has also influenced the product architecture to improve the sequence ofassembly as it flows through the manufacturing process

The following case study illustrates the key elements in applying DFM&A Figure 3

shows a product called the motor drive assembly This design consists of 17 parts and

assemblies Outwardly it looks as if it can be assembled with little difficulty The product ismade up of two sheet metal parts and one aluminum machined part It also has a motor

assembly and a sensor, both bought from an outside supplier In addition, the motor drive assembly has nine hardware items that provide other functions—or do they?

At this point, the design looks simple enough It should take minimal engineering effort

to design and detail the unique parts and develop an assembly drawing Has a UMC beendeveloped yet? Has a DFM&A analysis been performed? The DFA analysis will look ateach process step, part, and subassembly used to build the product It will analyze the time

it takes to ‘‘get’’ and ‘‘handle’’ each part and the time it takes to insert each part in theassembly (see Table 1) It will point out areas where there are difficulties handling, aligning,and securing each and every part and subassembly The DFM analysis will establish a costfor each part and estimate the cost of fabrication tooling The analysis will also point outhigh-cost areas in the fabrication process so that changes can be made

At this point, the DFA analysis suggested that this design could be built with fewerparts A review of Table 2, column 5, shows that the design team feels it can eliminate thebushings, stand-offs, end-plate screws, grommet, cover, and cover screws Also by replacingthe end plate with a new snap-on plastic cover, they can eliminate the need to turn the

2 Design for Manufacturing and Assembly 9

Table 1

Motor Drive Assembly Number of parts and assemblies 19 Number of reorientation or adjustment 1 Number of special operations 2 Total assembly time in seconds 213.4 Total cost of fabrication and assembly tooling $3590 Tool amortization at 10K assemblies $ 0.36 Total cost of labor at $74.50 / hr $ 4.42 Total cost of materials $ 42.44 Total cost of labor and materials $ 46.86

DFM Analysis

The DFM analysis provided the input for the fabricated part cost As an example, the base

is machined from a piece of solid aluminum bar stock As designed, the base has 11 differentholes drilled in it and 8 of them require taping The DFM analysis (see Table 3) shows that

it takes 17.84 minutes to machine this part from the solid bar stock The finished machinedbase costs $10.89 in lots of 1000 parts The ideal process for completing a DFM analysismight be as follows

In the case of the base, the design engineer created the solid geometry in Matra Data’sEuliked CAD system (see Fig 4) The design engineer then sent the solid database as anSTL file to the manufacturing engineer, who then brought the STL file into a viewing tool

called Solid View (see Fig 5) SolidView allowed the ME to get all the dimensioning and

geometry inputs needed to complete the Boothroyd Dewhurst design for manufacturing chining analysis of the base part SolidView also allowed the ME to take cut sections of thepart and then step through it to insure that no producibility rules had been violated.Today all of the major CAD supplies provide the STL file output format There aremany new CAD viewing tools like SolidView available, costing about $500 to $1000 Theseviewing tools will take STL or IGS files The goal is to link all of the early product devel-opment data together so each member can have fast, accurate inputs to influence the design

ma-in its earliest stage

In this example, it took the ME a total of 20 minutes to pull the STL files into SolidViewand perform the DFM analysis Engineering in the past has complained that DFM&A takestoo much time and slows the design team down The ME then analyzes the base as a diecasting part, following the producibility rule By designing the base as a die casting, it ispossible to mold many of the part features into the part This net shape die cast design willreduce much of the machining that was required in the original design The die cast partwill still require some machining The DFM die casting analysis revealed that the base

2 Design for Manufacturing and Assembly 11

Table 3 Machining Analysis Summary Report Setups

Time Minutes

Product life volume 10,000 Number of machine tool setups 3 Number of library operation setups 1 Workpiece weight, lb 0.85 Workpiece volume, cu in 8.80 Material density, lb / cu in 0.097

casting would cost $1.41 and the mold would cost $9050 Table 4 compares the two differentfabrication methods

This early DFM&A analysis provides the product development team with accurate laborand material estimates at the start of the project It removes much of the complexity of theassembly and allows each member of the design team to visualize every component’s func-tion By applying the basic principles of DFA, such as

