risk management using failure mode and effect analysis (fmea) · risk management using failure mode...
TRANSCRIPT
Risk Management Using Failure
Mode and Effect Analy sis (FMEA)
1123-77375_ch00_5P.indd 1 12/21/18 11:02 AM
Also available from ASQ Quality Press:
Failure Mode and Effect Analy sis: FMEA from Theory to Execution, Second EditionD.H. Stamatis
Advanced Quality Auditing: An Auditor’s Review of Risk Management, Lean Improvement, and Data Analy sisLance B. Coleman Sr.
The Certified Supplier Quality Professional HandbookMark Allen Durivage, editor
The ASQ Auditing Handbook, Fourth EditionJ.P. Russell, editor
Proactive Supplier Management in the Medical Device IndustryJames B. Shore and John A. Freije
Quality Risk Management in the FDA- Regulated Industry, Second EditionJosé Rodríguez- Pérez
The ISO 9001:2015 Implementation Handbook: Using the Pro cess Approach to Build a Quality Management SystemMilton P. Dentch
The Certified Six Sigma Yellow Belt HandbookGovindarajan Ramu
Handbook of Investigation and Effective CAPA Systems, Second EditionJosé Rodríguez- Pérez
Practical Design of Experiments (DOE): A Guide for Optimizing Designs and Pro cessesMark Allen Durivage
The Certified Phar ma ceu ti cal GMP Professional Handbook Second EditionFDC Division and Mark Allen Durivage, editor
The Quality Toolbox, Second EditionNancy R. Tague
To request a complimentary cata log of ASQ Quality Press publications, call 800-248-1946, or visit our Web site at http:// www . asq . org / quality - press.
1123-77375_ch00_5P.indd 2 12/21/18 11:02 AM
Risk Management Using Failure
Mode and Effect Analy sis (FMEA)
D. H. Stamatis
ASQ Quality PressMilwaukee, Wisconsin
1123-77375_ch00_5P.indd 3 12/21/18 11:02 AM
American Society for Quality, Quality Press, Milwaukee 53203© 2019 by ASQ. Printed in 2018.All rights reserved.Printed in the United States of Amer i ca.23 22 21 20 19 18 5 4 3 2 1
Library of Congress Cataloging- in- Publication Data
Names: Stamatis, D. H., 1947– author.Title: Risk management using failure mode and effect analy sis (FMEA) / D.H. Stamatis.Milwaukee, Wisconsin : ASQ Quality Press, 2018. | Includes bibliographical references
and index.Identifiers: LCCN 2018053259 | ISBN 9780873899789 (pbk. : alk. paper)Subjects: LCSH: Failure analy sis (Engineering) | Reliability (Engineering) |
Quality control. | Risk management.Classification: LCC TS176 .S75174 2018 | DDC 620/.00452— dc23LC rec ord available at https:// lccn . loc . gov / 2018053259
No part of this book may be reproduced in any form or by any means, electronic, mechanical, photocopying, recording, or other wise, without the prior written permission of the publisher. Portions of information contained in this publication/book are printed with permission of Minitab Inc. All such material remains the exclusive property and copyright of Minitab Inc. All rights reserved.
Director, Quality Press and Programs: Ray ZielkeManaging Editor: Paul Daniel O’MaraSr. Creative Ser vices Specialist: Randy L. Benson
ASQ Mission: The American Society for Quality advances individual, orga nizational, and community excellence worldwide through learning, quality improvement, and knowledge exchange.
Attention Bookstores, Wholesalers, Schools, and Corporations: ASQ Quality Press books, video, audio, and software are available at quantity discounts with bulk purchases for business, educational, or instructional use. For information, please contact ASQ Quality Press at 800-248-1946, or write to ASQ Quality Press, P.O. Box 3005, Milwaukee, WI 53201–3005.
To place orders or to request a free copy of the ASQ Quality Press Publications Cata log, visit our website at http:// www . asq . org / quality - press.
Printed on acid- free paper.
1123-77375_ch00_5P.indd 4 12/21/18 11:02 AM
v
List of Figures and Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ixPreface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiIntroduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii
Chapter 1 Risk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1Overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1In ISO 9001:2015 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3In IATF 16949 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7VDA (New FMEA Proposed Standard of 2018) . . . . . . . . . . . . 7Method or Pro cess of Conducting Any Risk Assessment . . . . . 10General Comments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
Chapter 2 Reliability and FMEA . . . . . . . . . . . . . . . . . . . . . . . . 19Overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19Need to Understand the Concept of Failure . . . . . . . . . . . . . . . . 20Design for Reliability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
Chapter 3 Prerequisites of FMEA . . . . . . . . . . . . . . . . . . . . . . . 23Creating an Effective Team . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23The Structure of the FMEA Team . . . . . . . . . . . . . . . . . . . . . . . 23Ingredients of a Motivated FMEA Team . . . . . . . . . . . . . . . . . . 25Potential FMEA Team Members . . . . . . . . . . . . . . . . . . . . . . . . 26Mind-set of Minimizing Failures . . . . . . . . . . . . . . . . . . . . . . . . 27
Chapter 4 What Is an FMEA? . . . . . . . . . . . . . . . . . . . . . . . . . . . 29Definition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29Is FMEA Needed? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30Benefits of FMEA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31The Pro cess of Conducting an FMEA . . . . . . . . . . . . . . . . . . . . 32Understanding Your Customers and Their Needs . . . . . . . . . . . 36
Table of Contents
1123-77375_ch00_5P.indd 5 12/21/18 11:02 AM
vi Table of Contents
What Happens after Completion of the FMEA? . . . . . . . . . . . . 37Vocabulary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
Chapter 5 Robustness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45Boundary Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45Interface Matrix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46P- Diagram. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
Chapter 6 The FMEA Form and Rankings . . . . . . . . . . . . . . . . 49Severity Rating . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50Occurrence Rating . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51Detection Rating. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51Classification and Characteristics . . . . . . . . . . . . . . . . . . . . . . . 52Understanding and Calculating Risk . . . . . . . . . . . . . . . . . . . . . 53Driving the Action Plan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
Chapter 7 Types of FMEA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57FMEA Challenges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
Chapter 8 The Common Types of FMEAs . . . . . . . . . . . . . . . . . 59Concept . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61Pro cess . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66Equipment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
Chapter 9 Health FMEA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85Comparison of Root- Cause Analy sis and HFMEA . . . . . . . . . . 86The Pro cess of the HFMEA . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87Forms and Rankings. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
Chapter 10 Failure Mode Effects and Criticality Analy sis (FMECA) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
Definition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97Pos si ble Sources for Identifying Functions . . . . . . . . . . . . . . . . 98The Pro cess of Conducting an FMECA. . . . . . . . . . . . . . . . . . . 99Qualitative Analy sis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101Quantitative Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101Form . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
Chapter 11 Control Plans . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113Purpose . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113Benefits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114Content of a Control Plan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114FMEA/Control Plan Linkage . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
1123-77375_ch00_5P.indd 6 12/21/18 11:02 AM
Table of Contents vii
Chapter 12 Linkages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119DFMEA to PFMEA to Pro cess Flow Diagram
to Control Plan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
Chapter 13 Tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123Overview of Typical Tools Used in FMEA . . . . . . . . . . . . . . . . 123
Chapter 14 Troubleshooting an FMEA . . . . . . . . . . . . . . . . . . . 141 After FMEA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
Chapter 15 Typical Concerns When Conducting an FMEA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
Common Team Prob lems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147Common Procedural Prob lems. . . . . . . . . . . . . . . . . . . . . . . . . . 147Institutionalizing FMEA in Your Com pany. . . . . . . . . . . . . . . . 149
Chapter 16 FMEAs Used in Selected Specific Industries . . . . . 151Automotive . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151Aerospace . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152Software . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152Chemical/Pharmaceutical . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152
Chapter 17 Warranty, Six Sigma, Lean, and FMEA . . . . . . . . 155Automotive Perception of Warranty . . . . . . . . . . . . . . . . . . . . . . 155Six Sigma . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158Lean . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159 After Improvements Are Made . . . . . . . . . . . . . . . . . . . . . . . . . 159
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161Selected Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165
1123-77375_ch00_5P.indd 7 12/21/18 11:02 AM
1123-77375_ch00_5P.indd 8 12/21/18 11:02 AM
ix
Figure 1.1 VDA- proposed FMEA form . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Table 1.1 A typical risk register for a proj ect that includes four steps: identify, analyze, plan response, monitor and control . . . . . . . 16
Figure 4.1 Overview of a DFMEA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
Figure 4.2 Overview of a PFMEA. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
Table 5.1 Robustness focus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
Figure 5.1 A typical boundary diagram . . . . . . . . . . . . . . . . . . . . . . . . . . 46
Table 5.2 Interface matrix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
Figure 5.2 A typical P- diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
Figure 6.1 A typical FMEA form . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
Figure 6.2 Area chart showing priority levels . . . . . . . . . . . . . . . . . . . . . . 53
Table 7.1 Types of FMEAs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
Table 8.1 DFMEA— severity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
Table 8.2 DFMEA— occurrence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
Table 8.3 DFMEA— detection. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
Figure 8.1 A typical PFMEA. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
Table 8.4 PFMEA— severity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
Table 8.5 PFMEA— occurrence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
Table 8.6 PFMEA— detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
Table 8.7 A typical control matrix for a manufacturing pro cess. . . . . . . 72
Figure 8.2 Explanation of the EFMEA form. . . . . . . . . . . . . . . . . . . . . . . 79
List of Figures and Tables
1123-77375_ch00_5P.indd 9 12/21/18 11:02 AM
x List of Figures and Tables
Table 9.1 Similarities and differences of RCA and HFMEA . . . . . . . . . 87
Table 9.2 8 wastes and 6S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
Table 9.3 A typical comparison of pro cess design and orga nizational change. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
Figure 9.1 A typical HFMEA worksheet . . . . . . . . . . . . . . . . . . . . . . . . . 92
Table 9.4 A typical severity ranking . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
Table 9.5 A typical matrix showing severity and probability . . . . . . . . . 94
Figure 10.1 A typical qualitative FMECA . . . . . . . . . . . . . . . . . . . . . . . . . 102
Figure 10.2 A typical quantitative FMECA . . . . . . . . . . . . . . . . . . . . . . . . 109
Figure 11.1 Linkage from DFMEA to CP. . . . . . . . . . . . . . . . . . . . . . . . . . 116
1123-77375_ch00_5P.indd 10 12/21/18 11:02 AM
xi
Change rarely comes in the form of a whirlwind, despite the currently popu lar notion to the contrary. Change is not “creative destruction” like we’ve been told. Change that expects us to throw out every thing we were and start over isn’t change at all, but a convulsion. A hiccup. The internet did not change every thing. Broadband did not change every thing. September 11 did not change every thing. Nor did Enron, WorldCom, or any other com pany. Nor will tomorrow’s horror, tomorrow’s amazing breakthrough, or tomorrow’s scandal.
If you follow the cataclysmic theory of change, you will reap a whirlwind indeed. There is a dif fer ent theory of change that no one talks about but that is much more significant for the wise professional. In the coastlines of any coun-try, state, or territory one can see it every day. The waves may crash against the rocks, but they are a distraction. The real action is the tide . When the tide changes, huge forces are put in motion that cannot be halted. (If you doubt the power of the tide, look at the suburbs of any fair- size town anywhere. A piece of farmland on the edge of most towns is worth its weight in gold. Why? Because it’s where the affluent middle class wants to bunk down every night.)
Risk is everywhere. It does not matter where we are or what we do. It affects us on a personal level, but it also affects us in our world of com-merce and our business. No matter if we are focused on identifying and/or analyzing risks, these risks are always balanced with an associated benefit. In the final analy sis, all types of risks exist, and they are generated by a variety of factors, such as customer requests, continual improvement phi-losophy, and competition.
Why do we do a risk analy sis? Primarily to answer the following two questions:
1. What can go wrong?
2. If something does go wrong, what is the probability of it happen-ing and what is (are) the consequence(s)?
Preface
1123-77375_ch00_5P.indd 11 12/21/18 11:02 AM
xii Preface
In the past, these questions were focused on “prob lem fixing.” The primary analy sis was to focus on “who” did it. Of course, the focus on prob-lems assumed that somebody was to blame, and action was taken. In other words, we operated on the princi ple that “If it’s not broken, don’t touch it.” Today, that paradigm has changed. The focus is on prevention. In other words: “If it is not broken improve it.” The focus is on “how it happened” and “why it happened.”
In this book we explore the evaluation pro cess of risk by utilizing one of the core methodologies available: failure mode and effect analy sis (FMEA). Our intent in this book is to make the concepts easy to understand and explain why FMEA is used in many industries with positive results to either eliminate or mitigate risk.
It is not a complete reference on FMEA; rather, it is a summary guide for every one who wants some fast information regarding failures and how to deal with them. Specifically, we cover the following topics:
• Risk as defined in International Organization for Standardization (ISO), International Automotive Task Force (IATF), and Verband der Automobilindustrie (VDA) standards and/or guidelines
• Reliability and FMEA
• Prerequisites of FMEA
• What an FMEA is
• Robustness
• The FMEA form and rankings
• Types of FMEA
• The common types of FMEAs
• Failure mode effects and criticality analy sis (FMECA)
• Health FMEA
• Control plans
• Linkages
• Tools
• Troubleshooting an FMEA
• Getting the most from FMEA
• FMEAs used in selected and specific industries
• Warranty, Six Sigma, and lean
1123-77375_ch00_5P.indd 12 12/21/18 11:02 AM
xiii
In the past 100 years or so, the United States has been the envy of the world. This country has been the leader in almost every major innovation people have made. The historical trend has been positive indeed. However, what about the future? Can the status quo be retained? Is there anything to worry about? Can the leadership for tomorrow be guaranteed by following past successes?
The United States wants to be among the leaders; it wants to be better; its citizens want to work smart and be efficient. But with leadership and gen-eral betterment comes change— change in be hav ior and technology. The old ways served workers well but not anymore. The following saying describes the situation best.
If you always do what you always did, you will always get what you always got .
What the United States has achieved in the past is not good enough anymore as world competition increases; it must improve or it will be left behind as others pursue technological and quality improvements for their products and/or ser vices. In simple terms, this means that the attitude and be hav ior toward quality must change.
A good starting point is for organ izations to use the 6S pro cess (sort- store- shine- standardize- sustain- safety), emphasizing sustain and safety. Both of these areas focus on prevention and will lead to good designs as well as excellent pro cesses.
As with any transformation, this change brings uncertainty and risk. However, this transformation may be successful if the organ ization has (1) vision, (2) mission, (3) strategy, (4) action plan, and (5) implementation strategy.
The recognition that all well- managed companies are interested in pre-venting or at least minimizing risk in their operations is the concept of
Introduction
1123-77375_ch00_5P.indd 13 12/21/18 11:02 AM
xiv Introduction
risk management analy sis. The requirements for performing such analy sis may be extensive and demanding. The elimination, control, or reduction of risk is a total commitment by the entire organ ization, and it is more often than not the responsibility of the engineering department. In this book we focus only on a small portion of this engineering responsibility— specifically, the FMEA methodology. Here we must emphasize that FMEA is only one methodology of many that can help in the strategy, action plan, and imple-mentation strategy for improvement.
1123-77375_ch00_5P.indd 14 12/21/18 11:02 AM
1
1Risk
OVERVIEW
Over the last five or so years, much discussion has been devoted to risk, although the fact is that the “risk concept” is immature in the International Organ ization for Standardization (ISO) documents. Why? Because of the 157 items in the ISO standards the word “risk” is used in 45 unique definitions, of which 21 specifically address hazards. On the other hand, in the International Automotive Task Force (IATF) 16949 standard, the word “risk” is sprinkled throughout the standard and it is especially emphasized in reference to pre-vention actions and failure mode and effect analysis (FMEAs).
In both ISO and IATF documents, the definitions are abundant as they consider events with negative outcomes using very specific language. For example, risk is “a function of the probability of occurrence of a given threat and the potential adverse consequences of that threat’s occurrence.”
To be sure, that is a very unique definition. However, it is not all inclu-sive. For example, the Project Management Body of Knowledge (PMBOK) (2000, ch. 11; 2017, ch. 11) goes further by including concerns that deal with the effect of uncertainty, the combination of the consequences of an event and the associated likelihood of its occurrence, and uncertain events or conditions that, if they occur, have a positive or negative effect on a proj ect’s objectives.
The term risk, used in the ISO standards and the new IATF standard, “pertains to safety and/or per for mance requirements in the context of meet-ing applicable regulatory requirements at minimum.” An FMEA is a standard technique used to assess and evaluate potential risks in the design devel-opment phase, which continues during production pro cess controls. Other related techniques are fault tree analy sis, warranty analy sis, SWOT (strengths- weaknesses- opportunities- threats) analy sis, event tree analy sis, business con-tinuity planning, BPEST (business, po liti cal, economic, social, technological) analy sis, real option modeling, decision making under conditions of risk and
1123-77375_ch01_5P.indd 1 12/21/18 11:02 AM
2 Chapter One
uncertainty, statistical inference, mea sures of central tendency and disper-sion, and PESTLE (po liti cal, economic, social, technical, legal, environ-mental) analy sis. Any of these vari ous risk assessment techniques can also be used to incorporate other aspects of the quality management system (Stamatis 2014). The specific definition, per the ISO standards, includes three related definitions:
1. Risk definition. Risk is the potential of gaining or losing something of value. Values (such as physical health, social status, emotional well- being, or financial wealth) can be gained or lost when taking risk resulting from a given action or inaction, foreseen or unforeseen. Risk can also be defined as the intentional interaction with uncertainty (EN ISO 14971:2012, 2.16).
2. Risk management. Risk management is the identification, assess-ment, and prioritization of risks (defined in ISO 31000:2018 as the effect of uncertainty on objectives), followed by coordinated and eco nom ical appli-cation of resources to minimize, monitor, and control the probability and/or impact of unfortunate events or to maximize the realization of opportuni-ties (EN ISO 14971:2012, 2.22).
3. Risk assessment. Risk assessment is the determination of a quantita-tive or qualitative estimate of risk related to a well- defined situation and a recognized threat (also called hazard). Quantitative risk assessment requires calculations of two components of risk (R): the magnitude of the potential loss (L) and the probability (p) that the loss will occur (Stamatis 2014).
The need for risk analy sis is defined in several clauses of the stan-dard. For example, in ISO 13485:2016, we find that risk is mentioned either directly or indirectly throughout the standard but specifically in sections (clauses) 4.1; 7.4.1; 7.4.2; 8.2; 8.2.1; 8.3; and 8.3.4.
Therefore, the issue of risk has become a very impor tant quality requirement and is viewed as a preventive tool in any quality management system (QMS), Advanced Product Quality Planning (APQP) pro cess, and FMEA of any organ ization. In fact, for most organ izations, there is no need to have a separate section to identify “preventive action” because the con-cept of preventive action is expressed through a “risk- based” approach in which the QMS requirements are identified and implemented. For example, ISO 9001:2015 makes risk- based thinking a requirement by specifically calling for risks and opportunities to be determined and addressed. This is a very strong statement to demonstrate that risks are to be identified, correc-tive actions must be planned, and implementation has to be monitored for the specific risks under consideration.
Reading the standards, one concludes without hesitation that their intent as far as risk is concerned is (a) to provide confidence in the organ-
1123-77375_ch01_5P.indd 2 12/21/18 11:02 AM
Risk 3
ization’s ability to consistently provide customers with conforming goods and ser vices and (b) to enhance customer satisfaction. The concept of “risk” in the context of the international standards relates to the uncertainty in achieving these objectives.
Since that is the primary concern, the standards articulate the notion of risk- based thinking, which is defined as something we all do automati-cally and often subconsciously. Although the concept of risk has always been implicit in ISO 9001, the 2015 revision makes it more explicit and builds it into the whole management system. So, risk- based thinking now has become part of the pro cess approach. The intent here is to make sure that risk- based thinking makes preventive action part of the routine pro cess. This is also very true in the FMEA pro cess.
To be sure, risk is often thought of only in the negative sense. However, risk- based thinking can also help to identify opportunities. This can be con-sidered the positive side of risk, which is a relatively recent concept. The result of this thinking is:
• To improve customer confidence and satisfaction
• To ensure consistency of quality of goods and ser vices
• To establish a proactive culture of prevention and improvement
• To have successful companies intuitively take a risk- based approach
IN ISO 9001:2015
Risk in ISO 9001:2015 and ISO 14001:2015 is implied throughout the stan-dards as it is referenced wherever the words “planning” or in clause 6.0 “prevention” and “effectiveness” are mentioned. The reason for this general inclusion is because the ISO standards consider risk to be an inherent char-acteristic in all aspects of a quality management system.
Therefore, in a given organ ization risk- based thinking must be preva-lent in any management review because it is a vital ele ment in the continual improvement pro cess that is focused on prevention. This risk- based think-ing must be demonstrated with risk register documentation and must be available for review during audits. A risk register is documented informa-tion that validates that an organ ization has done risk- based thinking.
It must be emphasized here that risk- based thinking is indeed preven-tive action and unequivocally is every body’s business! As a consequence, it must become an integral part of the orga nizational culture.
It is significant to note that ISO 9001:2015 does not automatically require anyone to carry out a full, formal risk assessment or to maintain
1123-77375_ch01_5P.indd 3 12/21/18 11:02 AM
4 Chapter One
a risk register per ISO 31000 (Risk management— Princi ples and guide-lines on implementation); although it will be a useful reference, it is not mandated. On the other hand, the standard requires the proper identifica-tion of what the risks and opportunities are in your organ ization, which depends on context. That is why the active participation of management is imperative.
The premise of the risk evaluation as a quality management system is founded in clause 4.4— Quality management system and its pro cesses. Here we find that the organ ization must establish, implement, maintain, and continually improve a quality management system, including the pro cesses needed and their interactions, in accordance with the requirements of this ISO standard. The organ ization must determine the pro cesses needed for the quality management system and their application throughout the organ-ization (in subsection (f) we read the risks and opportunities in accordance with the requirements of 6.1) and then plan and implement the appropriate actions to address them.
Furthermore, we need to read clause 4.6 (Planning for the quality man-agement system) and clause 6.1 (Actions to address risks and opportuni-ties, specifically, 4.1 and 6.1.1). When planning for the quality management system, the organ ization shall consider the issues referred to in 4.1 and the requirements referred to in 4.2 and determine the risks and opportunities that need to be addressed to (a) give assurance that the quality management system can achieve its intended result(s); (b) prevent, or reduce, undesired effects; and (c) achieve continual improvement.
In section 5 of the standard, Management responsibility, and clause 6.0— Planning—we see specific actions. For example, in 6.1 we find actions to address risks and opportunities. In 6.1.1 we find that when planning for the quality management system, the organ ization shall consider the issues referred to in 4.1 and the requirements referred to in 4.2 and determine the risks and opportunities that need to be addressed to (a) give assurance that the quality management system can achieve its intended result(s); (b) prevent, or reduce, undesired effects; and (c) achieve continual improvement. Fi nally, clause 6.1.2 tells us that the organ ization shall plan (a) actions to address these risks and opportunities, and (b) how to, one, integrate and implement the actions into its quality management system pro cesses (see 4.4) and, two, evaluate the effectiveness of these actions. Actions taken to address risks and opportunities shall be proportionate to the potential impact on the conformity of products and ser vices.
Of course, to appreciate what the ISO standard is requiring is to see the individual clauses in context. The context should be addressed from at least the following actions and questions:
1123-77375_ch01_5P.indd 4 12/21/18 11:02 AM
Risk 5
• Analyze and prioritize the risks and opportunities in your organ ization— what is acceptable? What is unacceptable?
• Plan actions to address the risks. How can I avoid or eliminate the risk? How can I mitigate the risk?
• Implement the plan— take action.
• Check the effectiveness of the action. Does it work? Learn from experience (continual improvement).
Overall, the ISO standard pres ents an outline of requirements for risk- based thinking as specific requirements. For example:
• Risks and opportunities are determined and addressed.
• Actions are implemented to address the specific risk(s) identified.
• There is no requirement for formal methods for risk management or a documented risk management pro cess.
• QMS is considered a preventive tool (as such, there is no need to have a separate clause or sub- clause titled “preventive action”; the concept of preventive action is expressed through a risk- based approach to formulating QMS requirements).
• Risk is defined in ISO 9000 as the effect of uncertainty. (Uncertainty is the state— even partial— of deficiency of information related to the understanding or knowledge of an event, its consequence, or its likelihood. Remember that not all pro cesses have the same level of risk in terms of the organ ization’s ability to meet its objectives.) Therefore, the consequences of pro cess, product, ser vice, or system nonconformities will not be the same for all organ izations.
• Opportunities can arise as a result of a situation favorable to achiev-ing an intended result, which means any opportunity is not the posi-tive side of risk. Rather, an opportunity is a set of circumstances that makes it pos si ble to do something— either from a positive or negative situation. This means that an opportunity pres ents dif fer-ent levels of risk and appropriate planning, definition, and therefore evaluation must take place, and the earlier the better.
Goal
So, generally speaking, ISO’s goal is to make sure that the customer is satis-fied with the quality system of the supplier in such a way that it is sufficient
1123-77375_ch01_5P.indd 5 12/21/18 11:02 AM
6 Chapter One
to deliver an acceptable product. In other words, it is a system- oriented standard.
On the other hand, the goal of the automotive QMS standard is an upgrade of ISO to the new IATF 16949 standard as defined on page 7 of it. The new upgrade provides for continual improvement, emphasizing defect prevention and reduction of variation and waste (in the entire supply chain) in much more detail. In essence, the idea is to harmonize the automotive standards globally.
Proj ect
Once the goal has been defined, the next concern is to make sure manage-ment understands the perimeters of the proj ect from at least four perspec-tives (Stamatis 2016, 1998). They are:
1. Resources: Of great importance here is to recognize the workload in resource planning.
2. Timing: Proj ect plan must be in place with internal and external milestones and critical paths that are practical and realistic to fulfill all the planned resources requirements.
3. Ability to address customer requirements: Any success depends on whether the customer requirements are met as defined and expected.
4. Change: Change is inevitable. Therefore, the first and perhaps the most impor tant ele ment of any change implementation is the recognition of an effective implementation.
The key new actions in ISO 9001 that address risks and opportunities (in addition to the ones that have already been mentioned) are:
• 6.1.1 Actions to Address Risks and Opportunities: Planning needs to consider the context (issues) of the organ ization (4.1) and the interested party expectations (4.2) and address the risks and oppor-tunities to help the QMS achieve “its intended results.”
• 6.1.2 Actions to Address Risks and Opportunities: Actions need to be developed, integrated, and implemented back into the QMS pro cesses (4.1).
• 9.1.3 Analy sis and Evaluation: Effectiveness of actions on risks and opportunities.
• 9.3 Management Review: (b) Changes in external and internal issues and (e) effectiveness of actions on risks and opportunities.
1123-77375_ch01_5P.indd 6 12/21/18 11:02 AM
Risk 7
• 10.1 (Improvement) General: “Preventing or reducing undesired effects.”
• 10.2 Nonconformity and Corrective Action: “Update risks and oppor-tunities” based on corrective actions.
IN IATF 16949
As of September 2018, IATF 16949 is the official automotive requirement replacing the ISO 9001/16949:2015. It has become an official standard and not an engineering specification. In the standard, we find that risk is evident because it is either directly or indirectly mentioned throughout the standard but specifically in sections (clauses) 4.4.1.2; 6.1.2.1; 6.1.2.2; 6.1.2.3; 7.2.3; 8.1.2; 8.3.2.1; 8.7.1.1; 8.7.1.2; and 8.7.1.6.
IATF 16949 supports all the core tools (statistical process control [SPC], measurement system analysis [MSA], FMEA, APQP, production part approval process [PPAP] of the Automotive Industry Action Group [AIAG]) and adds a number of specific risk- related requirements to minimize the chance of failure during new program development and to maximize (not optimize) the chance for realization of planned activities. Key clauses that deal directly with the managing and mitigating risk are 8.1.1; 8.2.2.1; 8.2.3.1.1; 8.2.3.1.3; and 8.6.6.
The comprehensiveness of the new standard is coming to light with further requirements that deal with additional controls for the management of development proj ects through the cycle, eventually concluding with a product approval pro cess. Key clauses that deal with this area are 8.3.3.1; 8.3.4.2; 8.3.4.3; 8.3.5.1; and 8.3.4.4.
VDA (NEW FMEA PROPOSED STANDARD OF 2018)
There is no ambiguity about the relationship of FMEA and risk. The two are very closely related. However, for all industries— except automotive— the requirements of all FMEAs are the same. That is, they follow the specifications of the Society of Automotive Engineers (SAE) J1739, ISO requirements, AIAG, FMEA, and their specific requirements, as appropriate, for their customers.
Princi ples
Generally, there are 12 princi ples that guide the way that risk management is integrated and deployed in ISO 9001/16949:2015; in IATF:16949:2016; and in Verband der Automobilindustrie (VDA) and ISO 31000:2018:
1123-77375_ch01_5P.indd 7 12/21/18 11:02 AM
8 Chapter One
1. Risk management creates and protects value. Resources expended to mitigate risk should be less than the consequence of inaction.
2. Risk management is an integral part of all orga nizational pro cesses.
3. Risk management is part of decision making.
4. Risk management explic itly addresses uncertainty and assumptions.
5. Risk management is systematic, structured, and timely.
6. Risk management is based on the best available information.
7. Risk management is tailored to specific tasks.
8. Risk management takes human and cultural factors into account.
9. Risk management is transparent and inclusive.
10. Risk management is dynamic, iterative, and responsive to change.
11. Risk management facilitates continual improvement and enhancement.
12. Risk management demands that organ izations continually or peri-odically reassess their systems.
In the automotive industry, under the leadership of a German consor-tium, a new standard is pending called the VDA. It is supposed to harmo-nize the automotive standards worldwide. In the United States the standard has been met with mixed feelings, although the AIAG has given the stan-dard the green light for further clarification and discussion between its members before full implementation. It is expected to be finalized either very late in 2018 or early 2019. However, it is not a mandatory requirement for Ford, General Motors (GM), and Fiat Chrysler (FC). So, it is up to these automakers if they want to use it or continue with the traditional FMEA approach.
The expected benefits are primarily in five areas. They are:
• Supplying German automakers with a good quality product
• Maintaining higher- quality standards than even IATF 16949
• Increasing overall quality of products
• Reducing warranty costs
• Minimizing risk issues/concerns and prob lems
1123-77375_ch01_5P.indd 8 12/21/18 11:02 AM
Risk 9
Model of the Expected VDA Standard: A Six- Step Approach
Step 1: Scope (Prob lem) Definition
Step 2: Structure Analy sis
Step 3: Function Analy sis
Step 4: Failure Analy sis
Step 5: Risk Analy sis
Step 6: Optimization
Specifically, these steps include:
2.1.1 Purpose: The purpose of the design FMEA (DFMEA) Scope Definition is to define what is included and excluded in the FMEA based on the type of analy sis being developed (i.e., system, subsystem, or component).
2.2.1 Purpose: The purpose of Design Structure Analy sis is to identify and break down the design into system, subsystem, and component parts for technical risk analy sis.
2.3.1 Purpose: The purpose of the Design Function Analy sis is to ensure that the functions specified by requirements/specifications are appropriately allocated to the system ele ments.
2.4.1 Purpose: The purpose of the Design Failure Analy sis is to iden-tify failure causes, modes, and effects and show their relationships in order to undertake risk assessment.
2.5.1 Purpose: The purpose of Design Risk Analy sis is to estimate risk by evaluating severity, occurrence, and detection and prioritize the need for actions.
2.6.1 Purpose: The purpose of the Design Optimization is to determine actions to mitigate risk and assess the effectiveness of those actions. The primary objective of optimization is to develop actions that reduce risk and increase customer satisfaction by improving the design.
Flow
• DFMEA contains information that is useful for pro cess FMEA (PFMEA):
– Failure causes related to piece- to- piece
1123-77375_ch01_5P.indd 9 12/21/18 11:02 AM
10 Chapter One
– End- user failure effects and severity for the failure causes related to product characteristics
• PFMEA contains information that needs alignment with the DFMEA:
– Failure effects and severity for failure modes that are also shown in the DFMEA
• Not all failure causes in a DFMEA are failure modes in a PFMEA.
Action Priority
High: The team must either identify an appropriate action to improve prevention and/or detection controls or justify and document why current controls are adequate.
Medium: The team should identify appropriate actions to improve prevention and/or detection controls or, at the discretion of the com pany, justify and document why controls are adequate.
Low: The team could identify actions to improve prevention or detec-tion controls.
METHOD OR PROCESS OF CONDUCTING ANY RISK ASSESSMENT
The generic method for a risk analy sis of any kind consists of the following ele ments, performed, more or less, in the following order:
1. Identify, characterize threats.
2. Assess the vulnerability of critical assets to specific threats.
3. Determine the risk (i.e., the expected likelihood and consequences of specific types of attacks on specific assets).
4. Identify ways to reduce those risks.
5. Prioritize risk reduction mea sures.
1123-77375_ch01_5P.indd 10 12/21/18 11:02 AM
Risk 11
General Evaluation Criteria for DFMEA
Severity: Whereas the evaluation criteria for a DFMEA for indus-tries other than automotive still follow the AIAG guidelines (or some modification of them), the automotive industry, by adopting the VDA standard, has quite dif fer ent proposed cri-teria. However, since they are not officially published, we can say that the severity is evaluated based on what the end-user experiences.
Occurrence: Whereas the evaluation criteria for a DFMEA for industries other than automotive still follow the AIAG guidelines (or some modification of them), the automotive industry by adopt-ing the VDA standard has quite dif fer ent proposed criteria. Again, though, since they are not officially published we can say that occurrence will be evaluated based on (a) estimated frequency, (b) product experience, and (c) prevention control.
Detection: Whereas the evaluation criteria for a DFMEA for indus-tries other than automotive still follow the AIAG guidelines (or some modification of them), the automotive industry by adopting the VDA standard has quite dif fer ent potential criteria. Once more, since they are not officially published we can say that the detec-tion will be based on the activity performed prior to delivery of the design for production.
General Evaluation Criteria for PFMEA
Severity: Whereas the evaluation criteria for a PFMEA for indus-tries other than automotive still follow the AIAG guidelines (or some modification of them), the automotive industry by adopting the VDA standard has quite dif fer ent proposed criteria. However, since they are not officially published, we can say that the severity is evaluated based on the failure effects rated for manufacturing, assembly, and end user.
Occurrence: Whereas the evaluation criteria for a PFMEA for indus-tries other than automotive still follow the AIAG guidelines (or some modification of them), the automotive industry by adopting the VDA standard has quite dif fer ent proposed criteria. Again, though, since they are not officially published we can say that the occurrence is evaluated based on (a) estimated frequency, (b) pro-cess experience, and (c) prevention control.
1123-77375_ch01_5P.indd 11 12/21/18 11:02 AM
12 Chapter One
Detection: Whereas the evaluation criteria for a PFMEA for indus-tries other than automotive still follow the AIAG guidelines (or some modification of them), the automotive industry by adopt-ing the VDA standard has quite dif fer ent proposed criteria. Once more, since they are not officially published we can say that the detection will be based on activities prior to shipment of the prod-uct so that nonconformances may be caught.
Monitoring and System Response (MSR)
The scope of a supplemental FMEA for monitoring and system response may be established in consultation between customer and supplier. Typical steps are:
1. Scope definition
2. Structure analy sis
3. Function analy sis
4. Failure analy sis
5. Risk analy sis
6. Optimization
FMEA Form
Whereas the form used for the FMEA for industries other than the automo-tive industry still follow the classic AIAG format, the automotive industry, by adopting the VDA standard, has modified the form quite extensively (see Figure 1.1).
Identification
After establishing the context, the next step in the pro cess of managing risk is to identify potential risks. Risks are about events that, when triggered, cause prob lems or benefits. Hence, risk identification can start with the source of our prob lems and those of our competitors (benefit) or with the prob-lem itself. It is impor tant to note here that the prob lem is recognized at the actionable root- cause level. It is also impor tant to identify the prob lem at the escape point (EP). The EP is the place of the pro cess that the root cause could have been caught but it was not. Unless we find the EP, we will have a perpetual prob lem. We may find out how to fix it but the fixing is going to be continual since the origination of the root cause will not be known.
1123-77375_ch01_5P.indd 12 12/21/18 11:02 AM
PreventionAction
DetectionAction
ResponsiblePerson
TargetCompletion
Date
OPTIMIZATIONStatus
(Untouched UnderConsideraton in
ProgressCompletedDecided)
Action Takenwith Pointerto Evidence
CompletionDate
Occ
urre
nce
(O)
AP
Det
ectio
n (D
)
Seve
rity
(S)
DESIGN FAILURE AND EFFECTS ANALYSIS (DFMEA)
1. System(Item)
STRUCTURE ANALYSIS FUNCTION ANALYSIS FAILURE ANALYSIS RISK ANALYSIS
2. SystemElement/Interface
3. Component Element(Item /
Interface)
2. Function ofSystem Element
and intendedPerformance
Output
3. Function ofComponent
Element and Requirement orIntended Outputor Characteristic
2. FailureMode (FM)
3. FailureCause (FC)
CurrentDetection
Control (DC)of FC or FM
Det
ectio
n (D
) of F
OFM
AP
Filte
r Cod
e (O
ptio
nal)
Current Prevention
Control (PC)of FC
Seve
rity
(S) o
f FE
Subject:DFMEA Start Date:
DFMEA Revison Date:FMEA Due Date:
Company Name:Engineering Location:
Customer Name:Model / year / Platform:
FMEA Team:
1. Function ofSystem and
Requirement or intended
Output
1. FailureEffects
(FE)
Occ
urre
nce
(O) o
f FC
DFMEA ID Number:Design Responsibility:
Security Classification:FMEA Intast: FMEA Test:
Figure 1.1 VDA- proposed FMEA form
1123-77375_ch01_5P.indd 13
12/21/18 11:02 AM
14 Chapter One
So, how do we proceed? Depending on the situation there are several options. However, most commonly they fall into two categories:
1. Source analy sis. Risk sources may be internal or external to the system that is the target of risk management (use mitigation instead of management since, by definition, risk deals with factors of decision making that cannot be managed). Examples of risk sources are owners of a proj ect, employees of a com pany, customer requirements, legal requirements, or events of nature.
2. Prob lem analy sis. Risks are related to identified threats (e.g., the threat of losing money, the threat of abuse of confidential information, or the threat of human errors, accidents, and casualties). The threats may exist with vari-ous entities, most impor tant with pro cess owners, customers, and legislative bodies such as the government. When either source or prob lem is known, the events that a source may trigger or the events that can lead to a prob lem can be investigated. For example, stakeholders withdrawing during a proj ect may endanger funding of the proj ect; confidential information may be stolen by employees even within a closed network; natu ral events may result in many casualties.
The chosen method of identifying risks may depend on culture, indus-try practice, applicable and thorough FMEA, and compliance with appro-priate standards and customer specifications.
GENERAL COMMENTS
Every thing we do in life has a risk. Generally, that risk is evaluated (assessed) based on two considerations:
1. The benefit one gets from performing a task
2. The negative consequences from either performing or not doing the task
In any environment there is always more than one risk involved. So, some options in identifying the available risk options are:
1. Design a new business pro cess with adequate built-in risk control and containment mea sures from the start.
2. Periodically reassess risks that are accepted in ongoing pro cesses as a normal feature of business operations and modify mitigation mea sures.
3. Transfer risks to an external agency (e.g., an insurance com pany).
1123-77375_ch01_5P.indd 14 12/21/18 11:02 AM
Risk 15
4. Avoid risks altogether (e.g., by closing down a par tic u lar high- risk business area).
The assessment is not always in financial terms. It may be safety, gov-ernment regulations, customer demands, and so on. If the assessment is a financial one, make sure an appropriate cost- benefit analy sis is performed. To be sure that the appropriate assessment is being evaluated, each option requires developing a plan that is implemented and monitored for effective-ness (i.e., reduction or mitigation of risk). Typical guides for this evaluation of increasing seriousness of the risk (ISO 31000:2009, revised with ISO 31000:2018: Risk management— Princi ples and guidelines on implementa-tion) are:
1. Intensified technical and management reviews of the engineering pro cess
2. Special oversight of designated component engineering
3. Special analy sis and testing of critical design items
4. Rapid prototyping and test feedback
5. Consideration of relieving critical design requirements
6. Initiation of fallback parallel developments
Essentially, the key capabilities of a risk- enabled quality management solution include but are not limited to the following:
• An integrated risk register. There needs to be a centralized place to rec ord and monitor individual hazards and risk items. The importance here is to demonstrate consistency. At minimum, this document should include dates, description of risk, risk type (busi-ness, proj ect, and stage), severity of effect, likelihood of occurrence (Low <30%, Medium 31–75%, or High >75%), countermea sures, status, and any other quantitative value that may be applicable and appropriate. A typical risk register based on the work of Kokcharov (2015) is shown in Table 1.1.
• Flexible risk tools. A matrix is needed to be able to activate risk assessment tools, to demonstrate the link between audits to risk deviations or regulatory requirements, and to track the effectiveness of the tools as they are applied to specific risks.
• Risk- based effectiveness checks. This requirement is used for ver-ifying corrective action and evaluating improvement requirements.
1123-77375_ch01_5P.indd 15 12/21/18 11:02 AM
16 Chapter One
Risk Management Plan
Select appropriate controls or countermea sures to mea sure each risk. Risk mitigation needs to be approved by the appropriate level of management. For instance, a risk concerning the image of the organ ization should have top management decision behind it whereas IT management would have the authority to decide on computer virus risks.
The risk management plan should propose applicable and effective security controls for managing the risks. For example, an observed high risk of computer viruses could be mitigated by acquiring and implementing antivirus software. A good risk management plan should contain a schedule for control implementation and the persons responsible for those actions.
According to ISO/IEC (International Electrotechnical Commission) 27001, the stage immediately after completion of the risk assessment phase consists of preparing a risk treatment plan, which should document the deci-sions about how each of the identified risks should be handled. Mitigation of risks often means se lection of security controls, which should be docu-mented in a statement of applicability that identifies which par tic u lar con-trol objectives and controls from the standard have been selected and why.
Potential Risk Treatments
Once risks have been identified and assessed, all techniques to manage the risk fall into one or more of these four major categories (Stamatis 2014; Dorf-man 2007):
1. Avoidance (eliminate, withdraw from, or not become involved). This includes not performing an activity that could carry risk. An example would be not buying a machine with high life- cycle costs even though the original price of the machine may be less than a machine with low life- cycle costs, even though the original price may be higher.
Table 1.1 A typical risk register for a proj ect that includes four steps: identify, analyze, plan response, monitor and control.
Identify AnalyzePlan
ResponseMonitor and
Control
ID Description Category ProbabilityEstimated
impact
Additional workload
daysStatus
Response cost
Costs
1123-77375_ch01_5P.indd 16 12/21/18 11:02 AM
Risk 17
2. Reduction (optimize and mitigate). Risk reduction or optimization involves reducing the severity of the loss or the likelihood of the loss from occurring. A classic example of this is whether or not to buy water sprin-klers that are designed to put out a fire and therefore reduce the risk of loss by fire. However, by installing the sprinklers you may cause another loss due to water damage, which may be a higher cost and not suitable for the bud get of the organ ization. The result may be in finding another alternative.
Yet another approach to reducing risk is by outsourcing it (Stamatis 2016, 2014; Roehrig 2006). A typical method is to insure the loss through a third party.
3. Sharing (transfer, outsource, or insure). This option is briefly defined as “sharing with another party the burden of loss or the benefit of gain, from a risk, and the mea sures to reduce a risk.”
The term “risk transfer” is often used in place of risk sharing in the mis-taken belief that you can transfer a risk to a third party through insurance or outsourcing. In practice, if the insurance com pany or contractor go bankrupt or end up in court, the original risk is likely to still revert to the first party. As such, in the terminology of prac ti tion ers and scholars alike, the purchase of an insurance contract is often described as a “transfer of risk.”
4. Retention (accept and bud get). As unthinkable as it may be, the fact is that risk retention is a real ity in every thing we do. It involves accepting the loss, or benefit of gain, from a risk when it occurs. In our common lan-guage we call it self- insurance. In other words, we will take a chance that the failure will not happen or if it does, it will not be detrimental to our organ-ization. Risk retention is a viable strategy for small risks where the cost of insuring against the risk would be greater over time than the total losses sus-tained. All risks that are not avoided or transferred are retained by default.
Ideal use of these risk control strategies may not be pos si ble. Some of them may involve trade- offs that are not acceptable to the organ ization or person making the risk management decisions. Another source of control may be the approach that the US Department of Defense takes in using the avoid- control- accept- transfer (ACAT) model.
Review and Evaluation of the Plan
Initial risk management plans will never be perfect. This is because we are not able to foresee all risks and all potential prob lems that may occur in the short or long term. Things happen unexpectantly and when they do, hope-fully we learn, adjust, modify, and/or completely change the task, system, or pro cess in the organ ization. Obviously, practice, experience, and actual loss results will necessitate changes in the plan and contribute information
1123-77375_ch01_5P.indd 17 12/21/18 11:02 AM
18 Chapter One
to allow pos si ble dif fer ent decisions to be made in dealing with the risks being faced.