• Combining or eliminating parts

• Eliminating assembly adjustments

• Designing part with self-locating features

• Designing parts with self-fastening features

• Facilitating handling of each part

• Eliminating reorientation of the parts during assembly

• Specifying standard parts

Figure 4

the design team is able to rationalize the motor drive assembly with fewer parts and assemblysteps Figure 6 shows a possible redesign of the original motor drive assembly The DFM&Aanalysis (Table 5) provided the means for the design team to question the need and function

of every part As a result, the design team now has a new focus and an incentive to changethe original design

Table 6 shows the before-and-after DFM&A results

If the motor drive product meets its expected production life volume of 10,000 units,the company will save $170,100 By applying principles of DFM&A to both the labor andmaterial on the motor drive, the design team is able to achieve about a 35% cost avoidance

on this program

2.2 Getting the DFM&A Process Started

Management from All of the Major Disciplines Must Be on Your Side

In order for the DFM&A process to succeed, upper management must understand, accept,

and encourage the DFM&A way of thinking They must want it It is difficult, if not

im-possible, for an individual or group of individuals to perform this task without managementsupport, since the process requires the cooperation of so many groups working together Thebiggest challenge of implementing DFM&A is the cooperation of so many individuals to-wards a common goal This does not come naturally, especially if it is not perceived by theleaders as an integral part of the business’s success In many companies, management doesnot understand what DFM&A is They believe it is a manufacturing process It is not; it is

a new product development process, which must include all disciplines (engineering, service,

program managers, and manufacturing) to yield significant results The simplest method toachieve cooperation between different organizations is to have the team members work in acommon location (co-located team) The new product development team needs some nur-

2 Design for Manufacturing and Assembly 13

Figure 5

Table 4

Die Cast and Machined

Machined from Bar Stock

$9050 die casting tooling / 10,000 $0.91 Machining time, min 3.6 17.84 Machining cost $3.09 $8.55

turing and stimulation to become empowered This is an area where most companies justdon’t understand the human dynamics of building a high-performance team Table 7 shouldaid in determining whether you are working in a team environment or a work group envi-ronment

Many managers will say that their people are working in a team environment, but theystill want to have complete control over work assignments and time spent supporting theteam In their mind, the team’s mission is secondary to the individual department manager’s

cover motor

2 motor screws

sensor

set screw

base

Figure 6 Redesign of motor assembly.

goals This is not a team; it is a work group The essential elements of a high-performanceteam are

• A clear understanding of the team’s goals (a defined set of goals and tasks assigned

to each individual team member)

• A feeling of openness, trust, and communication

• Shared decision-making (consensus)

• A well-understood problem-solving process

• A leader who legitimizes the team-building processManagement must recognize that to implement DFM&A in their organization, they must beprepared to change the way they do things Management’s reluctance to accept the need forchange is one reason DFM&A has been so slow to succeed in many companies Training isone way of bringing DFM&A knowledge to an organization, but training alone cannot beexpected to effect the change

2.3 The DFM&A Road Map

The DFM&A Methodology (A Product Development Philosophy)

• Form a multifunctional team

• Establish the product goals through competitive benchmarking

Table 6 Comparison of DFM&A Results

Motor Drive Assembly

Redesign of Motor Drive Assembly

Number of reorientation or adjustment 1 0

Total assembly time in seconds 213.4 86.34 Total cost of labor at $74.50 / hr $4.42 $1.79

Total cost of labor and material $46.86 $28.42 Total cost of fabrication tooling $3590 $17,888 Tool amortization at 10K assemblies $0.36 $1.79

Savings ⫽ $17.01

Table 7 Human Factors Test (check the box where your team fits) Yes Team Environment Yes Work Group Environment

Are the team members committed

to the group’s common goals?

 Are members loyal to outside

groups with conflicting interests (functional managers)?

Is there open communication with all members of the team?

 Is information unshared?

Is there flexible, creative leadership?

 Is there a dominating leadership?

Is the team rewarded as a group?  Is there individual recognition?

Is there a high degree of confidence and trust between members?

 Are you unsure of the group’s

authority?