Risk analy sis results and management plans should be updated periodi-cally for two primary reasons:
1. To evaluate whether the previously selected security controls are still applicable and effective.
2. To evaluate the pos si ble risk- level changes in the business envi-ronment. For example, information risks are a good example of rapidly changing business environments.
Limitations
Prioritizing the risk management pro cesses too highly could keep an organ-ization from ever completing a proj ect or even getting started. This is espe-cially true if other work is suspended until the risk management pro cess is considered complete.
It is also impor tant to keep in mind the distinction between risk and uncertainty. Risk can be mea sured by impacts times probability. If risks are improperly assessed and prioritized, time can be wasted in dealing with risk of losses that are not likely to occur. Spending too much time assessing and managing unlikely risks can divert resources that could be used more profit-ably. Unlikely events do occur, but if the risk is unlikely enough to occur it may be better to simply retain the risk and deal with the result if the loss does in fact occur. Qualitative risk assessment is subjective and lacks consistency. The primary justification for a formal risk assessment pro cess is legal and bureaucratic.
CONCLUSION
Risk- based thinking is an ele ment in the pro cess approach. That is:
• Risk- based thinking is an input to management review.
• Risk- based thinking is an ele ment in the continual improvement pro cess that is focused on prevention.
Risk- based thinking has been demonstrated during audits; a risk regis-ter is documented with information that validates an organ ization has done risk- based thinking.
1123-77375_ch01_5P.indd 18 12/21/18 11:02 AM
19
OVERVIEW
When one talks about reliability the implication is that there is some speci-fication for a product such that no failures will be pres ent in the system, subsystem, component, or pro cess. Therefore, reliability is an engineering discipline that focuses on prevention of failures by design, people, hard-ware, production, and maintenance personnel or pro cesses.
It is impossible to create something that is 100% reliable because reli-ability at time t is equal to 1 minus the failure rate: R(t) = 1 − F(t). Therefore, in practice we achieve acceptable failures only, and only if the risk is miti-gated to provide benefits that are within the definition of the acceptability guidelines that the customer and/or the team has identified. This is pos si ble if we have a thorough understanding of all the potential failure modes and then take appropriate steps to prevent them from occurring. Understanding potential failure modes is achieved by analyzing and testing during both the design and the production phases of a proj ect. Of course, there are several ways to do the analy sis. Here we focus on the FMEA. Other ways include but are not limited to:
• Reliability- centered maintenance (RCM): program development and implementation
• Equipment criticality analy sis
• Reliability engineering analy sis and support, including FMEA, failure code development, root- cause analy sis, and lean tools such as 6S (sort-store-shine-standardize-sustain-safety)
• Reliability engineering training: pro cesses, methods, and tools
• Preventive maintenance optimization
2Reliability and FMEA
1123-77375_ch01_5P.indd 19 12/21/18 11:02 AM
20 Chapter Two
• Predictive maintenance program
• Calibration program and optimization
• Maintenance engineering staff augmentation: planners, schedulers, work preparers, maintenance supervisors, RCM engineers
NEED TO UNDERSTAND THE CONCEPT OF FAILURE
Criteria
A failure by strict definition is a deviation from a standard and/or specifica-tion. However, the criteria for defining a failure are heavi ly dependent on context of use and may be relative to a par tic u lar observer or belief system. A situation considered to be a failure by one might be considered a success by another, particularly in cases of direct competition or a zero- sum game. Similarly, the degree of success or failure in a situation may be viewed dif-ferently by distinct observers or participants, such that a situation that one considers to be a failure another might consider to be a success, a qualified success, or a neutral situation.
It may also be difficult or impossible to ascertain whether a situation meets criteria for failure or success due to ambiguous or ill- defined defini-tions of those criteria. Finding useful and effective criteria, or heuristics, to judge the success or failure of a situation may itself be a significant task. That task depends on clear, simple, and concise operational definition as well as a team that has both knowledge and some owner ship ( either direct or indirect) of the system, subsystem, or component under consideration.
Therefore, in cases where there is a difference of opinion about the fail-ure, the FMEA team should decide to treat the failure under discussion in the most conservative way. This means that the failure exists for all customers.
Types
Once the criteria for failure have been identified, then the team is ready to proceed with the analy sis, always remembering that failure can be dif-ferentially perceived from the viewpoint of the evaluator. A person who is only interested in the final outcome of an activity would consider it to be an outcome failure if the core issue has not been resolved or a core need is not met. A failure can also be a pro cess failure, whereby although the activity is completed successfully, a person may still feel dissatisfied if the under-
1123-77375_ch01_5P.indd 20 12/21/18 11:02 AM
Reliability and FMEA 21
lying pro cess is perceived to be below an expected standard or benchmark. Fundamentally, there are three types of failures:
1. Failure to perceive
2. Failure to anticipate
3. Failure to carry out a task
The first two generally account for the concept and design FMEAs. The third one accounts for pro cess and ser vice FMEAs.
DESIGN FOR RELIABILITY
Reliability is of course an issue of design. Therefore, to minimize failures a good design must have at least the following items evaluated before the release of that design. The reader should note that the steps identified here are indeed part of a detailed FMEA analy sis. The steps are:
• Step 1: Design for maintainability.
• Step 2: Perform functional analyses to determine failure modes as well as their consequences, severity, and ways of early detection.
• Step 3: Analyze components with potential failures and determine their failure models.
• Step 4: Determine maintenance tasks, their frequency, and their effectiveness.
• Step 5: Define and optimize maintenance implementation plan.
1123-77375_ch01_5P.indd 21 12/21/18 11:02 AM
1123-77375_ch01_5P.indd 22 12/21/18 11:02 AM
23
CREATING AN EFFECTIVE TEAM
Perhaps one of the most impor tant issues in dealing with the FMEA is that this analy sis must be done with a team. An FMEA completed by an indi-vidual is only that individual’s opinion and does not meet the requirements or the intent of an FMEA.
An effective FMEA team:
• Has expertise in the subject (five to seven individuals)
• Is multilevel/consensus- based
• Includes representatives of all relevant stakeholders ( those who have owner ship)
• May change membership as work progresses
• Is cross- functional and multidisciplinary (one person doing his or her best cannot approach this level of knowledge)
• Allows appropriate and applicable empowerment
• Includes the operator for PFMEA
THE STRUCTURE OF THE FMEA TEAM
Core Team
The experts of the proj ect and those closest to the proj ect are the core team. They facilitate honest communication and encourage active participation. Support membership may vary depending on the stage of the proj ect. The
3Prerequisites of FMEA
1123-77375_ch01_5P.indd 23 12/21/18 11:02 AM
24 Chapter Three
leader for the DFMEA should be the design engineer and for the PFMEA the manufacturer engineer.
Champion/Sponsor
This team member:
• Provides resources and support
• Attends some meetings
• Supports the rest of the team
• Promotes team efforts and implements recommendations
• Shares authority/power with other team members
• Kicks off the team
• Comes from management (the higher up in management the better)
• Breaks down any bottlenecks that may surface
Team Leader
A team leader is the “watchdog” of the proj ect. Typically, this function falls on the lead engineer. A good team leader:
• Possesses good leadership skills
• Is respected by team members
• Leads but does not dominate
• Maintains full team participation
Recorder
This person documents the team’s efforts. The recorder is responsible for coordinating meeting rooms and times as well as distributing meeting minutes and agendas.
Facilitator
The watchdog of the pro cess, the facilitator keeps the team on track and makes sure that every one participates. In addition, it is the facilitator’s responsibility to make sure that team dynamics develop in a positive envi-ronment. For the facilitator to be effective it is imperative that this indi-
1123-77375_ch01_5P.indd 24 12/21/18 11:02 AM
Prerequisites of FMEA 25
vidual has no stake in the proj ect, possesses FMEA pro cess expertise, and communicates assertively.
Team Considerations
• Continuity of members
• Receptiveness and open- mindedness
• Commitment to success
• Empowered by the sponsor
• Cross- functionality
• Multidisciplinary
• Consensus- based
• Positive synergy
INGREDIENTS OF A MOTIVATED FMEA TEAM
To motivate people:
• Set realistic agendas.
• Have a good facilitator.
• Keep meetings short.
• Have the right people pres ent.
• Reach decisions based on consensus.
• Have open- minded, self- initiators, and volunteers on the team.
• Offer incentives.
• Establish ground rules.
• Make one individual responsible for coordination and accountabil-ity of the FMEA proj ect. Typically for the design, the design engi-neer is that person, and for the pro cess, the manufacturing engineer accounts for that responsibility.
To make sure the effectiveness of the team is sustained throughout the proj ect, it is imperative that every one concerned with the proj ect bring useful
1123-77375_ch01_5P.indd 25 12/21/18 11:02 AM
26 Chapter Three
information in the pro cess. Useful information may be derived from one’s education, experience, training, or a combination of these factors. At least three areas that are usually underutilized for useful information are (1) back-ground information, (2) surrogate data, and (3) the input of the operator.
1. Background information and supporting documents that may be helpful to complete the system, design, or pro cess FMEAs include:
– Customer specifications (OEMs)
– Previous or similar FMEAs
– Historical information on warranty/recalls, etc.
– Design reviews and verification reports
– Product drawings/bill of material
– Pro cess flowcharts/manufacturing routing
– Test methods
– Preliminary control and gage plans
– Maintenance history
– Pro cess capabilities
2. Surrogate data generated from similar proj ects may help in the initial stages of the FMEA. When surrogate data are used, extra caution should be taken. The surrogate data should be replaced with the actual data as soon as pos si ble
3. Operator input is very essential. As the person the closest to the operation, the operator is most qualified to discuss assignable causes.
POTENTIAL FMEA TEAM MEMBERS
The actual team composition for your organ ization will depend on your individual proj ect and resources. An appropriate team for your proj ect may include any of the following:
• Design engineers
• Manufacturing engineers
• Quality engineers
1123-77375_ch01_5P.indd 26 12/21/18 11:02 AM
Prerequisites of FMEA 27
• Test engineers
• Reliability engineers
• Maintenance personnel
• Operators (from all shifts)
• Equipment suppliers
• Customers
• Suppliers
• Anyone who has a direct or indirect interest in the proj ect:
– In any FMEA team effort, the individuals must have interaction with manufacturing and/or pro cess engineering while conduct-ing a DFMEA. This is impor tant to ensure that the pro cess will manufacture per design specification.
– On the other hand, interaction with design engineering while conducting a PFMEA or assembly FMEA is impor tant to ensure that the design is right.
– In either case, team consensus will identify the high risk areas that must be addressed to ensure that the design and/or pro cess changes are implemented for improved quality and reliability of the product.
Once the team is chosen for the given proj ect, spend 15 to 20 minutes creating a list of the biggest (however you define “biggest”) concerns for this product or pro cess. This list will be used later to make sure we have a complete list of functions.
MIND-SET OF MINIMIZING FAILURES
Another prerequisite for conducting an FMEA is to recognize that failures should be eliminated and/or minimized. As noble a goal as that proposition is, we all know that it is difficult to achieve. So, what we often end up doing is minimizing as much as pos si ble the potential for any system, pro cess, subsystem, or component failure. This is a team trade- off that may be dif-ficult to achieve. Remember, a failure is a nonconformance from a standard and/or a specification. These nonconformances may or may not be a con-cern for the customer and/or design and/or the pro cess. In fact, the design and/or pro cess may indeed operate with a given nonconformance.
1123-77375_ch01_5P.indd 27 12/21/18 11:02 AM
28 Chapter Three
Even though the previous statement is correct, it is imperative that all of us should be concerned with failures. Our goal is and should be to have an attitude of failure- free designs and pro cesses. This will facilitate cus-tomer satisfaction and improve efficiency within the organ ization that is undertaking the FMEA practice.
In the end, this translates to more profitability!
1123-77375_ch01_5P.indd 28 12/21/18 11:02 AM
29
DEFINITION
FMEA is an engineering “reliability tool” that:
1. Helps to define, identify, prioritize, and eliminate known and/or potential failures of the system, design, or manufacturing pro cess before they reach the customer. The goal is to eliminate the failure modes or reduce their risks.
2. Provides structure for a cross- functional critique of a design or a pro cess.
3. Facilitates interdepartmental dialogue. (It is much more than a design review.)
4. Creates the mental discipline that “ great” engineering teams go through when critiquing what might go wrong with the design, product, or pro cess.
5. Is a living document that reflects the latest design, product, and pro cess actions.
6. Ultimately helps prevent, and not react to, prob lems.
7. Identifies potential product- or process- related failure modes before they happen.
8. Determines the effect and severity of these failure modes.
9. Identifies the causes and probability of occurrence of the failure modes.
10. Identifies the “controls” and their effectiveness.
4What Is an FMEA?
1123-77375_ch01_5P.indd 29 12/21/18 11:02 AM
30 Chapter Four
11. Quantifies and prioritizes the risks associated with the failure modes.
12. Develops and documents action plans that will occur to reduce risk.
IS FMEA NEEDED?
If any answer of the following questions is positive, then you need an FMEA:
• Are customers becoming more quality conscious?
• Are reliability prob lems becoming a big concern?
• Are regulatory requirements harder to meet?
• Are you doing too much prob lem solving?
• Are you addicted to prob lem solving? This is a very impor tant con-sideration in the application of an active FMEA program because when the thrill and excitement of solving prob lems become domi-nant, your organ ization is addicted to prob lem solving rather than preventing the prob lem to begin with. A proper FMEA will help break your addiction by:
– Reducing the % time to prob lem solving
– Increasing the % time in prob lem prevention
– Increasing the efficiency of resource allocation
Note: Emphasis is always on reducing complexity and engineering changes.In more general terms we need an FMEA to emphasize the need to
improve our designs and/or pro cesses to be more effective (satisfy our cus-tomers) and efficient (optimize our resources). However, the most impor-tant reason for conducting an FMEA is the need to improve. This strongly implies that in order to receive all or some of the benefits of an FMEA program, the need to improve must be ingrained in the organ ization’s cul-ture. If not, the FMEA program will not succeed. Therefore, a successful FMEA is a customer, com pany, and supplier requirement for a world- class quality. Specifically, any FMEA can help the improvement pro cess in the following areas:
• Superior competitive advantage:
– Best- in- class value
– Quality per for mance
1123-77375_ch01_5P.indd 30 12/21/18 11:02 AM
What Is an FMEA? 31
– Sustainable cost advantage
– Flawless launch at the start of production or commencement of a program
• Superior orga nizational capability:
– Brings best- in- class design
– Uses or achieves breakthrough technology
– Moves fast
• Superior culture:
– “Can do” attitude
– An obsession with continual improvement
– Team spirit
– Knowing how to say “no” the right way
This translates into:
• Faster development time
• Reduction of overall cost
• Improved quality throughout the life of the product and/or ser vice
BENEFITS OF FMEA
When properly conducted, all types of FMEAs should lead to:
1. Confidence that all (reasonable) risks have been identified early and appropriate actions taken
2. Priorities and rationale for product and pro cess improvement actions
3. Reduction of scrap, rework, and manufacturing costs
4. Preservation of product and pro cess knowledge
5. Reduction of field failures and warranty cost
6. Document risks and actions for future designs and/or pro cesses
1123-77375_ch01_5P.indd 31 12/21/18 11:02 AM
32 Chapter Four
THE PROCESS OF CONDUCTING AN FMEA
To conduct an FMEA effectively one must follow a systematic approach. The recommended approach is an eight- step method that facilitates the sys-tem, design, product, pro cess, equipment, and ser vice FMEA. The steps are:
1. Select the team and brainstorm. Make sure the appropriate individuals are going to participate. The team must be cross- functional and multidisci-plined and the team members must be willing to contribute (i.e., share their experience and knowledge).
After the team has been identified and is in place, the team tries to pri-oritize the opportunities of improvement. Is the concern in a system, design, product, pro cess, or ser vice? What kind of prob lems are there and/or what kind of prob lems are anticipated with a par tic u lar situation? Is the customer and/or supplier involved or is continual improvement being pursued in de-pen dently? If the customer and/or supplier identified specific failures, then the job is much easier because direction has already been given. On the other hand, if continual improvement is being in de pen dently pursued, the brainstorm, affinity diagram, story book method, and/or cause- and- effect diagram may prove to be the best tools to identify some direction.
2. Create a functional block diagram and/or pro cess flowchart. For sys-tem and design FMEAs, the functional block diagram is applicable. For the pro cess and ser vice FMEAs, the pro cess flowchart is applicable. The idea is to make sure that every one is on the same wavelength. Does every-one understand the system, design, pro cess, and/or ser vice? Does every one understand the prob lems associated with the system, design, pro cess, and/or ser vice?
The functional block diagram focuses the discussion on the system and design while the pro cess flowchart focuses the discussion on the pro cess and ser vice. Both of these tools also provide an overview and a working model of the relationships and interactions of the systems, subsystems, components, pro cesses, assemblies, and/or ser vices and help in the under-standing of the system, design, product, pro cess, and/or ser vice.
3. Prioritize. After the team understands the background, the actual analy sis begins. Frequent questions are: What part is impor tant? Where should the team begin?
Sometimes, this step is completely bypassed because the prioritization is de facto. The customer has identified the priority, or due to warranty cost or some other input the determination has been made by management to start at a given point.
1123-77375_ch01_5P.indd 32 12/21/18 11:02 AM
What Is an FMEA? 33
4. Begin data collection. The team must collect data of the failures and categorize them appropriately. At this point the team begins to fill in the FMEA form. The failures identified are the failure modes of the FMEA.
5. Analyze the data. Now the data are utilized for a resolution. Remem-ber, the reason for the data is to gain information that is used to gain knowl-edge. Ultimately, that knowledge contributes to the decision. This flow can be shown as follows:
Data Information Knowledge Decision
Flow
The analy sis may be qualitative or quantitative. The team may use brainstorming, cause- and- effect analy sis, quality function deployment (QFD), design of experiments (DOE), SPC, another FMEA, mathematical modeling, simulation, reliability analy sis, and anything else that team members think is suitable.
Information from this step will be used to fill in the columns of the FMEA form in relationship to the effects of the failure, existing controls, and the estimation of severity, occurrence, and detection.
6. Rec ord results. The theme here is data driven. Based on the analy sis, results are derived. The information from this step will be used to quantify the severity, occurrence, detection, and risk priority number (RPN). The appropriate columns of the FMEA will be completed.
7. Confirm/evaluate/mea sure. After the results have been recorded, it is time to confirm, evaluate, and mea sure the success or failure. This evalua-tion takes the form of three basic questions.
– Is the situation better than before?
– Is the situation worse than before?
– Is the situation the same as before?
The information from this step will be used to recommend actions and to see the results of those actions in the corresponding columns of the FMEA form.
8. Do it all over again. Regardless of how step 7 is answered, the team must pursue improvement all over again because of the under lying philoso-phy of FMEA, which is continual improvement.
The long- term goal is to completely eliminate every single failure. The short- term goal is to minimize the failures if not eliminate them. Of course, the perseverance for those goals has to be taken into consideration in rela-tionship to the needs of the organ ization, costs, customers, and competition.
1123-77375_ch01_5P.indd 33 12/21/18 11:02 AM
34 Chapter Four
The philosophy of continual improvement embedded in the FMEA is that all types of FMEAs are living documents. ( Here we must note however that due to the short- term versus long- term expectations, the short- term results may be sufficient; therefore the team may change to a new one so that the long- term results may come to fruition.)
Getting Started
As with anything else, before the FMEA begins there are some assump-tions and preparations that must be taken care of. These assumptions are:
1. Define the FMEA proj ect and scope.
2. Know your customers and their needs.
3. Know the function.
4. Understand the concept of priority.
5. Develop and evaluate conceptual designs/pro cesses based on your customer’s needs and business strategy.
6. Be committed to continual improvement.
7. Create an effective team.
A general overview of the DFMEA and PFMEA may be seen in Figures 4.1 and 4.2, respectively.
Design FMEATeam
Scope Boundary diagram
Function Function tree
Failure modes
Path 1
Path 2
Path 3
EffectsEffects list
CauseIshikawa diagram
ControlsPreventive/detective
SeverityRanking table
OccurrenceRanking table
DetectionRanking table
ClassYC (S = 9 or 10)
ClassYS (S = 5 to 8 and O > 3)Action
Action
Action
Figure 4.1 Overview of a DFMEA
1123-77375_ch01_5P.indd 34 12/21/18 11:02 AM
What Is an FMEA? 35
Timing
The FMEA should be performed and/or updated whenever:
• A new cycle begins (new product/pro cess).
• Changes are made to the operating conditions.
• A change is made in the design/pro cess.
• New regulations are instituted.
• Customer feedback indicates a prob lem.
Uses
Among the primary uses of an FMEA are:
• Development of system requirements that minimize the likelihood of failures
• Development of designs and test systems to ensure that the failures have been eliminated or the risk is reduced to acceptable level
• Development and evaluation of diagnostic systems
• To help with design choices (trade- off analy sis)
Process FMEATeam
Scope Process flow—macro and micro
Function Purpose statement
Failure modes
Path 1
Path 2
Path 3
EffectsEffects list
CauseIshikawa diagram(MEPEM)ControlsPreventive/detective
SeverityRanking table
OccurrenceRanking table
DetectionRanking table
Class (S = 9 or 10)(when confirmed)ClassSC (S = 5 to 8 and O > 3)(when confirmed)Action
Action
Action
Figure 4.2 Overview of a PFMEA
1123-77375_ch01_5P.indd 35 12/21/18 11:02 AM
36 Chapter Four
Advantages
The pro cess of conducting an FMEA can:
• Improve the quality, reliability, and safety of a product/pro cess
• Improve a com pany’s image and competitiveness
• Increase user satisfaction
• Reduce system development time and cost
• Collect information to reduce future failures and capture engineer-ing knowledge
• Reduce the potential for warranty concerns
• Lead to early identification and elimination of potential failure modes
• Emphasize prob lem prevention
• Minimize late changes and associated cost
• Be a catalyst for teamwork and idea exchange between functions
• Reduce the possibility of same kind of failure in future
• Reduce the impact on a com pany’s profit margin
• Improve production yield
UNDERSTANDING YOUR CUSTOMERS AND THEIR NEEDS
A product or a pro cess may perform functions flawlessly, but if the func-tions are not aligned with customer needs, you may be wasting your time. Therefore, you must:
• Determine all (internal and/or external) relevant customers.
• Understand the customer’s needs better than the customers under-stand their own needs.