• Perform a design for assembly analysis

• Segment the product into manageable subassemblies or levels of assembly

• As a team, apply the design for assembly principles

• Use creativity techniques to enhance the emerging design

• As a team, evaluate and select the best ideas

• Ensure economical production of every piece part

• Establish a target cost for every part in the new design

• Start the detailed design of the emerging product

• Apply design for producibility guidelines

• Reapply the process at the next logical point in the design

• Provide the team with a time for reflection and sharing results

2 Design for Manufacturing and Assembly 17

This DFM&A methodology incorporates all of the critical steps needed to insure a successfulimplementation

Develop a multifunctional team of all key players before the new product architecture is defined This team must foster a creative climate that will encourage ownership of the new product’s design The first member of this team should be the project leader, the person who

has the authority for the project This individual must control the resources of the zation, should hand-pick the people who will work on the team, and should have the authority

organi-to resolve problems within the team

The team leader should encourage and develop a creative climate It is of utmost portance to assemble a product development team that has the talent to make the rightdecisions, the ability to carry them out, and the persistence and dedication to bring theproduct to a successful finish Although these qualities are invaluable, it is of equal impor-tance that these individuals be allowed as much freedom as possible to germinate creativesolutions to the design problem as early as possible in the product design cycle

im-The product development team owns the product design and the development process.The DFM&A process is most successful when implemented by a multifunctional team, whereeach person brings to the product design process his or her specific area of expertise Theteam should embrace group dynamics and the group decision-making process for DFM&A

to be most effective

Emphasis has traditionally been placed on the design team as the people who drive andown the product Designers need to be receptive to team input and share the burden of thedesign process with other team members

The team structure depends on the nature and complexity of the product Disciplinesthat might be part of a product team include

Clearly, there can be drawbacks to multidisciplinary teams, such as managing too manyopinions, difficulty in making decisions, and factors in general that could lengthen the prod-uct development cycle However, once a team has worked together and has an understanding

of individual responsibilities, there is much to gain from adopting the team approach Groupsworking together can pool their individual talents, skills, and insight so that more resourcesare brought to bear on a problem Group discussion leads to a more thorough understanding

of problems, ideas, and potential solutions from a variety of standpoints Group making results in a greater commitment to decisions, since people are more motivated tosupport and carry out a decision that they helped make Groups allow individuals to improveexisting skills and learn new ones

decision-Having the team located together in one facility makes the process work even better.This co-location improves the team’s morale and also makes communication easier Remem-bering to call someone with a question, or adding it to a meeting agenda, is more difficultthan mentioning it when passing in the hallway Seeing someone reminds one of an issuethat may have otherwise been forgotten These benefits may seem trivial, but the differencethat co-location makes is significant

As a team, establish product goals through a competitive benchmarking process: concept development Competitive benchmarking is the continuous process of measuring your own

products, services, and practices against the toughest competition, or the toughest competition

in a particular area The benchmarking process will help the team learn who the ‘‘best’’ areand what they do It gives the team a means to understand how this new product measures

up to other products in the marketplace It identifies areas of opportunities that need changing

in the current process It allows the team to set targets and provides an incentive for change.Using a DFM&A analysis process for the competitive evaluation provides a means for rel-ative comparison between those of your products and those of your competitors You deter-mine exactly where the competition is better

Before performing a competitive teardown, decide on the characteristics that are mostimportant to review, what the group wants to learn from the teardown, and the metrics thatwill be noted Also keep the teardown group small It’s great to have many people walkthrough and view the results, but a small group can better manage the initial task of dis-assembly and analysis Ideally, set aside a conference room for several days so the productcan be left out unassembled, with a data sheet and metrics available

Perform a design for assembly analysis of the proposed product that identifies possible candidate parts for elimination or redesign and pinpoints high-cost assembly opera- tions Early in this chapter, the motor drive assembly DFM&A analysis was developed This

example illustrates the importance of using a DFA tool to identify, size, and track the savings opportunities This leads to an important question: Do you need a formal DFAanalysis software tool? Some DFM&A consultants will tell you that it is not necessary touse a formal DFA analysis tool It is my supposition that these consultants want to sell you

cost-consulting services rather than teach the process It just makes no sense not to use a formal

DFA tool for evaluating and tracking the progress of the new product design through itsevolution The use of DFA software provides the team with a focus that is easily updated

as design improvements are captured The use of DFA software does not exclude the needfor a good consultant to get the new team off to a good start The selection of a DFA tool

is a very important decision The cost of buying a quality DFA software tool is easily justified

by the savings from applying the DFA process on just one project

At this point, the selection of the manufacturing site and type of assembly process should

be completed Every product must be designed with a thorough understanding of the bilities of the manufacturing site It is thus of paramount importance to choose the manu-facturing site at the start of product design This is a subtle point that is frequently overlooked

capa-at the start of a program, but to build a partnership with the manufacturing site, the siteneeds to have been chosen! Also, manufacturing facilities have vastly different processes,capabilities, strengths, and weaknesses, that affect, if not dictate, design decisions Whenselecting a manufacturing site, the process by which the product will be built is also beingdecided

As a team, apply the design for assembly principles to every part and operation to generate

a list of possible cost opportunities The generic list of DFA principles includes the

follow-ing:

• Designing parts with self-locating features

• Designing parts with self-fastening features

• Increasing the use of multifunctional parts

• Eliminating assembly adjustments

• Driving standardization of fasteners, components, materials, finishes, and processes

2 Design for Manufacturing and Assembly 19

Table 8 DFM&A Metrics

Old Design Competitive New Design Number of Parts & Assemblies

Number of Separate Assm Operations

Total Assembly Time

Total Material Cost

Totals

It is important for the team to develop its own set of DFA principles that relate to the specificproduct it is working on Ideally, the design team decides on the product characteristics itneeds to meet based on input from product management and marketing The product defi-nition process involves gathering information from competitive benchmarking and teardowns,customer surveys, and market research Competitive benchmarking illustrates which productcharacteristics are necessary

Principles should be set forth early in the process as a contract that the team draws uptogether It is up to the team to adopt many principles or only a few, and how lenient to be

Provide the team with a time for reflection and sharing results Each team member needs

to understand that there will be a final review of the program, at which time members will

be able to make constructive criticism This time helps the team determine what worked andwhat needs to be changed in the process

Use DFM&A Metrics

The development of some DFM&A metrics is important The team needs a method to sure the before-and-after results of applying the DFM&A process, thus justifying the timespent on the project Table 8 shows the typical DFM&A metrics that should be used tocompare your old product design against a competitive product and a proposed new redesign.The total number of parts in an assembly is an excellent and widely used metric If thereader remembers only one thing from this chapter, let it be to strive to reduce the quantity

mea-of parts in every product designed The reason limiting parts count is so rewarding is thatwhen parts are reduced, considerable overhead costs and activities that burden that part alsodisappear When parts are reduced, quality of the end product is increased, since each partthat is added to an assembly is an opportunity to introduce a defect into the product Totalassembly time will almost always be lowered by reducing the quantity of parts

Table 9 DFM&A New Products Checklist Design for Manufacturing and Assembly Consideration Yes No Design for assembly analysis completed:  

Has this design been analyzed for minimal part count?  

Have all adjustments been eliminated?  

Are more than 85% common parts and assemblies used in this design?  

Have assembly and part reorientations been minimized?  

Have more than 96% preferred screws been used in this design?  

Have all parts been analyzed for ease of insertion during assembly?  

Have all assembly interferences been eliminated?  

Have location features been provided?  

Have all parts been analyzed for ease of handling?  

Have part weight problems been identified?  

Have special packaging requirements been addressed for problem parts?  

Are special tools needed for any assembly steps?  

Does design capitalize on self-alignment features of mating parts?  

Have limited physical and visual access conditions been avoided?  

Does design allow for access of hands and tools to perform necessary assembly steps?

Has adequate access been provided for all threaded fasteners and drive tooling?  

Have all operator hazards been eliminated (sharp edges)?  

Has adequate panel pass-through been provided to allow for easy harness / cable routing?

Have harness / cable supports been provided?  

Have keyed connectors been provided at all electrical interconnections?  

Are all harnesses / cables long enough for ease of routing, tie down, plug in, and to eliminate strain relief on interconnects?

Does design allow for access of hands and tools to perform necessary wiring operations?