• Document the customer’s needs and develop concepts. For example, customers may need:
– Chewable toothpaste
– Smokeless cigarettes
1123-77375_ch01_5P.indd 36 12/21/18 11:02 AM
What Is an FMEA? 37
– Celery- flavored gum
– Other (identify and then document your customers’ specific needs)
In FMEA, a customer is anyone/thing that has functions/needs from your product or manufacturing pro cess. An easy way to determine cus-tomer needs is to understand the Kano model and QFD, especially for design issues.
WHAT HAPPENS AFTER COMPLETION OF THE FMEA?
Generally there are seven steps that the team must follow:
1. Review the FMEA. Make sure that the function, purpose, and objec-tive have been met. Make sure that all the loose ends have been addressed and the appropriate action has been recommended and/or implemented. Some helpful hints for this review follow.
• Is the prob lem identification specific?
• Was a root cause, an effect, or a symptom identified?
• Is the corrective action mea sur able?
• Is the corrective action proactive?
• Is the use of terminology current and consistent?
• Is the corrective action realistic and sustainable?
• Has a control plan been developed and linked to the critical as well as significant characteristics in the FMEA?
2. Highlight the high- risk areas. A visual inspection of the critical col-umn, the severity column, and the RPN column generally will identify the high- risk areas (see Figure 1.1). In the critical column, the high- risk item may be identified as such; in the severity column the high- risk item usually will have a number higher or equal to 7; and in the RPN column usually a number greater than or equal to 100 (on a 1 to 10 scale) will indicate that there might be a high- risk item. In some industries this is not recognized as a valid identification pro cess for high- risk items. In some cases, the high- risk item is identified by the numerical value of severity regardless of the value of RPN (see step 3).
3. Identify the critical, significant, and major characteristics. Upon completion of the FMEA, a visual check of the RPN and critical columns
1123-77375_ch01_5P.indd 37 12/21/18 11:02 AM
38 Chapter Four
should identify the critical, significant, and major characteristics. Make sure that there is a direct correlation between the critical column and the effects of the failure and the severity columns. Great care should be taken when reviewing the RPN because these numbers will indicate whether or not action should be taken. Here we must emphasize that even though many industries use the RPN as a clearing point for evaluating risks (the higher the number the riskier the failure mode cause), there is a better way to do the evaluation based on (1) severity, (2) criticality (Severity × Occurrence), and (3) RPN (Severity × Occurrence × Detection).
4. Ensure that a control plan exists and is being followed. As previously mentioned, the idea behind performing an FMEA is to eliminate and/or reduce known and potential failures before they reach the customer. In this step, make sure that all critical, significant, and major characteristics have a documented plan for controlling, improving, and/or handling changes. The control plan is the map that will allow prac ti tion ers to make the product and/or ser vice acceptable to the customer. Although the FMEA identifies the vital signs of the pro cess and/or ser vice, the control plan monitors those vital signs of the pro cess and/or ser vice.
5. Conduct capability studies. After the control plan is in place and sta-tistical control has been established, a potential capability or a long- term capability must be performed.
6. Work on pro cesses that have a Cpk less than or equal to 1.33. Although the 1.33 generally is accepted as the minimum goal, be aware that some companies require a Ppk = 1.33 (namely, automotive companies) or even a Cpk = 2.00 (Motorola). The point is to continually improve the pro cess by eliminating variation. Produce every thing around the target.
7. Work on pro cesses that have Cpk or a Ppk greater than or equal to 1.33. After the minimum standard is reached in step 6, try to go beyond that standard for further improvement. Reduce variation and try to reach or exceed a Cpk or Ppk greater than or equal to 2.00. Remember, all standards are minimum per for mance. Consequently, continual improvement dictates that one should at all times try to exceed all standards, including all Cpk or Ppk targets.
VOCABULARY
As in every methodology, including the FMEA, there is special jargon that is used to communicate functions, failures, and appropriate actions to
1123-77375_ch01_5P.indd 38 12/21/18 11:02 AM
What Is an FMEA? 39
remove or minimize these failures. It is imperative therefore to be familiar with the vocabulary and its significance to the FMEA. Following is a list of key terms used in all FMEAs.
Function: What is the intent of the design or pro cess? Specifically, the following items should be addressed:
• Describe the design/pro cess intent or engineering requirement.
• Write it in verb- noun mea sur able format.
• Represent all wants, needs, and requirements, both spoken and unspoken for all customers and systems.
Failure Mode: How can this function fail? There are usually six mini-mum failures for each function. They are:
1. No function. It does not work.
2. Degradation. The function over time fails.
3. Intermittent. The function sometimes works and sometimes does not.
4. Partial. The function does not work at full cycle.
5. Unintended. The function acts in a surprise manner.
6. Over function. The function does more than intended.
Effect of Failure: Describe the consequence(s) of failure. Typical considerations for design are:
• Part/sub-component
• Next higher assembly
• System
• Total product (as in vehicle)
• Government regulations
• Customer (internal and end user)
Typical considerations for pro cess are:
• Operator safety
• Next user
• Downstream users
1123-77375_ch01_5P.indd 39 12/21/18 11:02 AM
40 Chapter Four
• Machines/equipment
• Total pro cess operation
• Ultimate customer
• Compliance with government regulations
Severity (S): How serious is the effect on the failure mode? Generally the severity is the worst numerical effect value. Severity is a rela-tive ranking within the scope of the individual FMEA. A reduction in severity ranking index can be effected only through a design change.
Classification: If severity values are 9 or 10, that is where safety and/or government regulations are effected. This means that the classi-fication column should reflect the potential critical characteristics. When that happens, the team must:
• Develop a proactive design recommended action.
• Ensure that information is communicated to the PFMEA after causes have been generated.
If the severity is >4, this implies that the item is significant and there-fore proactive actions should be recommended. In some industries if the severity is 4–8 and the occurrence is >3, the item is considered to be significant and appropriate proactive actions are necessary.
Pos si ble Cause(s): This is an indication of a design weakness, the consequence of which is the failure mode. In other words: What causes the function to fail? A good source for answering this ques-tion may be found in the P- diagram and the interface matrix. For design concerns, a good rule to follow is to assume that:
• The item is manufactured and assembled within engineering specifications.
• The design may include a deficiency that may cause unaccept-able variation (e.g., misbuilds, errors, etc.).
For pro cess concerns, a good rule to follow is to ask:
• If incoming parts are correct, what would cause the operation to fail in this manner?
• What incoming sources of variation could cause the operation to fail in this manner?
1123-77375_ch01_5P.indd 40 12/21/18 11:02 AM
What Is an FMEA? 41
Common ways to determine causes are:
– Brainstorming.
– 5 Whys method.
– Fishbone diagram.
– Fault tree analy sis (FTA). This model uses a tree to show the cause- and-effect relationship between a failure mode and the vari ous contributing causes. The tree illustrates the logical hier-archy branches from the failure at the top to the root causes at the bottom.
– Classic five- step problem- solving pro cess:
1. What is the prob lem?
2. What can I do about it?
3. Put a star on the “best” plan.
4. Do the plan!
5. Did your plan work?
– Kepner Tregoe method (analy sis that asks, what is the prob lem, what is not the prob lem).
– Discipline (8D).
– Experience:
• Knowledge of physics and the sciences
• Knowledge of similar products
– Experiments. When many causes are suspect or specific cause is unknown:
• Classical
• Taguchi methods
Occurrence (O): How often does the cause of the function happen? Occurrence is the likelihood that a specific cause/mechanism (listed in the previous column) will occur during the life of the function (design or pro cess). The likelihood of occurrence ranking number has a relative meaning rather than an absolute value. Pre-venting or controlling the causes/mechanisms of the failure mode through a design change or design pro cess change (e.g., design
1123-77375_ch01_5P.indd 41 12/21/18 11:02 AM
42 Chapter Four
checklist, design review, design guide) is the only way a reduction in the occurrence ranking can be effected. At this point the highest S × O (criticality) failure mode– cause combinations determine if an appropriate recommended action can be taken. An action should be posted for any cause that received an occurrence rating of 10 (where the team could not reach consensus or where occur-rence could not be estimated).
Prevention Controls: This item is for the planning controls in order to avoid the cause from happening or to reduce the rate of occurrence.
Detection Controls: This item identifies the effectiveness of the plan-ning controls, which may be analytical, physical methods, before the item is released to production.
Detection (D): Detection is the rank associated with the best type of design control from the list in the previous column. Detection is a relative ranking, within the scope of the individual FMEA. To achieve a lower ranking, generally the planned design control (e.g., validation and/or verification activities) has to be improved.
RPN: This is the result of S × O × D. Based on the highest number, the priority is set for recommended action. However, RPN is not always the best importance indicator since the severity or occur-rence sometimes dominates the RPN factor. A better way to set priorities for a completed FMEA might be to first use the high severity number (9 or 10 and, in some organ izations, 5 and higher by agreement of the customer and supplier) followed by the highest criticality indices (S × O) and then the highest RPN numbers. Each failure cause must have its own RPN calculated. Be sure to recog-nize that some failure modes have the same solution or follow-up activity.
Recommendations: These are actions that must be taken to minimize or eliminate the cause of the failure. To be effective the actions must be (1) appropriate, (2) applicable, (3) completed within a reasonable time, and (4) cost- effective. This means that each action must be assigned to an individual with a specific due date. If no action is planned, enter “None” or “None at this time.” All identified corrective actions should first be directed at the high-est ranked concerns and critical items. In fact, the aim should be prevention and not increasing detection methods.
Actions Taken: Here we identify the specific action that is taken from the list of the recommendations. An FMEA without posi-
1123-77375_ch01_5P.indd 42 12/21/18 11:02 AM
What Is an FMEA? 43
tive and effective actions to prevent failures is not of much use. Once the actions have been implemented, the estimated values or “the future” become “the pres ent” and can be incorporated into the left- hand side of the form on the next FMEA. After action has been taken, enter a brief description of the action and its effective or actual completion date. At this point re- rate the severity, occur-rence, or detection based on the actions taken and enter into the revised severity, revised occurrence or revised detection columns as applicable.
New Severity: In order for a new number to be entered here, some or all the following must happen. (1) Change the design, (2) change standards, (3) change procedures and/or instructions, (4) change policies, and (5) make pro cess changes. Warning! There are two schools of thought here. One is that once the severity is identified it remains the same unless the design is changed. The second is that the severity may change if redundant systems are in place and/or if there is a combination of the five items mentioned.
New Occurrence: The number may change if redundant systems or one of the following items is incorporated in the design or pro cess. (1) Change the design, (2) change standards, (3) change procedures and/or instructions, (4) change policies, and (5) make pro cess changes.
New Detection: The number may change if controls are added or one of the following items is incorporated in the design or pro cess. (1) Change the design, (2) change standards, (3) change procedures and/or instructions, (4) change policies, and (5) make pro cess changes.
New RPN: The number will change if any of the contributing factors change in any way. The factors that make up the RPN are severity, occurrence, and detection. Any change in these will change the RPN.
1123-77375_ch01_5P.indd 43 12/21/18 11:02 AM
1123-77375_ch01_5P.indd 44 12/21/18 11:02 AM
45
All FMEAs have a robustness focus. This means that robustness tools are inputs to a good FMEA. A pictorial view is shown in Table 5.1.
BOUNDARY DIAGRAM
The idea of a boundary diagram is to identify as well as represent other components in the higher- level assembly. Typically, the boundary diagram includes all system attachments and mechanisms as well as interfaces with:
• Other systems
• Manufacturing/assembly tools
• Servicing/customer adjustment
• All user interfaces
Once that repre sen ta tion has been accomplished then the boundary diagram is constructed, usually with dotted lines around the item of concern for the FMEA. This means that the boundary diagram considers what is best included and excluded in the analy sis of the par tic u lar FMEA. The boundary in essence has two functions: It is used (1) to aid the identifica-tion of the pos si ble effects of failures, and (2) once the scope is defined, to focus the support team in the pro cess of conducting the FMEA. A typical diagram is shown in Figure 5.1.
In this case items G, B, and C will be considered for the FMEA.
5Robustness
1123-77375_ch01_5P.indd 45 12/21/18 11:02 AM
46 Chapter Five
INTERFACE MATRIX
The interface matrix is used to identify and prioritize interactions between items of concern and their four par ameters. Specifically, an interface matrix:
• Acts as an input to a design FMEA.
• Identifies and quantifies the strength of system interactions by:
– Showing whether the relationship is necessary or adverse
– Identifying the type of relationship
A typical interface matrix is shown in Table 5.2.
Figure 5.1 A typical boundary diagram
A
F G H
DCB
Table 5.1 Robustness focus.
DFMEA with Robustness Linkages Pro cess
Boundary diagram Qualifies and clarifies the relationships between systems
Interface matrix Identifies and quantifies the strength of system interactions
P- diagram Identifies and quantifies the strength of system interactions
FMEA with robustness linkages
Robustness checklist
This is a DFMEA pro cess output. It summarizes error states, noise factors, and the associated design controls. It is also an input for the DVP.
Design verification plan (DVP)
1123-77375_ch01_5P.indd 46 12/21/18 11:02 AM
Robustness 47
P- DIAGRAM
The P- diagram is a method for identifying the ideal function as well as the par ameters that will prevent the ideal function from occurring. It is recom-mended for the design FMEA because it:
• Is a structured tool to identify intended inputs and outputs for a function
• Describes noise factors, control factors, ideal function, and error states
• Assists in the identification of:
– Potential Causes for Failure
– Failure Modes
Table 5.2 Interface matrix.
Individual Items of Concern A B C D
A P E P E P E P E
I M I M I M I M
B P E P E P E P E
I M I M I M I M
C P E P E P E P E
I M I M I M I M
D P E P E P E P E
I M I M I M I M
P = Physical touching
I = Information exchange
E = Energy transfer
M = Material exchange
Each of the letters may be represented with numbers such as:
+2 = interaction is necessary
+1 = interaction is beneficial but not absolutely necessary for functionality
0 = interaction does not affect functionality
−1 = interaction causes negative effects but does not prevent functionality
−2 = interaction must be prevented to achieve functionality
1123-77375_ch01_5P.indd 47 12/21/18 11:02 AM
48 Chapter Five
– Potential Effects of Failure
– Current Controls
– Recommended Actions
The relationship of input → pro cess → output is the ideal function. This means that all inputs are utilized for the expected output. No waste! The P- diagram is based on the following par ameters:
• Input refers to the items that are used in the pro cess. They may be manpower, machine, method, material, mea sure ment, or environment.
• Pro cess is the “value added” activity under consideration. It is the reason for existence.
• Output is the expected result of the pro cess.
• Errors are the items that contribute to less than 100% of output— the failures.
• Control items are the items that we can have in the pro cess to make sure that the output is at optimum— the actions that mitigate the failures so that the output is reached.
• Noise factors are the factors that contribute to customer usage, piece- to- piece variation, external environment, interactions, and changes over time without there being an adverse reaction to the pro cess.
A pictorial view is shown in Figure 5.2.
Figure 5.2 A typical P- diagram
Input Process
Noise
Output
ErrorsControl
1123-77375_ch01_5P.indd 48 12/21/18 11:02 AM
49
The generic form for all types of FMEA is very simple and straightforward. For certain industries, however, this form may be modified to reflect the needs of that industry. Figure 6.1 pres ents a generic form that identifies all needed information for reducing or eliminating a root cause from either a design and/or a pro cess. The reader should note that the only difference between the design and pro cess forms is the header that identifies whether it is a design or pro cess.
The rankings or criteria as they are commonly known are not stan-dardized. In other words, there are no criteria that every one is using for all FMEAs and industries. What is impor tant to know is that the criteria must be based on logic, knowledge, and experience about the task at hand. Having said that, it is also impor tant for the reader to rec-ognize that certain industries such as aerospace, nuclear, automotive, and others have indeed recommended criteria lists for severity, occurrence, and detection. If your industry has these guidelines and you want to devi-ate from them, it is acceptable to do so, but you must attach an adden-dum with the FMEA to show the dif fer ent criteria so that when someone else reads the FMEA, they will know the deviations from the suggested guideline.
In the section for DFMEA and PFMEA we will identify criteria that are considered very common and used in several industries. Here we sum-marize some key items of concern for any FMEA and pres ent a general strategy for reducing risk. A more detailed analy sis is covered in the section of the specific FMEAs.
6The FMEA Form and Rankings
1123-77375_ch01_5P.indd 49 12/21/18 11:02 AM
50 Chapter Six
SEVERITY RATING
Severity is a relative rating of the seriousness of the effect. Specifically:
• Severity is a numerical rating of the impact on customers.
• When multiple effects exist for a given failure mode, enter the worst- case severity on the worksheet to calculate risk. (This is the accepted method for the automotive industry and for the SAE J1739 standard. It should also be recognized that some companies, while they will accept this approach, prefer to have individual ratings for each of the effects.)
• In cases where severity varies depending on timing, use the worst- case scenario.
Reducing Severity (or Reducing the Severity of the Failure Mode Effect)
• Design or manufacturing pro cess changes are necessary.
• Focus on reducing severity in order of: changing design, pro cess, standards, procedures, and/or instructions.
Figure 6.1 A typical FMEA form
FMEA Form
Failu
reEf
fect
Seve
rity
Cla
ssC
ause
Occ
urre
nce
Prev
entio
n co
ntro
lsD
etec
tion
cont
rols
Det
ectio
nR
PNR
ecom
men
datio
nsAc
tions
take
nSe
verit
yO
ccur
renc
eD
etec
tion
RPN
System:Subsystem:Component:Process:
Function Comments
Page __ of __ Team: Originaldate:
Reviseddate:
FMEA number
1123-77375_ch01_5P.indd 50 12/21/18 11:02 AM
The FMEA Form and Rankings 51
OCCURRENCE RATING
This is the estimated number of frequencies or cumulative number of fail-ures (based on experience) that will occur (in our design concepts for a given cause over the “intended life of the design”). Example: The cause of staples falling out is soft wood. The likelihood of occurrence is a 9 if we picked balsa wood but a 2 if we choose oak.
Just like severity, there are standard tables for occurrence for each type of FMEA. The ratings on these tables are estimates based on experience and/or similar products or pro cesses. Nonstandard occurrence tables may also be used, based on specific characteristics. However, reliability expertise is needed to construct occurrence tables. (Typical characteristics may be historical failure frequencies, Cpks’ theoretical distributions, and reliability statistics.)
Reducing Occurrence (or Reducing the Frequency of the Cause)
• Design or manufacturing pro cess changes are necessary.
• Focus on reducing occurrence in order of: changing design, pro cess, standards, procedures, and/or instructions.
DETECTION RATING
Detection rating is a numerical rating of the probability that a given set of controls will discover a specific cause or failure mode to prevent bad parts from leaving the operation/facility or getting to the ultimate customer. Among the key concerns:
• Assuming that the cause of the failure did occur, assess the capabil-ities of the controls to find the design flaw or prevent the bad part from leaving the operation/facility. In the first case, the DFMEA is at issue. In the second case the PFMEA is of concern.
• When multiple controls exist for a given failure mode, rec ord the best (lowest) to calculate risk.
• In order to evaluate detection, there are appropriate tables for both design and pro cess. Just as before, however, if there is a need to alter them, remember that the change and approval must be done by the FMEA team with consensus.
1123-77375_ch01_5P.indd 51 12/21/18 11:02 AM
52 Chapter Six
Reducing Detection (or Increasing the Probability of Detection)
• Improving the detection controls is generally costly, reactive, and does not do much for qualify improvement, but it does reduce risk.
• Increased frequency of inspection, for example, should only be used as a last resort. It is not a proactive corrective action.
CLASSIFICATION AND CHARACTERISTICS
These characteristics must be classified according to risk impact:
• Severity 9, 10: Highest Classification (Critical) These kinds of characteristics are product- or process- related
that:
– May affect compliance with government or federal regulations (e.g., EPA, OSHA, FDA, FCC, FAA)
– May affect safety of the customer
– Require specific actions or controls during manufacturing to ensure 100% compliance
• Severity between 5–8 and Occurrence Greater than 3: Secondary Classification (Significant)
These characteristics are product- or process- related that:
– Are noncritical items that are impor tant for customer satisfaction (e.g., fit, finish, durability, appearance)
– Should be identified on drawings, specifications, or pro cess instructions to ensure acceptable levels of capability
• High RPN: Secondary Classification
Product Characteristics/“Root Causes”
• Examples include size, form, location, orientation, or other physical properties such as color, hardness, or strength.
Pro cess Pa ram e ters/“Root Causes”
• Examples include pressure, temperature, current, torque, speed, feeds, voltage, nozzle dia meter, time, chemical concentrations, cleanliness of incoming part, and ambient temperature.
1123-77375_ch01_5P.indd 52 12/21/18 11:02 AM
The FMEA Form and Rankings 53
UNDERSTANDING AND CALCULATING RISK
Without risk, there is very little pro gress! Risk is inevitable in any system, design, or manufacturing pro cess.
The FMEA pro cess aids in identifying significant risks, then helps to minimize the potential impact of risk. It does that through the risk priority number or, as it is commonly known, the RPN index. In the analy sis of the RPN, make sure to look at risk patterns rather than just a high RPN.
The RPN is the product of severity, occurrence, and detection or:
RISK = RPN = S × O × D
Obviously the higher the number of the RPN the more the concern. A good rule of thumb analy sis to follow is 95%. That means that you will address all failure modes with a 95% confidence. It turns out the magic num-ber is 50 [(S = 10 × O = 10 × D = 10) − (1000 × .95)]. This number of course is only relative to what the total FMEA is all about and it may change as the risk increases in all categories and in all causes.
Special risk priority patterns require special attention through specific action plans that will reduce or eliminate the high risk factor. They are identified through:
1. High RPN
2. Any RPN with a severity of 9 or 10 and an occurrence >2
3. Area chart
The area chart in Figure 6.2 uses only severity and occurrence, which is more proactive.
Figure 6.2 Area chart showing priority levels
Severity
11 2 3 4 5 6 7 8 9 10
23456789
10
Occ
urre
nce
Highpriority
Mediumpriority
Lowpriority
1123-77375_ch01_5P.indd 53 12/21/18 11:02 AM
54 Chapter Six
The reader should recognize that this is the traditional and most com-mon method of determining risk. However, there are other ways that are in fact more sensitive and beneficial. For example, the priority may be identi-fied by:
1. Severity ranking of >5
2. Severity 3–4 and occurrence >4 (criticality)
3. RPN
DRIVING THE ACTION PLAN
• For each recommended action, the FMEA team must:
– Plan for implementation of recommendations.
– Make sure that recommendations are followed, improved, and completed.
• Implementation of action plans requires answering the classic questions:
– Who ( will take the lead)?