Does position of cable / harness impede air flow?  

Design for Manufacturing and Considerations Yes No Have all unique design parts been analyzed for producibility?  

Have all unique design parts been analyzed for cost?  

Have all unique design parts been analyzed for their impact of tooling / mold cost?

Has assembly tryout been performed prior to scheduled prototype build?  

Have assembly views and pictorial been provided to support assembly documentation?

Has opportunity defects analysis been performed on process build?  

Has products cosmetics been considered (paint match, scratches)?  

References 21

A simple method to test for potentially unnecessary parts is to ask the following threequestions for each part in the assembly:

1 During the products operation, does the part move relative to all other parts already

assembled? (answer yes or no)

2 Does the part need to be made from a different material or be isolated from all other

parts already assembled? (answer yes or no)

3 Must the part be separate from all other parts already assembled because of necessary

assembly or disassembly of other parts? (answer yes or no)

You must answer the questions above for each part in the assembly If your answer is ‘‘no’’for all three questions, then that part is a candidate for elimination

The total time it takes to assemble a product is an important DFM&A metric Time ismoney, and the less time needed to assemble the product, the better Since some of the mosttime-consuming assembly operations are fastening operations, discrete fasteners are alwayscandidates for elimination from a product By examining the assembly time of each andevery part in the assembly, the designer can target specific areas for improvement Totalmaterial cost is self-explanatory

The new product DFM&A checklist (Table 9) is a good review of how well your teamdid with applying the DFM&A methodology Use this check sheet during all phases of theproduct development process; it is a good reminder At the end of the project you should

have checked most of the yes boxes.

3 WHY IS DFM&A IMPORTANT?

DFM&A is a powerful tool in the design team’s repertoire If used effectively, it can yieldtremendous results, the least of which is that the product will be easy to assemble! The mostbeneficial outcome of DFM&A is to reduce part count in the assembly, which in turn willsimplify the assembly process, lower manufacturing overhead, reduce assembly time, andincrease quality by lessening the opportunities for introducing a defect Labor content is alsoreduced because with fewer parts, there are fewer and simpler assembly operations Anotherbenefit to reducing parts count is a shortened product development cycle because there arefewer parts to design The philosophy encourages simplifying the design and using standard,off-the-shelf parts whenever possible In using DFM&A, renewed emphasis is placed ondesigning each part so it can be economically produced by the selected manufacturing proc-ess

REFERENCES

1 G Boothroyd, P Dewhurst, and W Knight, Product Design for Manufacturing and Assembly, Marcel

Dekker, New York, 1994.

2 Boothroyd Dewhurst Inc., Design for Assembly Software, Version 8.0, Wakefield, RI, 1996.

1.4 Inefficiencies with the Traditional Approach 31 1.5 A Summary of the Problems 31

2.2 What the New Technology Is

2.3 Tuszynski’s Relational Algorithm (TRA) 33

1.1 A Historical Perspective on Technological Development

The main thrust of technological development has been to explore the relationships betweencauses and effects We do this for many reasons, two of which are prime The first reason

is so that we can understand the natural processes that surround us and comprise our ronment The second reason is the basic premise that if we understand what the relationshipsare between causes and effects, we will be able to produce the effects we want by activatingthe causes that produce the results we want and minimizing the causes that detract from theresults we want This has been true from the earliest glimmerings of technology developed

envi-by mankind

For example, what engineer has not learned Newton’s law that F⫽ ma? If we know the acceleration (a) that we want and the mass (m) of the item to be accelerated, then we can compute the force (F) required If the mass of the item is fixed, then there will be only

two variables and the acceleration will be proportional to the force applied In this instance,

if the force is doubled the acceleration is doubled If the force is quadrupled, so is theacceleration What we learn from Newton is that acceleration is related to force In thisinstance, the two variables are ‘‘co-related.’’ We also say, with the same meaning, that theyare ‘‘correlated.’’

When did we start thinking this way? Perhaps it was when our early ancestors firstthrew rocks at animals to defend themselves or get food, or perhaps even earlier when earlyhumans tried to figure out what kind of behavior it took to survive or reproduce Thisapproach is ingrained in the human mentality and has been the foundation on which we have

Edited by Myer Kutz Copyright  2006 by John Wiley & Sons, Inc.

1 Introduction 23

NOISE VARIABLES

CONTROL VARIABLES

Figure 1 The traditional manufacturing flow diagram.

built our technology This approach has been necessary, useful, and productive It is hard toimagine our existence without our understanding the linkages between causes and effects.Converting raw material to useful products in a manufacturing process involves con-verting inputs into outputs More generally, the result of any natural process is change Thepurpose of man-made processes is to produce change toward a desired goal or objective

1.2 The Traditional Approach

Manufacturing Process Flow Diagram

The balance of this chapter discusses process technology in the context of design, tooling,manufacturing, and quality engineering.* Figure 1 is a manufacturing process flow diagram

It shows inputs into and output from the manufacturing process Control and noise variablesinfluence the output for any given input

Inputs

Inputs are items input into the process and can be physical or nonphysical Inputs are usuallyphysical items when the process produces physical output, but they can be nonphysical itemswhere the process is an algorithmic or computational process Nonphysical items can be ofmany types In a manufacturing or simulation process, inputs are usually either variable orattribute data Inputs can also be combinations of physical and nonphysical items

The manufacturing process is a single step or a series of sequential steps that modifiesthe inputs to the process Each stage in a manufacturing process will have an input and anoutput.†The input of any one stage will be the output of the preceding stage and the output

of any one stage will be the input to the subsequent stage

Control Variables

Control variables are the process parameters‡controlled by an operator or process engineer

In essence, these are the knobs and dials, whether manually or automatically controlled, on

* Hopefully, those individuals schooled in sciences and technologies other than engineering and facturing will see applications from this chapter to their respective areas of expertise.

manu-† For ease of reading, multiple inputs and / or outputs are referred to here as an input or an output.

‡ Pressures, temperatures, speeds, times, chemical concentrations, orientations, power settings, cies, intensities, agitation levels, etc are typical process control settings.

frequen-INPUTS THE MFG PROCESS OUTPUTS

NOISE VARIABLES

CONTROL VARIABLES

SUPPLY VOLTAGE HUMIDITY

AMBIENT TEMPERATURE

PROCESS PRESSURES PROCESS TEMPERATURES PROCESS TIMES

PROCESS SPEEDS

PART DIMENSION 1 PART DIMENSION 2 PART DIMENSION 3 PART WEIGHT PART PERFORMANCE

}

CAUSES

EFFECTS

Figure 2 Process causes and effects.

the manufacturing equipment that are adjusted to produce conforming parts.* Figure 2 showstypical process control variables such as pressures, temperatures, times, and speeds Theseare the variables that are controlled by the process operator

Noise Variables

Noise variables are those variables that influence the output of the process Figure 2 showsvarious noise variables for a particular process Noise variables are not controlled becauseeither they cannot be controlled or we choose to not control them Noise variables may beleft uncontrolled for numerous reasons, including circumstances where

• They are unknown

• They are too expensive to control

• They are too time-consuming to control

• It is not possible to control them

• Controlling them would not make an appreciable difference in the quality or bility of the manufactured part

produci-Output Variables

The output of the process consists of the item to be manufactured or produced The output

of the process will have different characteristics In the instance where parts are being

man-* Conforming parts meet engineering specification or drawing values Conforming parts are defined as

‘‘good parts’’ and nonconforming parts are defined as ‘‘bad parts.’’

1 Introduction 25

ufactured, the characteristics will be referred as part characteristics Figure 2 shows varioustypical part characteristics such as dimensions, weight, or performance as outputs of themanufacturing process

For purposes of this chapter, part characteristics are divided, arbitrarily and for ience, into four categories:

conven-1 Variable characteristics

2 Attribute characteristics

3 Material characteristics

4 Performance characteristics

Variable characteristics are most typically dimensions Dimensions are usually

subcatego-rized into critical and noncritical dimensions.*

Attribute characteristics are data describing a part characteristic not measurable on a

number scale Typical attribute characteristics are on or off, the presence or absence ofundesirable characteristics, supplier A, B, or C, material type m, n, or o, etc Some attributes,such as color, can be converted to variable data (for example, a combination of red, green,and blue) if it is worth the cost and effort In the context of this chapter, visual attributecharacteristics can be thought of as the presence or absence of some desirable or undesirablepart characteristic as determined through visual inspection