– What (specifically is to be done)?
– Where ( will the work get done)?
– Why (which should be obvious)?
– When (should the actions be done)?
– How ( will we start)?
• Accelerate implementation by getting buy-in (owner ship).
• Draw out and address objections. When plans address objections in a constructive way, stakeholders feel owner ship in plans and actions. Owner ship aids in successful implementation!
• Typical questions that begin a fruitful discussion are:
– Why are we . . . ?
– Why not this?
– What about this?
– What if . . . ?
1123-77375_ch01_5P.indd 54 12/21/18 11:02 AM
The FMEA Form and Rankings 55
• Timing and actions must be reviewed on a regular basis to:
– Maintain a sense of urgency
– Allow for ongoing facilitation
– Ensure work is progressing
– Drive team members to meet commitments
– Surface new facts that may affect plans
• Fill in the actions taken:
– The Actions column should not be filled out before the actions are totally complete.
Rec ord final outcomes in the Action Plan and Action Results section of the FMEA form. Remember, because of the actions you have taken you should expect changes in severity, occurrence, detection, RPN, and new characteristic designations.
1123-77375_ch01_5P.indd 55 12/21/18 11:02 AM
1123-77375_ch01_5P.indd 56 12/21/18 11:02 AM
57
There are many types of FMEAs, each one specifically relating the causes of failures to the specific industry. For example, one may encounter an FMEA in areas such as phar ma ceu ti cal, environmental, industrial, defense, ser vice, healthcare, software, equipment, aerospace, automotive, petroleum, oil/gas, transportation, nuclear, marine, and many other specialized forums.
However, as variable as FMEAs may be, fundamentally they all are the same because they all try to prevent failures from happening or minimize their effect if they do. Because of the similarity in both analy sis and reac-tion approach, these dif fer ent FMEAs may be categorized primarily in the following categories shown in Table 7.1.
The reader will notice that even though we said that there are many FMEAs only five are identified in the table. The reason is because all others fall into either the design or pro cess category of FMEA. The difference is in the application and specific terminology used for the specific application. For example, in the phar ma ceu ti cal industry we may use an FMEA to:
• Implement a plan that introduces redundancy into the pro cess and interventions that are best suited to minimize risks of a product by using multiple stakeholders in the medi cation use pro cess.
• Prepare strategic contingency plans to respond promptly to Food and Drug Administration (FDA) questions and requests that come late in the approval pro cess.
• Establish a rigorous framework for the under lying approach to risk mitigation in the development of a proposed risk management pro-cess for a drug or biologic.
• Assess the risk management pro cess per for mance through identi-fied safety signals and adverse events of interest to assist in the
7Types of FMEA
1123-77375_ch01_5P.indd 57 12/21/18 11:02 AM
58 Chapter Seven
overall understanding of how, when, and where those risks may occur and ways to improve either the design and/or the pro cess.
FMEA CHALLENGES
FMEA is a living document and as such it must be reviewed and updated as needed or at least once a year. Because of this constant possibility of review, the pro cess of conducting an FMEA is considered to be a:
• Continuous brainstorming activity
• Lengthy consensus- building pro cess
• Pro cess that may not capture all pos si ble issues
• Team- dependent environment only
• Pro cess that determines and implements actions that drive reduction in risk
• Pro cess that ensures that the high- risk failure modes are addressed
• Pro cess that includes interfaces
Table 7.1 Types of FMEAs.
Types of FMEAs
Design Pro cess Ser vice Equipment Concept
Characteristics Component Machines Machines Component Component
Subsystem Methods Methods Subsystem Subsystem
System Material Material System System
Manpower Manpower
Mea sure ment Mea sure ment
Environment Environment
Focus Minimize failure effects on the system
Minimize production pro cess failure effects on the system
Minimize issues and prob lems interfering with the ser vice
Minimize safety issues
Minimize failure effects on the system, pro cess, or product from safety and government regulations
Objective Maximize system quality, reliability cost, and maintainability
Maximize the system quality, reliability, cost, maintainability, and productivity
Maximize quality and customer satisfaction
Minimize production issues and protect operator safety
Maximize system quality, reliability, cost, and maintainability in design, pro cess, or product
1123-77375_ch01_5P.indd 58 12/21/18 11:02 AM
59
There are many types of FMEAs. However, the most common ones are (1) concept, (2) design, and (3) pro cess. All of them, without exception, fol-low the same methodology except for specific failures in the specific indus-tries. For example, failures in the nuclear industry will be dif fer ent from the health industry’s failures, and in turn they will be dif fer ent from the automotive industry’s failures.
CONCEPT
Purpose
Fundamentally any FMEA analy sis is a risk assessment methodology. However, in the case of a concept FMEA (CFMEA), the focus is in the feasibility phase for the new, innovative or updated designs. In a CFMEA only potential customers are considered.
Use
A CFMEA is used as part of an early engineering assessment to identify the potential feasibility of a system/subsystem/component.
Benefit
The CFMEA is a way to test “what-if situations” for new, revolutionary, inno-vative ways of doing things. The benefits of an upfront CFMEA are:
• It identifies the success of potentiality of engineering as well as economic feasibility.
8The Common Types
of FMEAs
1123-77375_ch01_5P.indd 59 12/21/18 11:02 AM
60 Chapter Eight
• It identifies the necessity for redundant systems in the design.
• It identifies potential interaction and adverse effects of system/subsystem/components.
• It helps in the se lection of optimized alternatives for a par tic u lar design.
• It helps in identifying as early as pos si ble all potential effects of a proposed concept’s failure modes.
• It identifies potential system- level testing requirements.
• It helps determine the serious and/or catastrophic failures for the system/subsystem/component.
Form and Risk Criteria
For all intents and purposes, the form for the CFMEA is exactly the same as the one used for the DFMEA. However, most CFMEAs are never completed because of timing requirements. They usually stop after identifying major shortcomings in the proposed design, such as safety and/or government regulations. Another reason is that the timing requirements are overlapping with the DFMEAs and therefore the DFMEA is completed instead.
The criteria are also the same as that for the DFMEA. Very seldom they will be dif fer ent. If they are, it is to accommodate the specific require-ments of the industry.
Tools
• Computer simulation
• Functional diagrams
• Mathematical models
• Force field analy sis
• Breadboard tests
• Laboratory tests on surrogate ele ments
• QFD
• Benchmarking
• Internal past corporate knowledge
1123-77375_ch01_5P.indd 60 12/21/18 11:02 AM
The Common Types of FMEAs 61
DESIGN
Purpose
The purpose of a DFMEA is to perform a risk analy sis of all reasonable design flows of the proposed product prior to manufacturing. To do this there are two assumptions in determining flaws/failures:
1. Item is manufactured and assembled within engineering specifications.
2. Design may include a deficiency that may cause unacceptable varia-tion (misbuilds, errors, etc.) and may be addressed in the PFMEA.
Use
With an appropriate and applicable team in place, the primary use of DFMEA is to facilitate the following:
• Prevention planning
• Changing requirements
• Cost reduction
• Increased throughput
• Decreased waste
• Decreased warranty costs
• Reduction of non- value- added operations
Benefit
There are many benefits to conducting a DFMEA. Among the key benefits are the following:
• Increases the probability that potential failure modes and their effects have been considered in the design/development pro cess
• Helps in the objective evaluation of design requirements and design alternatives
• Establishes a priority system for design improvements based on potential failure modes ranked according to their effect on the customer— generally the external customer
1123-77375_ch01_5P.indd 61 12/21/18 11:02 AM
62 Chapter Eight
• Provides additional information to help plan thorough and efficient test programs for control
• Helps in the initial design for manufacturing and assembly requirements
• Provides an open issue format for recommending and tracking risk- reducing actions in both design and pro cess
• Provides future reference to aid in analyzing field concerns
Form and Ratings
The form for the DFMEA is the same as the form in Figure 6.1. How-ever, the ratings may differ from industry to industry and organ ization from organ ization; the ratings in Tables 8.1–8.3 are very common and can be used as a default guideline.
Special Note: There is nothing special about these guidelines. They may be changed to reflect the industry, the organ ization, the product/design, and/or pro cess. To modify these guidelines:
1. List the entire range of pos si ble consequences (effects).
2. Force rank the consequences from high to low.
3. Resolve the extreme values (rate 10 and rate 1).
4. Fill in the “other” ratings.
5. Use consensus.
Strategies for Lowering Risk: (Concept/Design)— High Severity or Occurrence
Change the Product Design to:
• Eliminate the failure mode cause or decouple the cause and effect.
• Eliminate or reduce the severity of the effect.
• Make cause less likely or impossible to occur.
• Eliminate function or eliminate part (functional analy sis).
“Tools” to Consider
• QFD
• FTA
1123-77375_ch01_5P.indd 62 12/21/18 11:02 AM
Table 8.1 DFMEA— severity.
Effect Description Rate
None No effect noticed by customer. The failure will not have any perceptible effect on the customer.
1
Very minor Very minor effect, noticed by discriminating customers. The failure will have little perceptible effect on the discriminating customers.
2
Minor Minor effect, noticed by average customers. The failure will have minor perceptible effect on average customers.
3
Very low Very low effect, noticed by most customers. The failure will have some small perceptible effect on most customers.
4
Low Primary product function is operational, however, at a reduced level of per for mance. Customer is somewhat dissatisfied.
5
Moderate Primary product function operational, however, secondary functions inoperable. Customer is moderately dissatisfied.
6
High Failure mode greatly affects product operation. Product or portion of the product is inoperable. Customer is very dissatisfied.
7
Very high Primary product function is nonoperational but safe. Customer is very dissatisfied.
8
Hazard with warning
Failure mode affects safe product operation and/or involves nonconformance with government regulation with warning.
9
Hazard with no warning
Failure mode affects safe product operation and/or involves nonconformance with government regulation without warning.
10
1123-77375_ch01_5P.indd 63 12/21/18 11:02 AM
64 Chapter Eight
• Benchmarking
• Brainstorming
• TRIZ (theory of inventive problem solving)
Evaluate Ideas Using Pugh Concept Se lection (Specific Examples)
• Change material, increase strength, decrease stress.
• Add redundancy.
• Constrain usage (exclude features).
• Develop fail- safe designs and early- warning system.
Strategies for Lowering Risk: (Concept/Design)— High Detection Rating
Change the Evaluation/Verification/Tests to:
• Make failure mode easier to perceive.
• Detect causes prior to failure.
Table 8.2 DFMEA— occurrence.
Occurrence Description Frequency Rate
Remote Failure is very unlikely <1 in 1,500,000 1
Low Relatively few failures 1 in 150,000 2
1 in 15,000 3
Moderate Occasional failures 1 in 2000 4
1 in 400 5
1 in 80 6
High Repeated failures 1 in 20 7
1 in 8 8
Very high Failure is almost inevitable 1 in 3 9
>1 in 2 10
1123-77375_ch01_5P.indd 64 12/21/18 11:02 AM
Table 8.3 DFMEA— detection.
Detection Description Rate
Almost certain Design control will almost certainly detect the potential cause of subsequent failure modes.
1
Very high Very high chance the design control will detect the potential cause of subsequent failure mode.
2
High High chance the design control will detect the potential cause of subsequent failure mode.
3
Moderate high Moderately high chance the design control will detect the potential cause of subsequent failure mode.
4
Moderate Moderate chance the design control will detect the potential cause of subsequent failure mode.
5
Low Low chance the design control will detect the potential cause of subsequent failure mode.
6
Very low Very low chance the design control will detect the potential cause of subsequent failure mode.
7
Remote Remote chance the design control will detect the potential cause of subsequent failure mode.
8
Very remote Very remote chance the design control will detect the potential cause of subsequent failure mode.
9
Very uncertain There is no design control or the control will not or cannot detect the potential cause of subsequent failure mode.
10
1123-77375_ch01_5P.indd 65 12/21/18 11:02 AM
66 Chapter Eight
“Tools” to Consider
• Benchmarking
• Brainstorming
• Pro cess control (automatic corrective devices)
• TRIZ
Evaluate Ideas Using Pugh Concept Se lection (Some Specific Examples)
• Change testing and evaluation procedures.
• Increase failure feedback or warning systems.
• Increase sampling in testing or instrumentation.
• Increase redundancy in testing.
Tools
• Reliability modeling
• QFD
• Benchmarking
• Block diagram
• P- diagram
• Interface diagram
• Cause- and- effect diagram
• Function tree
PROCESS
Purpose
The purpose of a PFMEA is to resolve issues/concerns/problems in the pro cess that result in low- quality product being shipped to the customer. Here the customer may be internal or external. There are two assumptions that must always be considered for optimum results:
1123-77375_ch01_5P.indd 66 12/21/18 11:02 AM
The Common Types of FMEAs 67
1. Assuming incoming parts are correct, what would cause the opera-tion to fail in this manner? In other words, the design is okay as is.
2. What incoming sources of variation could cause the operation to fail in this manner? As a last resort, evaluate issues that may be associated with design.
Use
Fundamentally, the PFMEA is used to identify each manufacturing step and to determine what functions are associated with each manufacturing pro cess step, and then to ask:
1. What does the pro cess step do to the part?
2. What are you doing to the part/assembly?
3. What is the goal, purpose, or objective of this pro cess step?
There is no standardized form for a PFMEA. You may use the one in Figure 6.1 or a simplified form, such as the one shown in Figure 8.1. Any organ ization may modify these forms to reflect their own pro cesses and needs.
Benefit
A PFMEA brings the following benefits:
• Identifies potential product- related pro cess failure modes
• Assesses the potential customer effects of the failures
• Identifies the potential manufacturing or assembly pro cess causes and identifies pro cess variables on which to focus controls or monitoring
• Develops a ranked list of potential failure modes, establishing a priority system for corrective action considerations
• Documents the results of the manufacturing or assembly pro cess
• Identifies pro cess deficiencies
• Identifies confirmed critical characteristics and/or significant characteristics
• Identifies operator safety concerns
• Feeds information on design changes required and manufacturing feasibility back to the designers
1123-77375_ch01_5P.indd 67 12/21/18 11:02 AM
Seve
rity
How
sev
ere
is th
eef
fect
to th
e cu
stom
er?
Seve
rity
Occ
urre
nce
Det
ectio
n
RPN
RPN
Item/function
What is theprocessstep?
In whatwaysdoes the key input go wrong?
What is theimpact on the key outputvariables(customerrequirements)or internalrequirements?
Whatcausesthe keyinputto gowrong?
Potentialfailuremode
Potentialeffect(s)of failure
Potentialcause/mechanismof failure
What are the existingcontrols andproceduresthat preventeither thecause or the failure mode?
Currentprocesscontrols(prevent/direct)
What are theactions forreducing theoccurrence ofthe cause, orimprovingdetection?
Recommendedaction(s)
__ System
__ Subsystem
__ Component
Model year(s)/Program(s): _____________
Team: _______________
Process responsibility: _____________
Key date: _______________
FMEA name/number: _____________
Prepared by: _______________
FMEA date (orig.): _________ (Rev.): _________
Clas
s
How
ofte
n do
esca
use
of F
M o
ccur
?O
ccur
renc
e
How
wel
l can
you
dete
ct c
ause
or F
M?
Det
ectio
n
Who isresponsiblefor therecommendedaction?
Responsibilityand targetcompletiondate
Whatare thecompletedactions to take with therecalculatedRPM? Besure toincludecompletionmonth/year
Actionstaken
Action results
Figure 8.1 A typical PFMEA
1123-77375_ch01_5P.indd 68
12/21/18 11:02 AM
The Common Types of FMEAs 69
Ratings
There are no standard or universal criteria for ranking any FMEA. How-ever, typical rating guidelines are shown in Tables 8.4–8.6.
Special Note: There is nothing special about these guidelines. They may be changed to reflect the industry, the organ ization, the product/
Table 8.4 PFMEA— severity.
Effect Description Rate
None No effect noticed by customer. The failure will not have any effect on the customer.
1
Very minor
Very minor disruption to production line. A very small portion of the product may have to be reworked. Defect noticed by discriminating customers.
2
Minor Minor disruption to production line. A small portion (<5%) of product may have to be reworked online. Pro cess up but minor annoyances.
3
Very low Very low disruption to production line. A moderate portion (<10%) of very low product may have to be reworked online. Pro cess up but minor annoyances.
4
Low Low disruption to production line. A moderate portion (<15%) of product may have to be reworked online. Pro cess up but some minor annoyances.
5
Moderate Moderate disruption to production line. A moderate portion (>20%) of product may have to be scrapped. Pro cess up but some incon ve niences.
6
High Major disruption to production line. A portion (>30%) of product may have to be scrapped. Pro cess may be stopped. Customer dissatisfied.
7
Very high Major disruption to production line. Close to 100% of product may have to be scrapped. Pro cess unreliable. Customer very dissatisfied.
8
Hazard with warning
May endanger operator or equipment. Severely affects safe pro cess operation and/or involves noncompliance with government regulations. Failure will occur with warning.
9
Hazard with no warning
May endanger operator or equipment. Severely affects safe pro cess operation and/or involves noncompliance with government regulations. Failure occurs without warning.
10
1123-77375_ch01_5P.indd 69 12/21/18 11:02 AM
70 Chapter Eight
design, and/or pro cess. Keep in mind that to modify these guidelines, you should:
1. List the entire range of pos si ble consequences (effects).
2. Force rank the consequences from high to low.
3. Resolve the extreme values (rate 10 and rate 1).
4. Fill in the “other” ratings.
5. Use consensus.
Manufacturing Pro cess Control Matrix
In any pro cess there are several dominant factors that should be evaluated for failures. Table 8.7 shows some of the factors involved and the appropri-ate data used in the evaluation and control of these failures.
Table 8.5 PFMEA— occurrence.
Occurrence Description Frequency Cpk Rate
Remote Failure is very unlikely, no failures associated to similar pro cesses.
<1 in 1,500,000 >1.67 1
Low Few failures. Isolated failures associated with like pro cesses.
1 in 150,000 1.50 2
1 in 15,000 1.33 3
1 in 2000 1.17 4
Moderate Occasional failures associated with similar pro cesses, but not in major proportions.
1 in 400 1.00 5
1 in 80 0.83 6
High Repeated failures. Similar pro cesses have often failed.
1 in 20 0.67 7
1 in 8 8
Very high Pro cess failure is almost inevitable.
1 in 3 0.51 9
>1 in 2 0.33 10
1123-77375_ch01_5P.indd 70 12/21/18 11:02 AM
The Common Types of FMEAs 71
Manufacturing Pro cess Control Examples
Statistical Pro cess Control
• X- bar/R control/charts (variable data)
• Individual X- moving range charts (variable data)
• p- , n- , u- , and c- charts (attribute data)
Nonstatistical Control
• Check sheets, checklists, setup procedures, operational definitions/instruction sheets
• Preventive maintenance (PM)
– Tool usage logs/change programs
– Mistake proofing/error proofing/poka- yoke
Table 8.6 PFMEA— detection.
Detection Description Rate
Almost certain
Pro cess control will almost certainly detect or prevent the potential cause of subsequent failure mode.
1
Very high Very high chance pro cess control will detect or prevent the potential cause of subsequent failure mode.
2
High High chance the pro cess control will detect or prevent the potential cause of subsequent failure mode.
3
Moderate high
Moderately high chance the pro cess control will detect or prevent the potential cause of subsequent failure mode.
4
Moderate Moderate chance the pro cess control will detect or prevent the potential cause of subsequent failure mode.
5
Low Low chance the pro cess control will detect or prevent the potential cause of subsequent failure mode.
6
Very low Very low chance the pro cess control will detect or prevent the potential cause of subsequent failure mode.
7
Remote Remote chance the pro cess control will detect or prevent the potential cause of subsequent failure mode.
8
Very remote
Very remote chance the pro cess control will detect or prevent the potential cause of subsequent failure mode.
9
Very uncertain
There is no pro cess control or the control will not or cannot detect the potential cause of subsequent failure mode.
10
1123-77375_ch01_5P.indd 71 12/21/18 11:02 AM
72 Chapter Eight
• Training and experience
• Automated inspection
• Visual inspection
It is very impor tant to recognize that inspection is not a very effective control because it is a reactive task and quite often very subjective, especially with attribute data.
Table 8.7 A typical control matrix for a manufacturing pro cess.
Dominance Factor Attribute Data Variable Date
Setup Check sheet Checklist
X- bar/R chart X- MR chart
Machine p- or c- chart Check sheet
Run chart X- bar/R chart X- MR chart
Operator Check sheet Run chart
X- bar/R chart X- MR chart
Component/material Check sheet Supplier information
Check sheet Supplier information
Tool Tool logs Check sheet p- or c- chart
Tool logs Capability study X- MR chart
Preventive maintenance Time to failure chart Supplier information
Time to failure chart Supplier information X- MR chart
Fixture/pallet/work holding Time to failure chart Check sheet p- or c- chart
Time to failure chart X- bar/R chart X- MR chart
Environment Check sheet Run chart X- MR Chart
1123-77375_ch01_5P.indd 72 12/21/18 11:02 AM
The Common Types of FMEAs 73
Strategies for Lowering Risk: (Manufacturing)— High Severity or Occurrence
Change the Product or Pro cess Design to:
• Eliminate the failure cause or decouple the cause and effect.
• Eliminate or reduce the severity of the effect (recommend changes in design).
“Tools” to Consider
• Benchmarking
• Brainstorming
• Mistake proofing
• TRIZ
Evaluate Ideas Using Pugh Concept Se lection (Specific Examples):
• Developing a “robust design” (insensitive to manufacturing variations)
• Changing pro cess par ameters (time, temperature, etc.)
• Increasing redundancy and adding pro cess steps
• Altering pro cess inputs (materials, components, consumables)
• Using mistake proofing (poka- yoke) to reduce handling
Strategies for Lowering Risk: (Manufacturing)— High Detection Rating
Change the Pro cess Controls to:
• Make failure mode easier to perceive.
• Detect causes prior to failure mode.
“Tools” to Consider
• Benchmarking
• Brainstorming
1123-77375_ch01_5P.indd 73 12/21/18 11:02 AM
74 Chapter Eight
Evaluate Ideas Using Pugh Concept Se lection (Specific Examples)
• Change testing and inspection procedures/equipment.
• Improve failure feedback or warning systems.
• Add sensors/feedback or feed forward systems.
• Increase sampling and/or redundancy in testing.
• Alter decision rules for better capture of causes and failures (i.e., more sophisticated tests).
At this stage, now you are ready to enter a brief description of the recommended actions, including the department and individual respon-sible for implementation, as well as both the target and finish dates on the FMEA form. If the risk is low and no action is required write “No action needed.”