Material characteristics are the physical properties of the manufactured part Typical

material characteristics could be tensile strength, surface hardness, density, or reflectivity.Material characteristics are usually variable data The categorization of part characteristicsinto these first three categories is not crucial, but is a matter of convenience

Performance characteristics refer to those characteristics that are measures of how well

the part performs relative to its functional requirements Performance characteristics are ally variable characteristics but can also be attribute characteristics

usu-Causes and Effects

Figure 2 identifies the control variables as causes and the output characteristics as effects.The foundation of the traditional approach is to relate causes and effects

The Traditional Approach to Cause and Effect

Figure 3 shows how the essence of the traditional approach is to determine the linkagebetween process causes and effects This approach is premised on the logical belief that if

we adequately understand the relationships between causes and the effects, then we should

be able to set the control variables to values that produce the desired part characteristics Inessence, the traditional approach looks for the correlations that model the relationships be-tween causes and effects

The Traditional Approach to Determining Correlations

The traditional approach to determining the correlations between causes and effects is toevaluate the part characteristics when parts are manufactured under different control settings(causal conditions) In some instances, the control settings will be deliberately changed toinduce variation in the manufactured parts In other instances, there may be enough natural

* Critical dimensions are those considered by the design engineer to be critical to form, fit, or function (performance).

INPUTS THE MFG

NOISE VARIABLES

CONTROL

Figure 3 The traditional approach to cause and effect.

variation in the control settings that, over time, enough data with enough variation will begenerated so that the correlations can be determined

Prior to the invention and application of efficient statistical methods, variation in partcharacteristics was usually induced by changing one control variable at a time and thendetermining how each part characteristic changed Sequentially changing each control settingone at a time is very time and cost inefficient and can lead to erroneous conclusions

Design of Experiments

Design of experiments (DOE) is a statistical methodology that has greatly improved theefficiency of determining the relationships between causes and effects DOE is a large stepforward in improving the time and cost efficiencies and in reducing erroneous conclusions.However, as discussed below, DOE does not universally explain all of the relationshipsbetween causes and effects

1.3 Problems with the Traditional Approach

As useful as the traditional approach is, there are many situations in which it is difficult,impossible, or uneconomical to determine the relationships between causes and effects Thesesituations can occur irrespective of whether DOE is used This section discusses severalsituations that make determining the relationships between causes and effects impractical

More Than a Few Control Variables—Many Relationships

Some processes have relatively few control variables, while others can have many As shown

in Fig 4, plastic injection-molding processes, for example, can have over 20 control bles When many control variables are involved, there are many cause-and-effect relation-ships to be evaluated and understood

varia-More Than a Few Control Variables—Many Control Setting Combinations

Further, as the number of control variables increases, the possible number of combinations

of control variables increases geometrically For most process variables, there are an infinite(analog) or large (digital) number of settings for each control variable However, the situationcan get complex even when there are only two or three levels chosen for each controlvariable For example, if there are 20 control variables and each control variable is examined

at only two settings—a high and a low value—then there are over one million possiblecombinations If each of the 20 control variables is examined at three settings—a high, anominal, and a low value—then there are over three billion possible combinations

1 Introduction 27PRESSURES

DIMENSIONS BLISTERS FLASH COLOR WEIGHT STRENGTH

BOWING SPECKS SINK MARKS KNIT LINES DELAMINATIO N BLUSH

INJECTION MOLDING PROCESS

Figure 4 Typical plastic injection molding variables.