For each entry that has a designated characteristic in the “class” (iden-tification) column:
• Review the issues that impact cause/occurrence, detection/control, or failure mode.
• Generate recommended actions to reduce risk.
• Examine special RPN patterns. They suggest that certain “char-acteristics”/root causes are impor tant risk factors that need special attention.
Guidelines for Pro cess Control System
1. Select the pro cess.
2. Conduct the FMEA on the pro cess.
3. Conduct gauge system analy sis.
4. Conduct a pro cess potential study.
5. Develop a control plan.
6. Train operators in control methods.
7. Implement the control plan.
8. Determine long- term pro cess capability.
9. Review the system for continual improvement.
1123-77375_ch01_5P.indd 74 12/21/18 11:02 AM
The Common Types of FMEAs 75
10. Develop an audit system.
11. Institute improvement actions.
Tools
• Mistake proofing
• Inspection
• Cause- and- effect diagram
• Affinity diagram
• Engineering testing (specific to the cause being evaluated)
EQUIPMENT
An equipment FMEA (EFMEA) is a systematic approach that applies the generic form of an FMEA to aid the thought pro cess used by engineers to identify the equipment’s potential failure modes and their effects. The focus however is on the operator’s safety.
Purpose
The basic purposes of any EFMEA are to:
1. Identify potential failure modes and rate the severity of their effects.
2. Rank- order potential design and pro cess deficiencies.
3. Help engineers focus on eliminating equipment design and pro cess concerns and help prevent prob lems from occurring.
Use
Fundamentally there are three reasons for using EFMEA. They are to:
1. Identify potential failure modes that may adversely affect environ-ment safety.
2. Identify potential failure modes that may adversely affect operator safety.
3. Identify potential design deficiencies before releasing machinery to production.
1123-77375_ch01_5P.indd 75 12/21/18 11:02 AM
76 Chapter Eight
Benefit
There are at least five basic benefits in completing an EFMEA. They are:
1. Improving the quality, reliability, and safety of the customer’s equipment
2. Improving the com pany’s image and competitiveness
3. Helping to increase customer satisfaction
4. Reducing equipment development, timing, and cost
5. Documenting and tracking actions taken to reduce risk
General Information about EFMEA
The EFMEA is a special FMEA; as such, some items have to be addressed specifically for equipment:
How is an EFMEA prepared? The equipment supplier is responsible for preparing the initial EFMEA. The customer generally only assists as a team member.
When is an EFMEA started? Generally, there are four points of concern for when the EFMEA should be started. They are:
1. When new systems, subsystems components, equipment, and pro cesses are being designed.
2. When existing equipment or pro cesses are modified in any way.
3. When carryover equipment and/or pro cesses are used in new applications or new environments.
4. After completing a problem- solving methodology (e.g., 8D) to prevent recurrence of a prob lem.
Who prepares the EFMEA? The EFMEA pro cess is a team effort between the supplier and customer. The team should be cross- functional and multidisciplinary. The responsible equipment engineer is the leader of any EFMEA team. However, the sup-plier’s equipment design engineer is expected to involve repre-sentatives from all affected activities. It is suggested that at a minimum the following repre sen ta tion should be part of an active team:
1123-77375_ch01_5P.indd 76 12/21/18 11:02 AM
The Common Types of FMEAs 77
Purchasing Supervisors TestingSkilled trades Quality engineering OperatorsIndustrial Reliability Customer engineering engineering representative
It is imperative that the team must realize that the team members may change as the equipment matures through the design, build, and test phases. Also, team members may be added as ad hoc personnel to aid in specific issues but not as core team members.
Who updates the EFMEA? The supplier’s design engineer is responsi-ble for keeping the EFMEA up to date. It is also the responsibility of suppliers to keep their own copy of the EFMEA.
When is an EFMEA updated? There are at least three conditions for updating the EFMEA:
1. Whenever a new machine’s proj ect time line changes
2. Whenever design modifications or new failure modes are discovered
3. Whenever a change is being considered to a machine’s design, application, environment, material usage, or operational pro cess
When is an EFMEA completed? It is considered complete when the equipment is installed, has passed its reliability testing, and is signed off by the plant staff. However, remember that an EFMEA, just like the traditional FMEA, is a living document and must be updated whenever significant changes occur in the equipment’s design and/or application.
When can an EFMEA be discarded? Depending on the industry it varies from cradle to grave (nuclear industry) to specific years as defined in the rec ord retention requirement of the organ ization’s policies and procedures. The retention period is reviewed regularly for effectiveness and appropriateness and is part of the organ-izations quality system.
What is the form that may be used in an EFMEA? A typical form for the EFMEA is the shown in Figure 8.2. Obviously it may be modified to reflect specific issues with a par tic u lar industry and/or equipment.
1123-77375_ch01_5P.indd 77 12/21/18 11:02 AM
78 Chapter Eight
Form
Figure 8.2 is coded with numbers from 1 to 23. The explanations follow the form.
FMEA number (1): Each FMEA must have a number to help track the document and its information. After all, the FMEA is a controlled document.
Equipment Name (2): It is used to identify the equipment’s name for reference and accountability.
Design responsibility (3): This is the place where the design respon-sibility is identified. For example, it may be the supplier and/or specific department or group responsible for the design of the par tic u lar equipment.
Prepared by (4): This item must identify the name, phone number, and e- mail for the primary contact for the EFMEA.
Model (5): This identifies the model of the equipment, if applicable.Review Date (6): It is the initial date that the EFMEA started. This
date should fall within the design and development phase of the equip-ment’s life- cycle pro cess. Sometimes this is called original date.
Page_ of _ : Identifies all pages of the FMEA for reference purpose.FMEA Date (7): This is the initial date the EFMEA is completed or
the revision date.Core Team (8): This item covers all the team members that participate
in the EFMEA. It should identify them by name, department, telephone, e- mail, and address.
System, Subsystem, Component (9a): This is information used to clas-sify the analyzed machine’s subsystem. The intent here is to identify the hierarchy of the machine so that it is quickly formulated and then trans-ferred to the column listing all the subsystems in the appropriate order. (Special Note: The subsystem name column and the function and per for-mance requirements column have the same location. However, to make the distinction easier it is suggested that the two be separated.)
Function or Per for mance Requirements (9b): This column relates directly to the subsystem name information and lists all the subsystem’s associated functions and the design intent of that system. This column provides information that corresponds to each of the machine’s identified subsystems. In addition, using these four recommended steps to fill out this column simplifies subsystem function identification: (1) brainstorm, (2) evaluate the results of brainstorming, (3) convert to verb- noun format, and (4) establish the appropriate and applicable mea sure ment system.
Potential Failure Mode (10): This is defined as the manner in which the equipment could potentially fail to meet the design intent. These fail-ures are generally the ones that the operator sees. Typically, there are two approaches for identifying these failures: (1) functional, which relates to
1123-77375_ch01_5P.indd 78 12/21/18 11:02 AM
Seve
rity
(12)
Seve
rity
Occ
urre
nce
Det
ectio
n
Rev
ised
RPN
(23)
RPN
(19)
System,subsystem,component(9a)
Functionorperformancerequirements(9b)
Potentialfailuremode(10)
Potentialcauses of failure(14)
Currentdesign andequipmentcontrols(17)
Recommendedaction(s) (20)
FMEA number (1)
Equipment name (2)
Design responsibility (3)
Prepared by (4)
Model (5)
Review date (6)
Page ___ of ___
FMEA date (7)
Core team (8)
Clas
sific
atio
n (1
3)
Occ
urre
nce
(15)
Det
ectio
n (1
8)Preventioncontrols andequipmentcontrols(16)
Responsibilityand targetcompletiondate (21)
Actiontaken(22)
Action results
Figure 8.2 Explanation of the EFMEA form
1123-77375_ch01_5P.indd 79
12/21/18 11:02 AM
80 Chapter Eight
some form of loss of function; and (2) hardware, when detailed part designs are available. If the failures are identified as part of a static model that means that the identified failures have no effect on other failures of other subsystems or components under investigation. It is very impor tant for the team to identify as many failures as pos si ble in the design phase because this action will reduce the number of failures that may be seen during the debug phase, start-up, and the useful life of the equipment.
Potential Effects of Failure (11): These represent the potential conse-quences of the failure mode. Typical areas of concern may be breakdowns; reduced cycle time; tooling; setup and adjustments; defective parts; govern-ment regulations; idling and minor stoppages; and safety. Of these, special attention must be given to government regulations and safety issues.
Severity (12): Represents the seriousness of the effects listed in column 11 and is made up of three components: (1) safety, (2) equipment downtime, and (3) product scrap. Each effect (in column 11) is assigned a ranking between 1 and 10 from the agreed-on criteria and the highest ranking is entered in this column. Only the highest ranking is entered because it rep-resents the most serious effect that may occur, if the failure mode occurs.
Classification (13): In the EFMEA the only time that this column is used is if the severity is 9 or 10, which has to do with government regula-tions or operator safety. If indeed it is used, then action must be taken to correct the prob lem. Typical designations are OS (for operator safety) and Y (which stands for government regulations).
Potential Causes of Failure (14): This column represents design defi-ciency or pro cess variation that results in the failure mode. To have an effective EFMEA, all first level failure mode causes must be identified and listed in this column. This may be accomplished by considering the follow-ing three questions: (1) What would cause the subsystem/component to fail in this manner? (2) What circumstances could cause the subsystem/com-ponent to fail to perform its function? (3) What can cause the subsystem/component to fail to deliver its intended function? To help in this identifica-tion pro cess the team should consider the causes relating to the highest risk failure modes and then review the following for validating their selected options: (1) historical test reports, (2) warranty data reports, (3) surrogate EFMEAs, (4) concern reports, (5) field reports, and (6) product recalls. The team must remember that all 9 or 10 severity rankings must identify the root cause of the failure mode. A root cause is the under lying reason for a first level cause to occur. To identify the root cause(s) one may use simple or difficult methodologies/tools. Options include a problem- solving methodol-ogy such as 8D or DOE, cause- and- effect diagrams, and FTA.
Occurrence (15): This column contains the rating relating to the likeli-hood that a par tic u lar failure mode cause will occur within a specific time
1123-77375_ch01_5P.indd 80 12/21/18 11:02 AM
The Common Types of FMEAs 81
period. To identify a reasonable occurrence, the team may use the Poisson distribution for theoretical numbers; historical data; maintenance rec ords; surrogate data; and warranty data. Each cause must have its own occur-rence number.
Prevention Controls and Equipment Controls (16): This column docu-ments all the prevention controls that are planned to minimize the risk of the failure mode.
Current Design and Equipment Controls (17): This column documents the effectiveness of the planning controls in column 16. For detection pur-poses the best control item will be carried over to column 18. For example, if there are four controls with the following individual ratings such as 8, 5, 6, 7, and a 2, then the 2 will be carried over to column 18. The reason is that the control with a 2 rating is the strongest of them all and therefore the most effective.
Detection (18): It is the column indicating the likelihood that the pre-vention controls (item 17) can detect and prevent the failure from reaching the customer. This evaluation is based on preset criteria that every one has agreed with. A numerical value of 1 indicates the prob lem is caught at the source, whereas a 10 indicates the failure reaches the customer.
RPN (19): The RPN is the product of severity, occurrence, and detection. It is very impor tant here to remember that each root cause must have its own RPN. The RPN often is used as a value that determines the priority for either design improvements and/or operational changes.
Recommended Action (20): This column documents all the pos si ble (appropriate) alternatives that may reduce the risk of the failure mode’s RPN. There should be more than one action available. The focus should be on reducing safety concerns, then on modes with high severity and occurrence and fi nally on the combination of severity, occurrence, and detection (RPN).
Responsible Individual and Target Completion Date (21): This column documents who is assigned the responsibility to review or implement the recommended action—if appropriate. A target date must also be identified. Without a name and/or date, no one is responsible and the due date becomes infinity.
Action Taken (22): This column briefly describes the specific action taken from the alternatives identified in column 20 with the intent to lower the risk of a failure mode. A completed EFMEA is of very limited value without positive actions to eliminate potential injury, safety issues, gov-ernment regulation violations, and machine downtime or to prevent part defects. The action taken here is essential to implement high failure risk solutions. It must be also mentioned that the primary design responsibility belongs to the supplier and therefore all EFMEA updating remains the supplier’s responsibility, even after installation at the customer’s facilities.
1123-77375_ch01_5P.indd 81 12/21/18 11:02 AM
82 Chapter Eight
(Obviously, if the design is the responsibility of the customer, then the cus-tomer bears the responsibility of the design EFMEA.)
Revised RPN (23): The revised RPN is calculated from the new severity, occurrence, and detection values resulting from the implemented actions in item 22. These new numerical values are a result of the team consensus. If there were no actions taken, then these new columns are left blank. If the actions have indeed reduced any of the three numbers or all of them, then the EFMEA team must repeat steps 20 through 23 on a new EFMEA form and a new revision number must be assigned. A word of caution here is appropriate. In order to change the severity number a design change must occur, other wise the occurrence and/or the detection may be the only items that will change.
Ratings
Generally, the ratings for the EFMEA are the same as that of the design FMEA.
Tools
As machine builders design new equipment a variety of techniques and tools exist to improve quality, safety, and per for mance. However, from experi-ence, an effective EFMEA can produce very good results if the EFMEA and FTA are used in combination.
The reader should remember that the EFMEA identifies all the machine’s potential failure modes and their first- level causes. In other words, it tries to identify the breadth of prob lems with a new design (or a modified one). On the other hand, the FTA is used to analyze the root cause of significant failures and establish the probabilities of each cause. The importance of this approach lies with the fact that in the pro cess of evaluating the probabilities it shows graphically the relationship of each of the causes. In other words, the FTA focuses on the depth of each individual failure. The fundamental question in any FTA is: What are all pos si ble causes for one failure?
Other typical tools that may be used are:
• Block diagrams
• Interface matrix
• P- diagram
• Brainstorming
1123-77375_ch01_5P.indd 82 12/21/18 11:02 AM
The Common Types of FMEAs 83
• Cause- and- effect diagram
• Reliability formulas (for defining failure)
• Poisson distribution (for identifying specific failure rates)
• DOE
• Others (see Chapter 13 on Tools)
1123-77375_ch01_5P.indd 83 12/21/18 11:02 AM
1123-77375_ch01_5P.indd 84 12/21/18 11:02 AM
85
Just like any other industry, the health industry is very much interested in reducing failures and become “fault tolerant” as much as pos si ble. To this end the Joint Commission on Accreditation of Healthcare Organ izations (JCAHO), now named as just The Joint Commission (TJC), has identified several proactive risk standards for conducting risk assessments. One of them is the FMEA. In healthcare it is designated as HFMEA. The H is for health.
For healthcare organ izations the basic definition of FMEA is still appropriate because it is recognized as a systematic method of identifying and preventing product and pro cess prob lems before they occur. This defi-nition of course is a drastic mind-set change for most health organ izations. It is drastic because in the past the practice was to reactively change as errors and/or failures were encountered and not to take proactive action to either prevent or absorb, in a systematic fashion, medical or healthcare errors or even risks to patient safety.
Prevention in healthcare means that the organ ization must have policies and procedures that prevent adverse occurrences, rather than simply react-ing when they occur. In addition, it means that barriers created by hindsight bias, fear of disclosure, embarrassment, blame, or retaliation (punishment in any form) must be identified and corrective action must be in place.
Some major issues could be minimized or even avoided had an FMEA been used. Examples include:
• Medical gas usage
• Patient safety (bed rail and enclosed bed entrapment)
• MRI incidents (ferromagnetic objects)
• Major medical center power failure
9Health FMEA
1123-77375_ch01_5P.indd 85 12/21/18 11:02 AM
86 Chapter Nine
In healthcare a barrier that eliminates or substantially reduces the like-lihood of a hazardous event occurring is considered an effective control mea sure. Therefore, in order to have a good system for preventing issues, prob lems concerns, and hazardous situations, there are six steps that a health organ ization must follow to be successful:
1. Identify high- risk pro cesses.
2. Prioritize these pro cesses.
3. Identify potential failure modes.
4. Identify the effect(s) for each failure mode.
5. Conduct a root- cause analy sis (RCA) for each critical effect.
6. At least once a year select one high- risk area for evaluation.
Typical actions from the above six items are:
• Redesign the pro cess to minimize or eliminate the risk of the fail-ure mode or to protect patients from its effects.
• Test the redesign pro cess.
• Validate that the pro cess is working.
• Implement the change.
• Identify and implement and mea sure effectiveness. (Remember that effectiveness is an issue of customer satisfaction and efficiency is an issue of optimizing resources.)
• Implement a control and monitor strategy for maintaining the effec-tiveness of the “new” pro cess over time.
COMPARISON OF ROOT- CAUSE ANALY SIS AND HFMEA
The purpose of RCA is to get to the root cause and escape point. The pur-pose of the HFMEA is to simulate potential failure modes and remove them or minimize them as much as pos si ble from the pro cess. Table 9.1 shows some similarities and differences.
Yet another significant difference is the usage of hazard analy sis as part of the FMEA. In healthcare it is common to use the term “hazard analy sis” to identify the pro cess of collecting and evaluating information on
1123-77375_ch01_5P.indd 86 12/21/18 11:02 AM
Health FMEA 87
hazards associated with par tic u lar pro cesses. The idea here is to develop a list of what are significant and reasonable items likely to cause injury and/or illness if not effectively controlled.
By contrast, the FMEA proper is evaluating dif fer ent ways that a pro-cess or subpro cesses can fail to provide the anticipated ideal function (i.e., the anticipated result without any errors). A good source for finding areas of improvement, especially in healthcare, is to evaluate the classic 8 wastes as well as to study the 6S. A simple overview is shown in Table 9.2.
THE PROCESS OF THE HFMEA
Fundamentally there are six steps in conducting a HFMEA. They are:
1. Define the topic.
2. Assem ble the team.
3. Graphically describe the pro cess.
4. Conduct the analy sis.
5. Identify actions and outcome mea sures.
6. Repeat as needed.
Table 9.1 Similarities and differences of RCA and HFMEA.
Similarities Differences
• Interdisciplinary and cross- functional team
• Focus on systems issues
• Use of flow diagram
• Actions and outcomes expected
• Matrix scoring card expected
• Use of triage to develop or trigger questions, cause- and- effect diagram, brainstorming, and so on
• Focus on pro cess vs. chronological flow diagram
• Prospective analy sis (e.g., what if)
• Choose specific topic for evaluation
• Include and consider detectability and criticality in evaluation
• Focus on testing intervention
1123-77375_ch01_5P.indd 87 12/21/18 11:02 AM
88 Chapter Nine
Define the Topic
Perhaps the most impor tant step in conducting an HFMEA is to define the scope and have a good and clear definition of the pro cess to be evaluated. It is highly recommended that the evaluation at this stage be defined as a mea sur able output.
The items of interest may be orga nizational and/or process- oriented. For pro cess orientation issues it is easy to identify the gaps(s) for analy sis.
Table 9.2 8 wastes and 6S.
Item 8 Wastes 6S
1 Unused human potential: Untapped creativity/talent/injuries. How does this affect the healthcare organ ization?
Sort: Remove items not needed daily (red tag pro cess).
2 Waiting: Patients/provider/material. What is the effect on the healthcare organ ization?
Set in order: Label items and make it obvious where they belong.
3 Inventory: Stacks of work/piles of supplies. How does this affect the efficiency of the health organ ization?
Shine/sweep: Clean and inspect every thing, inside and out. Visually sweep area to make sure every thing is in its place.
4 Transportation: Transporting people and/or paperwork. Does this hinder productivity? If so, how it can be improved and what are the bottlenecks?
Safety: All required safety information is posted. All exits and emergency equipment are clearly marked and functional.
5 Defects: Wrong information/rework. What is causing the wrong information and rework? What can we do about it?
Standardize: Establish policies and standard work to ensure 6S.
6 Motion: Finding information/double entry/searching: What is causing the extra work in searching and double entry? Why can’t we find the information when it is needed?
Sustain: Provide and/or make sure that training, discipline, daily activities, and self- audits are part of the new system so that improvements are continued and/or improved.
7 Overproduction: Duplication/extra information. Why does the system allow the organ ization to have duplicate and/or extra information beyond the legal requirements?
8 Pro cessing: Extra steps/checks/workarounds. Where is the inefficiency that creates this extra pro cessing?
1123-77375_ch01_5P.indd 88 12/21/18 11:02 AM
Health FMEA 89
The reason is that you know where you are and you also know where you want to be. That difference is the gap, and it can be mea sured in time, money, material utilization, and many more ways. For orga nizational change that may be little more difficult, but it can be accomplished if there is a willingness from management to change. A typical comparison is shown in Table 9.3.
Assem ble the Team
The team by definition must be cross- functional and multidisciplinary. All FMEAs and HFMEAs must be completed by a team. Team members must have owner ship and knowledge about the pro cess. All decisions must be based on a consensus basis.
Graphically Describe the Pro cess
To understand the pro cess with all its activities the team must be able to use some type of a graphical form, such as a flowchart, emphasizing the pro cess sequence rather than the chronological one. If the pro cess is com-plex, break down the area of concern and focus on “doable,” manageable
Table 9.3 A typical comparison of pro cess design and orga nizational change.
Pro cess Redesign Orga nizational Change
• Fail- safe designs
• Simplicity of a pro cess
• Standardization
• Elimination of excess sound/noises
• Simulate
• Looser coupling of systems (interaction and interfacing)
• Forcing functions where they do not belong
• Usability testing
• Redundancy
• Reduction of reliance on memory
• Reduction of complexity
• Leadership commitment
• Drive out fear
• Teamwork
• Empowerment
• Free flow of information
• Feedback (horizontal and vertical)
• Encouragement of ideas to improve
1123-77375_ch01_5P.indd 89 12/21/18 11:02 AM
90 Chapter Nine
activities. All activities of a simple and/or complex pro cess must be num-bered and related to the HFMEA as well as to the control/reaction plan. In the case of a very complex area, break down each subarea, do a pro cess flow diagram for each subarea, and evaluate accordingly.
Conduct the Analy sis
A systematic approach is necessary here. A typical one is the following:
• List failure mode by following the function and the pro cess flow diagram.
• Determine both severity and probability.
• Decide on a course of action based on some formal method (e.g., decision tree, FTA).
• Determine all failure mode causes.
• Determine the frequency of these causes.
• Identify actions and outcome mea sures.
To accomplish these tasks as part of the analy sis, the team needs appro-priate and applicable forms, worksheets, scoring criteria, and a formal RCA methodology.