42 DIMENSIONS

Figure 5 Typical plastic injection molding bles: a single part with 42 critical dimensions.

varia-More Than a Few Part Characteristics

For products with multiple part characteristics, process engineers and operators have thedifficult task of attempting to adjust process settings to produce parts with all dimensionssimultaneously at target values Some parts have a large number of critical characteristics.For example, the single plastic injection-molded part shown in Fig 5 has 42 critical dimen-sions In this instance, one must determine the relationship between each control variableand each of the 42 critical dimensions

Different Responses to Control Variable Changes

Different part characteristics can have different responses to changes in control settings.Figure 6 shows how the length of a part increases when the process temperature setting isincreased However, the diameter of the part decreases as the process temperature setting isincreased One cannot increase both the length and diameter of the part by changing theprocess control setting

Multiunit Processes

Some processes produce multiple parts for each process cycle Injection-molded parts, forexample, are frequently made with multicavity molds Figure 7 illustrates this point for apart that has 5 critical dimensions and is manufactured in an 8-cavity mold It is not unusualfor molded plastic parts to be made with 8-, 16-, or 32-cavity molds Molded rubber parts

3 5

4

3 5

4

3 5

4 2

1 3 5

4

3 5

4

3 5

4

3 5

4 2

1 3 5

4

3 5

4

3 5

4

3 5

4

2 1

3 5

4

2 1 3 5

4

3 5

4

2 1 3 5

4

3 5

4

2 1 3 5

4

3 5

4

2 1 3 5

4 2

1 3 5

4

3 5

4

3 5

4

3 5

4

2 1

3 5

4

2 1 3 5

4

3 5

4

2 1 3 5

4

3 5

4

2 1 3 5

4

3 5

4

2 1 3 5

4 2 PARALLEL PROCESSES

Figure 7 Multiunit processes greatly increase the number of part characteristics.

can be made with molds having hundreds of cavities Semiconductor wafer fabrication esses can result in thousands of circuits on each wafer

proc-Multiunit processes can get quite complex when each part has many part characteristics.For example, when a part that has 42 critical dimensions is produced in a 32-cavity mold,each machine cycle produces 1344 separate critical dimensions

Simple Interactions

Figure 8 illustrates a simple interaction between two control variables In this example, ifthe pressure control variable is at setting level 1, the length of the part increases as thetemperature control variable increases However, if the pressure control variable is at settinglevel 2, the length of the part decreases as the temperature increases Put more simply, theresponse of the length to changes in temperature depends on the value of the pressure Simpleinteractions are common in many manufacturing processes

Complex Interactions

Figure 9 illustrates a complex interaction between three control variables In this instance,

if the pressure control variable is at setting level 1, the length of the part

• Decreases as the temperature increases when the speed is at level 1

• Remains unchanged as the temperature increases when the speed is at level 2

• Increases as the temperature increases when the speed is at level 3

A different set of three response curves will also exist for the pressure control variablesetting at level 2 Put more simply, the response of the length to changes in temperaturedepends not only on the value of the pressure but also on the value of the speed Although

1 Introduction 29

LENGTH

TEMP ERATURE

PRESSURE 1 PRESSURE 2 SIMPLE INTERACTION

Figure 8 Simple interactions make it difficult to termine responses.

de-TEMP ERATURE

LENGTH

PRESSURE 1, SPEED 3 COMPLEX INTERACTION

PRESSURE 1, SPEED 2 PRESSURE 1, SPEED 1

Figure 9 Complex interactions make it even more difficult to determine responses.

complex interactions are not as common as simple interactions, they do occur occasionally

in manufacturing processes

Nonlinear Responses

Figure 10 shows a nonlinear response between a part characteristic and a process controlvariable The length increases, reaches a maximum, and then decreases as the temperaturecontrol variable increases If one is sampling only two temperature levels to determine theresponse of the length to changes in temperature, one could conclude, depending on the twopoints chosen, that

• Length increases as temperature increases

• Length is insensitive to changes in temperature Or

• Length decreases as temperature increases

Two types of errors can occur The first is the linear approximation of a nonlinear response.The second is that a reversal, as noted immediately above, can occur, which would invalidatethe conclusion on how much and in which direction to change the temperature

DOE May Not Be Helpful

Design of experiments has proven to be a useful tool for circumstances where (1) one part

or performance characteristic needs to be optimized and (2) there are few process ities When the first criterion is not met, DOE can and generally does give conflicting results.When the second criterion is not met, it is difficult to model the process and get usefulresults

complex-Small or Nonexistent Producibility Windows

The producibility window may be nonexistent, i.e., it can be impossible to produce goodparts Operators and process engineers can waste significant time learning this Even if a set