Identify Actions and Outcome Mea sures
The purpose of the previous step is to basically decide whether to “elimi-nate,” “reduce,” or “accept” the failure mode cause. As such, the analy-sis will drive the team to a par tic u lar action and/or reaction depending on the expected outcome. It is impor tant here to realize that each fail-ure mode cause needs its own action and/or reaction. The action and/or reaction must be mea sur able so that the redesign or improvement may be mea sured appropriately and without individual bias. When the root cause is identified, then the effort must be made to ensure that the escape point is also identified and the specific action is assigned to a specific individual with a specific due date. Fi nally, in this stage management must agree with the root cause and appropriate action taken. (Note: An escape point is where the failure occurred and could have been caught but was not.)
1123-77375_ch01_5P.indd 90 12/21/18 11:02 AM
Health FMEA 91
Repeat as Needed
In the spirit of continual improvement, the cycle must be repeated until the failure mode is completely eliminated or another failure is identified with a higher risk to purse.
FORMS AND RANKINGS
Worksheet
A worksheet is a temporary form that is used to facilitate the discussion of a HFMEA. There is no standard for such a form. Figure 9.1 shows a typical worksheet.
Ranking
A typical severity ranking for a HFMEA is shown in Table 9.4.
Typical Probability Ranking for HFMEA
There are many ways to rank probability ratings. However, the most com-mon one for HFMEA is the one based on Mil- Std 1629, which is combined with the severity ranking to give a numerical value for setting a priority. The items are:
Frequent: Likely to occur immediately or within a short period (may happen several times in one year).
Occasional: Probability will occur (may happen several times in one to two years).
Uncommon: Pos si ble to occur (may happen sometime in two to five years).
Remote: Unlikely to occur (may happen sometime in five to 30 years).
These items are multiplied with the following severity levels:
Category I— Catastrophic: A failure that may cause death or a system loss (FMEA rating 9–10).
Category II— Critical: A failure that may cause severe injury, major property damage, or major system damage (rating 7–8).
1123-77375_ch01_5P.indd 91 12/21/18 11:02 AM
Failuremode
Flow
Potentialcauses
Escapepoint(weakness) Proceed?
Action type(eliminate,control,accept)
Actions orrationale forstopping
Actions and outcomesHFMEA analysis
HFMEA Worksheet
Prob
lem
Seve
rity
Ran
k
Det
ectio
n
Outcomemeasures
Personresponsible
Managementconcurrence
Figure 9.1 A typical HFMEA worksheet
1123-77375_ch01_5P.indd 92
12/21/18 11:02 AM
Table 9.4 A typical severity ranking.
Severity Rating
Catastrophic event Major event Moderate event Minor event
(traditional FMEA rating of 9 or 10 indicates government regulation or safety issues— death or injury— with or without warning)
(traditional FMEA rating of 7 or 8 means that the failure causes a high degree of customer dissatisfaction)
(traditional FMEA rating of 3 to 6 means that the failure may be overcome with modifications to the pro cess or product or ser vice, but there is a minor per for mance loss)
(traditional FMEA rating of 1 or 2 means failure would not be noticeable to the customer and would not affect delivery of the service or product)
Patient outcome: Death or major permanent loss of function (sensory, motor, physiological, or intellectual), suicide, rape, hemolytic transfusion reaction, surgery/procedure on the wrong patient or wrong body part, infant abduction or infant discharge to the wrong family
Visitor outcome: Death or hospitalization of 3 or more visitors
Staff outcome: Death or hospitalization of 3 or more staff members
Fire: Any fire that is more than an incident
Equipment or facility: Damage over $250,000
Patient outcome: Permanent lessening of bodily function (sensory, motor, physiological, or intellectual), disfigurement, surgical intervention required, increased length of stay for 3 or more patients, increased level of care for 3 or more patients
Visitor outcome: Hospitalization of 1 or 2 visitors
Staff outcome: Hospitalization of 1 to 3 (or more) staff members, experiencing lost time or restricted duty due to injuries or illness
Fire: Not applicable
Equipment or facility: Damage over $100,000
Patient outcome: Increased length of stay or increased level of care for 1 or 2 patients
Visitor outcome: Evaluation and treatment for 1 or 2 visitors (less than hospitalization)
Staff outcome: Medical expenses, lost time, or restricted duty from injuries or illness for 1 or 2 staff members
Fire: Very small or insignificant
Equipment or fa cil i ty: Damage between $10,000 and $100,000
Patient outcome: No injury, no increased length of stay or increased level of care
Visitor outcome: Evaluation and no required or refused treatment
Staff outcome: First aid treatment only with no lost time; no restricted duty injuries or illness
Fire: Not applicable
Equipment or facility: Damage less than $10,000 or loss of any utility without adverse patient outcome (e.g., power, natu ral gas, electricity, water, communications, transport, heat/air conditioning)
1123-77375_ch01_5P.indd 93 12/21/18 11:02 AM
94 Chapter Nine
Category III— Marginal: A failure that may cause minor injury, minor property damage, or a minor system damage, resulting in delay or loss of availability of the system or even degradation of the system (rating 3–6).
Category IV— Minor: A failure not serious enough to cause injury, property damage, or system damage. However, it will cause some unscheduled delays (rating 1–2).
As a result of this multiplication, Table 9.5 is generated and appropriate action is taken based on the numerical value of Severity × Probability.
Detection
In a typical HFMEA the detection is based on the effectiveness ability to control the failure. So, a rank is generally the same as in the traditional FMEA.
Tools
There are many tools one may use in any FMEA and/or HFMEA. However, the most common ones are:
• Pro cess flowchart
• Brainstorming
• Check sheets
• Affinity chart
• Statistical pro cess control charts
Table 9.5 A typical matrix showing severity and probability.
Severity
Catastrophic Major Moderate Minor
Pro
bab
ility
Frequent
Occasional
Uncommon
Remote
1123-77375_ch01_5P.indd 94 12/21/18 11:02 AM
Health FMEA 95
• Correlation analy sis
• Scatter plots
• Box plots
• Decision tree
For more advanced tools and methodologies, see Chapter 13.
1123-77375_ch01_5P.indd 95 12/21/18 11:02 AM
1123-77375_ch01_5P.indd 96 12/21/18 11:02 AM
97
DEFINITION
FMEA is a bottom-up, inductive analytical method that may be performed at either the functional or piece- part level. FMECA extends FMEA by includ-ing a criticality analy sis, which is used to chart the probability of failure modes against the severity of their consequences. The result highlights failure modes with relatively high probability and severity of consequences, allow-ing remedial effort to be directed where it will produce the greatest value.
Unique Terms and Definitions
Function: The intent of the design or pro cess.
Function Failure: How this function will fail.
Failure Mode: What specific failure is identified for this function.
Failure Effect: What is the consequence(s) of this failure mode?
Severity Classification: How serious is this failure mode?
Occurrence: Frequency of the cause of the failure mode.
Mean Time between Failure: What is the average time between failures?
Failure Detection: How effective is the detection mechanism to “catch” the cause of the failure?
In addition to these primary definitions, the following considerations must be made in order to identify the function of concern:
• Consider all functions of the concern.
• Describe the functions with a verb and a noun in very specific terms.
10Failure Mode Effects and
Criticality Analy sis (FMECA)
1123-77375_ch01_5P.indd 97 12/21/18 11:02 AM
98 Chapter Ten
• State the function in terms of its utility rather than capability.
• Consider separating functions if they are complicated and convoluted.
For secondary considerations of the function, the following should be considered:
• Control
• Warning and status of failure mode
• Environmental issues
• Safety (both operational and personnel)
• Fluid and gas containment
• Explosions
• Support (structural)
• Comfort and blending of the environment
POS SI BLE SOURCES FOR IDENTIFYING FUNCTIONS
Functions for potential failure modes are all over an organ ization. However, for an FMECA, the common function sources may be found in the follow-ing areas:
• Maintenance (both from experience and theory)
• Operating manual, procedures, and instructions
• Data from system descriptions (with past failures or questionable practices)
These three areas are very fundamental and should be explored with an FMECA so that errors may be avoided. Typical errors as a result of over-looking the above items are:
• Missing either or both primary and secondary functions
• Listing simple and insignificant functions of lower priority
• Confusion of potential failures
• Lack of specificity
1123-77375_ch01_5P.indd 98 12/21/18 11:02 AM
Failure Mode Effects and Criticality Analy sis (FMECA) 99
• Failure to apply common sense in the name of expediency
• Failure to discuss thoroughly all functions
• Missing the opportunity to include compensating and/or restorative actions
• Missing the opportunity to miss effects at the point of functional failure
THE PROCESS OF CONDUCTING AN FMECA
Every methodology used to solve a prob lem of any kind is a pro cess and therefore must have the appropriate planning. This planning is done through the following steps:
Planning and Preparation
• Identify task.
• Identify team and responsibilities.
• Establish ground rules and assumptions.
• Identify analy sis items (functions).
• Prioritize items.
• Identify and document review pro cess.
• Conduct orientation and training.
Analy sis
• Equipment kick- off meeting
• Initial gathering
• Block diagram (hardware partition)
• Function
• Function failure
• Failure mode
• Failure effects
1123-77375_ch01_5P.indd 99 12/21/18 11:02 AM
100 Chapter Ten
• Failure consequences
• Task evaluation
• Task se lection
• Identifying responsible person for action
• Identifying due date of resolution
Implementation of Results
• Packaging maintenance task
• Implementing other pertinent actions
Sustaining
• Emergency issues
• Hardware changes
• Document reviews and update
• Trend and degradation analy sis
• Time effects as necessary
Criticality analy sis is a very special way of dealing with a failure. The basic difference between the FMEA and the FMECA is that in FMECA the priority of taking action is based on the product of effect (severity) and occurrence (frequency of the root cause). In some cases, the FMECA will include an analy sis of a “significant function” or SF. This means that the FMECA will include items that may have an adverse effect on the end- task with re spect to either individual or, in an interaction effect, other items. Some of these items include but are not limited to:
• Safety
• Financial functions
• Operations
• Environmental health
There are four basic questions that may identify an SF if the answer is a “Yes.” They are:
1. Does the loss of function have an adverse effect on environment and/or safety?
1123-77375_ch01_5P.indd 100 12/21/18 11:02 AM
Failure Mode Effects and Criticality Analy sis (FMECA) 101
2. Does the loss of function have an adverse effect on operations?
3. Does the loss of function have an adverse financial impact?
4. Does the function have protection by an existing control to prevent a failure?
There are two approaches to this action:
1. Qualitative analy sis
2. Quantitative analy sis
The qualitative analy sis is primarily used when specific item failures are not available. On the other hand, a quantitative analy sis is used when a suf-ficient failure rate is available to calculate criticality numbers.
QUALITATIVE ANALY SIS
Since there are no failure rate data available, failure mode probabilities are not used. This means that the criticality or risk associated with each failure is sub-jectively classified by the team members. The ranking subjectivity is for both severity and occurrence of the failures. The probability of occurrence of each failure is grouped into discrete levels that establish the qualitative failure probability level for each entry based on the judgment of the team. In addition, this analy sis may provide the initial steps for RCA, FTA, and logistical analy-sis, as necessary. Typical probability levels are:
• Frequent
• Reasonably probable
• Occasional
• Remote
• Extremely unlikely
The generic form for the qualitative analy sis is the same as the DFMEA or the PFMEA, depending on the task. However, there is also an alternative to this form and it looks like Figure 10.1.
QUANTITATIVE APPROACH
The quantitative method is used when failure rates, failure modes, failure ratios, and failure effects probabilities are known. These numbers are used
1123-77375_ch01_5P.indd 101 12/21/18 11:02 AM
Itemnumber
Itemfunction
Redundant systemSingle
component
Original date:
Revised date:
Team:
Approved by:
Page ___ of ___
System:
Part name:
Reference drawing:
Objective:
FMECA number:
Qualitative Failure Mode, Effects, and Criticality Analysis
Potentialfailuremode
Failurecause
Failureeffects O
ccur
renc
e
Seve
rity
RPN
(O ×
S)
Hav
e (N
)
Nee
d (M
)
Occ
urre
nce
Seve
rity
RPN
(O ×
S)
Remarks
Figure 10.1 A typical qualitative FMECA
1123-77375_ch01_5P.indd 102
12/21/18 11:02 AM
Failure Mode Effects and Criticality Analy sis (FMECA) 103
to calculate a criticality number to be used to prioritize items of concern. It is impor tant here to mention that these numbers are usually used after the design has been completed when confident data on the system can be used. (Often surrogate data are used for these numbers but they are substituted as soon as pos si ble with the actual data.) The quantitative analy sis is based on the failure mode criticality (Cm), which is the portion of the criticality num-ber for an item due to one of its failure modes. This results in a par tic u lar severity classification. Typical classifications based on the Mil- Std 1629 are:
Category I— Catastrophic: A failure that may cause death or com-plete system failure.
Category II— Critical: A failure that may cause severe injury, major property damage, or major system failure.
Category III— Marginal: A failure that may cause minor injury, minor property damage, or a delay/loss that will result in a degra-dation of the system.
Category IV— Minor: A failure not serious enough to cause injury, property damage, or system damage, but which will result in unscheduled delays of any kind.
This analy sis provides a substantial confidence for the risk being evaluated and may be used for other types of analyses, including FTA and RCM.
Mathematically, one may use several formulas to identify criticality. How-ever, before we introduce the formulas, let us examine some of the variables:
N = the amount of components that are redundant.
M = the required amount of components necessary.
Beta (β ) = the failure effect probability, which is used to quantify the described failure effect for each failure mode shown in the FMECA. This beta value represents the conditional probability or likelihood that the described failure effect will result in the identi-fied criticality classification, given that the failure mode occurs. The beta value also represents the team’s best judgment as to the likelihood that the loss or end- effect will occur. (For most items this probability will be 1, to indicate the worst pos si ble end- effect as a result of a failure mode.)
Alpha (α) = the probability (expressed as a decimal fraction) that the given part or item will fail in the identified mode. Determining alpha is done as a two- part pro cess for each component being ana-
1123-77375_ch01_5P.indd 103 12/21/18 11:02 AM
104 Chapter Ten
lyzed. First, the failure modes are determined and, second, modal probabilities are assigned. (If all alphas are identified for each item considered, their sum will be equal to 1.) Modal failures represent the dif fer ent ways a given part is known, or has been “observed, to fail.
On the other hand, a failure mechanism is a physical or chemical pro cess flaw caused by design defects, quality defects, part misap-plication, wear- out, or other pro cesses. It describes the basic reason for failure or the physical pro cess by which deterioration proceeds to failure. Once common part failure modes have been identified, modal probabilities (alpha) are assigned to each failure mode. This number represents the percentage of time, in decimal format, that the device is expected to fail in that given mode. ( Because the alpha and beta are very commonly confused, it is best to memo-rize that alpha is the failure mode ratio, the percentage of time or in what manner an item is going to fail. Beta, on the other hand, is the conditional probability of a failure effect occurring, given a specific failure mode. When a failure mode occurs, what percent-age of time is this going to be the end- effect?)
Failure rate (λp) = the number of failures per unit of time, typically expressed in failures per million hours or failures/106 hours. At the beginning of the evaluation, this failure rate is based more often than not on surrogate data. However, as information becomes avail-able the surrogate data are replaced with the actual data. (When ana-lyzing system failure rates where redundant or like components are used to accomplish a task, the failure rate must be adjusted to reflect the system failure rate. Furthermore, the source of the failure rate should be identified and recorded so that the validity of the data may be proved if there is a question or challenge about it.)
Now that we know the variables, let us see the actual formulas. Let us begin with the failure mode (modal):
1. Modal failure rate is the fraction of the item’s total failure rate based on the probability of occurrence of that failure mode. It is presented in a formula format as:
λm = αλp,
where
λm = the modal failure rate
α = the probability of occurrence of the failure mode (failure mode ratio)
1123-77375_ch01_5P.indd 104 12/21/18 11:02 AM
Failure Mode Effects and Criticality Analy sis (FMECA) 105
λp = the item failure rate
(Special Note: If there are three dif fer ent failure modes, then all failure rates will equal the item failure rate.)
2. Failure mode (modal) criticality number is a relative mea sure of the frequency of a failure mode. In other words, it is a mathematical calcula-tion to figure the rank importance based on the failure rate. It is shown in a formula format as:
Cm = βαλpt,
where
Cm = failure mode criticality
β = conditional probability of occurrence of next higher failure effect
α = failure mode ratio
λp = part failure rate
t = duration of applicable task
3. To identify a single failure probability, one may use the Poisson dis-tribution. A Poisson distribution is a discrete distribution with pa ram e ter (usually this is the mean) λ > 0, if for k = 0, 1, 2, . . . the probability mass function of X is given by:
f (k;λ)=Pr (X = k) = λ ke−λ
k!
where
e = the base of the natu ral logarithm (e = 2.71828 . . .)
k ! = the factorial of k (remember that 0! = 1)
The positive real number λ is equal to the expected value of X and also to its variance:
λ = E(X) = Var(X).
The Poisson distribution can be applied to systems with a large number of pos si ble events, each of which is rare. The Poisson distribution is some-times called a Poissonian.
4. To identify an item with criticality within a par tic u lar severity level, this approach identifies the item criticality (Cr). It is the criticality number associated with the item under analy sis. One may think of this Cr as the
1123-77375_ch01_5P.indd 105 12/21/18 11:02 AM
106 Chapter Ten
sum of the item’s failure mode criticality numbers, Cm, which result in the same severity classification. It is represented as:
Cr = (βαλpn=1
j
∑ t)n = (Cm )∑
where
Cr = item criticality
n = the current failure mode of the item being analyzed
j = the number of failure modes for the item being analyzed
β = conditional probability of occurrence of next higher failure effect
α = failure mode ratio
λp = part failure rate
t = duration of applicable task
Cm = failure mode criticality number
Once the functions have been completed, then the failure mode has to be identified. Typical ave nues for this identification are:
• Identify all known and potential (reasonable) failure modes.
• Be descriptive and as specific as pos si ble.
• Realize that “significant” and “reasonable” vary by proj ect.
• Identify all pos si ble failure causes.
5. Mean time between failure (MTBF) is the average time between fail-ures. It is used in maintenance and equipment failures. The equation for the MTBF is the sum of the operational periods divided by the number of observed failures. If the “down time” (with space) refers to the start of down-time (without space) and “up time” (with space) refers to the start of uptime (without space), the formula will be:
Mean timebetweenfailure = MTBF =(start of downtime-start of uptime)∑
number of failures
The MTBF is often denoted by the Greek letter θ or MTBF = θ. The MTBF can also be defined in terms of the expected value of the density function ƒ(t):
MTBF = tf (t)dt0
∞∫
1123-77375_ch01_5P.indd 106 12/21/18 11:02 AM
Failure Mode Effects and Criticality Analy sis (FMECA) 107
where ƒ is the density function of time until failure, satisfying the standard requirement of density functions: −MTBF = f (t)dt
0
∞∫ = 1.
In case of effects of redundancy, there are two possibilities to reduce system vulnerability and increase uptime. Failure rates need to be adjusted prior to using the above formulas considering repair time and without consideration of repair. Mathematically, we use the exponential distribution with constant time between failures. So, the formula with repair is:
λ(n−q)/n =n!(λ)q+1
(n − q −1)!(µ)q
where
n = number of active on line units
n! = n factorial
q = number of online units that can fail without system failure
μ = repair rate (μ = 1/MTTR, where MTTR is the mean time to repair in hours)
λ = failure rate for online unit (failures/hour)
The formula without repair is:
λ(n−q)/n =λ1ii=n−q
n
∑
If there is a situation of one standby off- line unit with n active online units required for success, with the off- line spare assumed to have a failure rate of zero, then the equations with and without repair are, respectively, shown as:
λn/n+1 =n[nλ + (1− P)µ]λµ + n (P +1)λ
, with repair
λn/n+1 =nλP +1
, without repair
where
n = number of active online units
n! = n factorial
q = number of online units that can fail without system failure
1123-77375_ch01_5P.indd 107 12/21/18 11:02 AM
108 Chapter Ten
μ = repair rate (μ = 1/MTTR, where MTTR is the mean time to repair in hours)
λ = failure rate for online unit (failures/hour)
P = probability that the switching mechanism will operate properly when needed (P = 1 with perfect switching)
On the other hand, if two active online units have dif fer ent failure and repair rates and one of the two is required for success, then we have:
λ1/2 =λAλB[(µA + µB)+ (λA + λB)](µA)(µB)+ (µA + µB)(λA + λB)
, with repair
λ1/2 =λA2λB + λAλB2
λA2 + λB2 + λAλB, without repair
These last two failure rates (λ), once calculated, should be substituted in the item 4 equation
Cr = (βαλpn=1
j
∑ t)n
to calculate the new criticality number that accounts for redundancy.
FORM
The generic form for the quantitative analy sis is the same as the DFMEA or the PFMEA, depending on the task. However, taking advantage of the formulas just discussed, there is also an alternative to this form; it is shown in Figure 10.2.
Sources for the Failure Mode
• Existing preventive maintenance task rec ords
• Operating rec ords
• Operator inputs
• Prior FMEAs, FTA, and FMECAs
• Input from engineering and subject matter experts
• Simulation studies data
Errors may be due to:
1123-77375_ch01_5P.indd 108 12/21/18 11:02 AM
Itemnumber
Itemfunction
Redun-dancy
Original date:
Revised date:
Team:
Approved by:
Page ___ of ___
System:
Part name:
Reference drawing:
Objective:
FMECA number:
Quantitative Failure Mode, Effects, and Criticality Analysis
Potentialfailuremode
Failurecause
Failurerate (λP)(Source)
Failureeffectprobability(β )
Failuremoderatio(α )
Operatingtime
Failuremodecriticalitynumber(Cr)
Itemcriticalitynumber(Cm)Se
verit
y
Hav
e (N
)
Nee
d (M
)
Remarks
Figure 10.2 A typical quantitative FMECA
1123-77375_ch01_5P.indd 109
12/21/18 11:02 AM
110 Chapter Ten
• Overdependence on data from simulation, experts, and nonspecific failure data
• Lack of specificity of the failed part/location
• Expediency (tackling insignificant as opposed to impor tant prob lems)
• Confusing failures with effects and vice versa
Ave nues for Identifying Effects
• List most severe effects.
• List reasonable effects.
• Differentiate between potential and pos si ble effects.
• Identify effects at point of functional failure.
When the effects are properly identified, good things happen in the totality of the analy sis. The effects must be evaluated for (a) local effect on the local part; (b) next higher effect on the function of the system/subsystem being analyzed; and (c) end- effect, or what the failure means to the task at hand. Specifically:
• The analy sis for the FMECA becomes much easier.
• Secondary damages are easier to be identified.
• Hidden failures are more easily identified.
• Functional failures are addressed more efficiently with the appro-priate effects assigned to them.
Sources for Effects
• Operating manuals
• Operator/engineer input
• Troubleshooting charts
• Test reports
• Failure/engineering investigations
• Subject matter experts
• Maintenance operators
1123-77375_ch01_5P.indd 110 12/21/18 11:02 AM
Failure Mode Effects and Criticality Analy sis (FMECA) 111
When appropriate effects are not identified, some of the prob lems may be attributed to these common errors:
• Assuming preventive maintenance exists
• Ignoring secondary issues
• Ignoring treatment of hidden failures
• Ignoring the appropriate level of the effect
Detection
Detection describes the method(s) by which functional failures are detected. It is impor tant to remember that the planned activities for control are not counted for the control ranking. What counts in the ranking is how effec-tive the control is in minimizing or eliminating the root cause of the failure. Typical controls are:
• Mistake proofing
• Visual alarms
• Reliability testing
• Specific testing at the end of the line (EOL)
• Gages and/or indicators
Of note here is the fact that (a) inspection is not a good control, and (b) operator error and training are not good control items.
Risk Priority Number
Generally, the RPN is not the final item when taking a specific action. The decision is based on priority, in the following order:
1. Severity
2. Severity × Occurrence
3. Severity × Occurrence × Detection
In other words, if we have the following situations:
(First option) 10 × 2 × 2 = 40
(Second option) 10 × 3 × 2 = 60
(Third option) 3 × 10 × 4 = 120
1123-77375_ch01_5P.indd 111 12/21/18 11:02 AM
112 Chapter Ten
The priority will be to solve the first item because of the high severity. Then we will address the second one because the severity and occurrence product is critical. Notice that the product of the third option is also 30, but the second option has a high severity, which takes pre ce dence. Lastly, we will evaluate the third one with the highest total RPN. (If one chooses the quantitative form, then the priority will be based on the Cr or Cm, depend-ing on the specific proj ect.)
Benefits
There are many benefits of conducting FMECA. Some are listed here:
• Prevents accidents
• Increases customer satisfaction
• Reduces costs for late changes
• Optimizes both design and pro cess robustness
• Documents the system and/or pro cess in addition to equipment
• Standardizes the methodology of risk assessment
• Allows for “ free” exchange of ideas and knowledge
• Reduces potential warranty costs
• Documents all the evidence of “due care” for liability concerns
• Improves the goodwill of the organ ization because it satisfies the corporate citizenship and accountability requirements for potential failures as well as catastrophic failures
Tools
Some of the common tools used in FMECA are:
• Scenario
• Brainstorming
• Simulation methods
• Reliability methods
• Affinity charts
• Cause- and- effect diagrams
• Checklists
1123-77375_ch01_5P.indd 112 12/21/18 11:02 AM
113
A control plan (CP) is a written description of the systems used to control and minimize product and pro cess variation. In addition, it specifies the pro cess monitoring and control methods used to control special character-istics. Special characteristics are the critical and significant characteristics for the product and/or pro cess. They are usually identified in the drawings and the FMEAs.
Generally speaking, a CP includes provisions for ongoing monitoring of pro cess control, stability, and capability.
PURPOSE
The purpose of a CP is to ensure that the customer requirements are met by supporting the manufacturing of quality products. This is done by hav-ing a plan that the operator follows and that monitors the key pro cess input variables (KPIV) as well as the key pro cess output variables (KPOV). As deviations occur from plan, the CP also includes a reaction plan that allows the operator to react to a given prob lem.
The primary intent of a CP is to create a structured approach for con-trol of pro cess and product characteristics while focusing the organ ization on the characteristics that are impor tant to the customer. Because of this, a CP must define the necessary systems that need to be in place to control the pro cess and minimize the occurrence of the failure modes identified in the FMEA.
When Used
The CP feeds into the operator work instructions, which must answer the following four questions:
11Control Plans
1123-77375_ch01_5P.indd 113 12/21/18 11:02 AM
114 Chapter Eleven
1. What has to be monitored?
2. How often must the monitoring occur?
3. How is the monitoring performed?
4. How will the operator react when a deviation occurs?
Types of Control Plans
There are three types of CPs:
1. Prototype
2. Preproduction
3. Production
During the product development cycle, CPs are used to document and communicate the initial plan for pro cess control. All of them are consid-ered legal documents as well as living documents. Therefore, if you are constructing and/or reviewing these documents, make sure they are correct because they are used often in legal disputes. A good practice is to initial all pages of the CP and FMEA.
In the automotive industry, there is also a dynamic control plan. This is a combination of the FMEA and the CP. No additional requirements are needed. The left side of the form is the FMEA proper and the right hand is exclusively for the CP.
BENEFITS
Fundamentally, the CP contributes to two basic benefits:
1. Assurance that customers will receive what they are paying for
2. Assurance that the operator will know what to do if there is a deviation in the pro cess or product
CONTENT OF A CONTROL PLAN
There are several items that a CP may cover. However, most often the most impor tant items are standardized. They are:
1. Part/pro cess number
2. Pro cess name/operation description
1123-77375_ch01_5P.indd 114 12/21/18 11:02 AM
Control Plans 115
3. All machines, devices, templates, tools, jigs for manufacturing
4. Characteristics to be controlled
5. Pro cess/product specifications/tolerances
6. Evaluation/mea sure ment techniques, including gage number and calibration information
7. Sampling plan, including size and frequency
8. Control method, including chart type, chart champion, and chart location
9. Reference to the reaction plan
All these items are usually kept in a tabular format. In some cases, depending on the industry, the CP may include more information.
FMEA/CONTROL PLAN LINKAGE
There must be a linkage between an FMEA, a CP, and a reaction plan. That linkage is based on three things:
1. Engineering documents such as:
a. Government regulations
b. Design requirements
c. Engineering material specifications
d. Critical manufacturing pro cess par ameters (if applicable), such as casting, welding, and heat treat
2. Quality data such as:
a. Warranty
b. Capability
c. Productivity
d. First- time throughput
e. Scrap
f. Things gone wrong (TGW)
g. Customer concerns
h. Employees and customer safety
1123-77375_ch01_5P.indd 115 12/21/18 11:02 AM
116 Chapter Eleven
i. Quality rejects
j. Field actions and/or stop ships
3. Pro cess knowledge such as:
a. Lessons learned
b. Installation drawings
c. Pro cess sheets
It is very impor tant to remember that CPs, DFMEAs, and PFMEAs are both legal and living documents; therefore, these documents must be updated whenever changes are made. Consequently, appropriate signa-tures must validate these changes. Figure 11.1 shows a typical linkage of DFMEA to PFMEA to CP.
Deficiencies in a Typical Control Plan
• Special characteristics not included in the FMEA or the CP
• Special controls not included in the FMEA
• Understanding severity, occurrence, and detection missing in FMEAs
• Failure mode not identified in FMEA
• Inadequate or no linkage between FMEA and CP
Design FMEA
Function
Kano modelinformation and/orQFD informationand/or corporateknowledge
System designspecifications
Partcharacteristics12345and so on
Failure Effect Severity Class Cause ControlsRecommendedaction
Process FMEA
Function Failure Effect Severity Class CauseControlsspecial
Recommendedaction
Design verificationplan and report
Sign-offreport
Regular control plan or dynamiccontrol plan. May change the classification symbol from critical to significant or significant to critical
Part drawing(inverted deltaand specialcharacteristics)
Figure 11.1 Linkage from DFMEA to CP
1123-77375_ch01_5P.indd 116 12/21/18 11:02 AM
Control Plans 117
• In effec tive control and/or gauging strategies
• Pro cess pa ram e ter controls not included or detailed in the CP
• In effec tive and/or inappropriate sampling plans (size and frequency)
• Inadequate and/or no reaction plan
• CP and/or reaction plan not followed
Tools Used
• SPC charts
• Sampling
• Mea sure ment system analy sis
• Jigs
• Pro cess flowchart
• Inspection
1123-77375_ch01_5P.indd 117 12/21/18 11:02 AM
1123-77375_ch01_5P.indd 118 12/21/18 11:02 AM
119
DFMEA TO PFMEA TO PROCESS FLOW DIAGRAM TO CONTROL PLAN
It is imperative for the reader to understand that in generating any FMEA there must be “inputs” and “outputs.” These linkages, therefore, determine the inputs and the outputs as well as their interrelationships with the ele-ments of the FMEA as they are identified in the discussion and/or the form. In a sense, they are the pro cess output that summarizes error states, noise factors, and the associated design controls, and also an input into the design verification plan. These linkages are summarized for concept, design, and pro cess FMEAs, as follows.
Design Concept Input
• Corporate requirements
• Regulatory requirements
• Customer requirements
• Benchmarking techniques
• Historical per for mance information
• Product- specific needs, wants, and expectations (results of a quality function deployment) ranked by customer’s importance
• Generic system design specifications (SDSs)
• Preproduct development targets for system per for mance
• SDSs for the system
• Corporate knowledge
12Linkages
1123-77375_ch01_5P.indd 119 12/21/18 11:02 AM
120 Chapter Twelve
Pro cess Concept Input
• Customer requirements
• Regulatory requirements
• Historical per for mance information
• Benchmarking techniques
• Corporate knowledge
Design Concept Output
• Program target values or recommendations
• Recommendations for new generic testing (now required DVP input)
• Specific system/subsystem or component design specification (specific SDSs; geometric dimensioning and tolerancing (GDT) information; validation criteria including engineering specifica-tions, reliability targets, and robustness needs)
Pro cess Concept Output
• Program target values or recommendations
• Recommendations for new generic testing (now required DVP input)
Design Input
• Concept FMEA → Recommendations for new generic testing now required DVP input → design validation pro cess (DVS) and methods and schedule
• P- diagram
• Boundary diagram
• Historical design per for mance
• Information including reliability
• Interface matrix
• Specific system/subsystem or component design specifications (specific SDS; GDT information; validation criteria including engi-neering specifications, reliability targets and robustness needs)
1123-77375_ch01_5P.indd 120 12/21/18 11:02 AM
Linkages 121
Design Output
• Potential criteria and/or significant characteristics → Prototype control plans
• Design information related to potential strategies
• Reliability and checklist
• New DVS
• Test methods or revisions based on FMEA analy sis
• Other recommended actions for product robustness → Target per-for mance review and validation
• Other recommended actions for future products or programs → Target per for mance review and validation
Pro cess Input
• Design FMEA → Potential critical and/or significant characteris-tics → Prototype control plans
• Design FMEA → Design information related to potential strategies
• Design FMEA → Reliability and robustness checklist
• Design Concept FMEA → Program target values or recommendations
• Pro cess Concept FMEA → Program target values or recommendations
• Pro cess Concept FMEA → Recommendation to new generic pro cess controls
• Historical controls and control plan information
• Gaging information specified using GDT
• Problem- solving and FMEA data
• Characteristic matrix
• Pro cess flow and specification information
• P- diagram
• Engineering specification
• Historical manufacturing per for mance information
1123-77375_ch01_5P.indd 121 12/21/18 11:02 AM
122 Chapter Twelve
Pro cess Output
• Safety sign- off
• Confirmed critical and significant characteristics → D(esign) and R(eview) sign- off and prelaunch control plans
• Prelaunch control plans → Production control plans → D and R sign- off
• Recommended manufacturing actions for product robustness
• Other recommended actions for future products or programs
Machinery Output
• Operator safety sign- off
• Production control plan
• Design information related to potential strategies
• New design/equipment methods or revisions based on FMEA analy sis
• Other recommended actions for equipment specifications → Target per for mance review and validation
• Other recommended actions for future equipment → Target per for-mance review and validation
1123-77375_ch01_5P.indd 122 12/21/18 11:02 AM
123
OVERVIEW OF TYPICAL TOOLS USED IN FMEA
This chapter provides the reader with a quick reference of some typical and most often used tools in the problem- solving pro cess and especially in the FMEA. The most basic of all problem- solving methodologies we believe is the eight- stage pro cess, which of course is a derivative of the scientific approach. The individual stages are:
1. Identify
2. Scope
3. Define
4. Analyze
5. Implement
6. Evaluate
7. Follow-up
8. Continually improve
Affinity Diagram
You create an affinity diagram by using a number of small cards (1″ × 3″), each inscribed with an idea or solution. The affinity diagram is based on brainstorming and a cause- and- effect diagram.
13Tools
1123-77375_ch01_5P.indd 123 12/21/18 11:02 AM
124 Chapter Thirteen
What it does: This tool is useful when (1) facts/thoughts are in chaos, (2) a breakthrough in traditional concepts is needed, and (3) support for justify-ing a proposed implementation is needed.
When to use it:
Stage 1: Identify
Stage 3: Define
Stage 4: Analyze
Box and Whisker Plot
An alternative to a histogram, this plot has the appearance of a rectangle (the box) with a horizontal and a vertical line passing through its center and extending outside the box (the whisker).
What it does: Displays the main features of a data set and permits simple comparisons of several data sets.
When to use it:
Stage 4: Analyze
Stage 5: Implement
Stage 6: Evaluate
Brainstorming
This idea- generating technique relies on team participation and interaction. All ideas are noted before any less practical ones are discarded.
What it does: Enables a team to create as many ideas as pos si ble in as short a time as pos si ble.
When to use it:
Stage 1: Identify
Stage 5: Implement
1123-77375_ch01_5P.indd 124 12/21/18 11:02 AM
Tools 125
Cause- and- Effect Diagram
This diagram is a simple means for finding the causes of an effect (prob-lem) by an individual or a team. It is also known as the fishbone diagram because of its shape.
What it does: Graphically shows the relationship of causes and sub- causes to an identified effect. Helps reveal potential root causes.
When to use it:
Stage 4: Analyze
Stage 6: Evaluate
Computer Simulation
A computer- based technique prob ably requires the assistance of operations research to prepare the programs.
What it does: Creates a pictorial repre sen ta tion of an area layout showing the movement of items within that area. Computer simulation is a means of solving what-if questions and examining the effects of vari ous related data over long- and short- term periods.
When to use it:
Stage 4: Analyze
Stage 6: Evaluate
Stage 7: Follow-up
Stage 8: Continually improve
Control Chart- c
Standard control chart for the total number of nonconformities, based on a constant sample size.
What it does: Graphically displays stability of a pro cess (e.g., total num-ber of errors in a batch of 100 forms rather than just the number of faulty forms).
When to use it:
Stage 4: Analyze
1123-77375_ch01_5P.indd 125 12/21/18 11:02 AM
126 Chapter Thirteen
Stage 5: Implement
Stage 6: Evaluate
Control Chart- Median and R
Standard chart that is an alternative to the X- bar and R chart for the control of pro cesses. It is less sensitive to trends, however, and under some circum-stances is considered to be more difficult to construct.
What it does: Graphically displays stability of a pro cess. Yields similar information to X- bar and R charts but has several advantages. (1) It is easier to use, because daily calculations are not required; (2) individual values and medians are plotted, with the median chart showing spread of pro cess output and giving an ongoing view of pro cess variation; (3) it shows where noncon-formities are scattered through a more or less continuous flow of a function; and (4) it shows where nonconformities from dif fer ent areas may be evident.
When to use it:
Stage 4: Analyze
Stage 5: Implement
Stage 6: Evaluate
Control Chart- np
Standard control chart similar to the c- chart; must be used if the sample sizes vary.
What it does: Graphically displays stability of a pro cess. Mea sures the actual number of nonconforming items rather than total number of faults (e.g., total number of faulty forms in a batch irrespective of faults in any one form).
When to use it:
Stage 4: Analyze
Stage 5: Implement
Stage 6: Evaluate
1123-77375_ch01_5P.indd 126 12/21/18 11:02 AM
Tools 127
Control Chart- p
Standard control chart requiring a constant sample size. It charts either con-forming or nonconforming items.
What it does: Graphically displays stability of a pro cess. Mea sures the actual number of conforming and nonconforming items rather than total number of faults. Expresses numbers in either fractional or percentile terms ( whether conforming or nonconforming items are used) of a total sample (e.g., total number of faulty forms in a batch irrespective of the number of faults in any one form).
When to use it:
Stage 4: Analyze
Stage 5: Implement
Stage 6: Evaluate
Control Chart- u
Standard control chart that is similar to the c- chart; must be used if the sample sizes vary.
What it does: Graphically displays stability of a pro cess (e.g., total number of errors in a batch of 100 forms rather than just the number of faulty forms).
When to use it:
Stage 4: Analyze
Stage 5: Implement
Stage 6: Evaluate
Control Chart X- bar and R
This standard control chart is the most used chart. It requires that a number of consecutive units be taken n times per work period and analyzed for specific criteria.
What it does: Graphically displays pro cess stability. Shows data in terms of spread (piece- to- piece variability) and location (pro cess average). X- bar covers averages of values in small subgroups (sample taken); it is known as mea sure of location. R chart deals with a range of values within each sample (highest minus lowest); it is known as a mea sure of spread.
1123-77375_ch01_5P.indd 127 12/21/18 11:02 AM
128 Chapter Thirteen
When to use it:
Stage 4: Analyze
Stage 5: Implement
Stage 6: Evaluate
Control Chart X- bar and S
Standard control chart similar to X- bar and R chart, however, the S chart considers standard deviation and is more complicated to calculate.
What it does: Graphically displays stability of pro cess S factor, which is a more accurate indicator of pro cess variability, especially with larger sample sizes. This chart is less sensitive in detecting special causes of varia-tion that produce only one value in a subgroup as unusual.
When to use it:
Stage 4: Analyze
Stage 5: Implement
Stage 6: Evaluate
Cross- Functional Pro cess Map
A series of columns representing departments across which the flow of a pro cess is mapped.
What it does: Allows a map of the pro cess to be shown, its order of pre-ce dence, and which departments it is routed through.
When to use it:
Stage 4: Analyze
Stage 5: Implement
Stage 6: Evaluate
Decision Tree
Decision trees are excellent tools for helping you to choose between several courses of action. They provide a highly effective structure within which you can lay out options and investigate the pos si ble outcomes of choosing
1123-77375_ch01_5P.indd 128 12/21/18 11:02 AM
Tools 129
those options. They also help you to form a balanced picture of the risks and rewards associated with each pos si ble course of action.
What it does: A decision tree lays out the prob lem so that all options can be challenged. It helps in the analy sis of the pos si ble consequences of a deci-sion by providing the framework to quantify the values of outcomes and the probabilities of achieving them. It is an excellent tool to identify and make the best decisions on the basis of existing information and best guesses.
When to use it:
Stage 4: Analyze
Stage 6: Evaluate
Design of Experiments
Several methods are available, of which the following are examples: (1) Taguchi method, including signal to noise (S/N) ratios; (2) accelerated testing methods— not in the reliability sense but rather in maximizing the results of multiple tests rather than one at a time; and (3) factorial and frac-tional factorial designs.
What it does: Factors common- cause variation into its components in order to optimize pro cess/product variables and reduce variation.
When to use it:
Stage 3: Define
Stage 4: Analyze
Stage 5: Implement
Stage 6: Evaluate
Dot Plot
A display that is somewhat similar to a histogram, but the axis is divided into many more divisions.
What it does: Usually it is used when there are insufficient criteria to construct a histogram or a box and whisker plot. It is used for comparison purposes.
1123-77375_ch01_5P.indd 129 12/21/18 11:02 AM
130 Chapter Thirteen
When to use it:
Stage 4: Analyze
Stage 5: Implement
Stage 6: Evaluate
Stage 7: Follow-up
Failure Mode and Effect Analy sis
A what-if approach to evaluating design weaknesses that starts at the com-ponent level and proceeds through the complete system.
What it does: This bottom-up approach identifies potential product/pro-cess weaknesses. It begins with a study of known failure modes for each component of a product or pro cess. Using physical analy sis or mathematical models, a determination is made of the effect of failure on a component, sub-system, or complete system.
When to use it:
Stage 3: Define
Stage 4: Analyze
Stage 6: Evaluate
Stage 7: Follow-up
Stage 8: Continually improve
Fault Tree Analy sis
A graphical display, resembling the shape of tree roots, that permits one to identify multiple causes of failures and display interactions between causes.
What it does: Begins with the definition of an undesirable event and traces that event through the system to identify basic causes; it is a top- down appraisal.
When to use it:
Stage 3: Define
Stage 4: Analyze
Stage 6: Evaluate
1123-77375_ch01_5P.indd 130 12/21/18 11:02 AM
Tools 131
Stage 7: Follow-up
Stage 8: Continually improve
Function Tree
A graphical repre sen ta tion of functions to ensure clear, total team under-standing of actionable and mea sur able items. From left to right, the question examined is, “How is function achieved?” and from right to left, the ques-tion is, “Why is function included?”
What it does: Provides an or ga nized brainstorm (verb- noun) approach to identify the essential features of a product (or pro cess sometimes). In addi-tion, a function tree helps to ensure that “unspoken” and “spoken” require-ments are defined.
When to use it:
Stage 1: Identify
Stage 3: Define
Stage 4: Analyze
Stage 6: Evaluate
Stage 7: Follow-up
Stage 8: Continually improve
Gauge Repeatability and Reproducibility
A mea sure ment of the repeatability and reproducibility of a gauge and the operator, respectively.
What it does: Mea sures variations in gauges and test equipment to ascer-tain (1) bias in accuracy due to improper calibration, (2) variation in preci-sion due to operation of the device, (3) variation in reproducibility when dif fer ent people use the equipment, and (4) variations in stability due to changes in environment, power fluctuations, and so on.
When to use it:
Stage 4: Analyze
1123-77375_ch01_5P.indd 131 12/21/18 11:02 AM
132 Chapter Thirteen
Graphs- Bar Chart
An x- y type of graph that uses narrow rectangular bars to signify frequen-cies of occurrence.
What it does: Compares discrete data from a number of sources (e.g., absenteeism on specific days in several offices).
When to use it:
Stage 4: Analyze
Stage 6: Evaluate
Graphs- Gantt Chart
An x- y type of graph that uses narrow rectangles or lines, usually parallel to the x axis, to represent periods of time on a specific task or tasks.
What it does: Displays, to scale, time to perform a unit of work that occurs at a given point in the pro cess. Allows a comparison of its position in the pro cess with other units of work and how they relate. It is a helpful to use with a pro cess flowchart to highlight and quantify information in both pre- and post- investigation situations.
When to use it:
Stage 4: Analyze
Stage 6: Evaluate
Stage 7: Follow-up
Graphs- Pie Chart
A circle divided into sectors, each of which represents a factor. The pie chart’s area is a proportion of the whole, expressed as a percentage.
What it does: Shows all the criteria involved in a pro cess/survey and individual percentages of the total. An area of the circle can be used to demonstrate a change/compare circumstances (e.g., a chart showing the car market by year and a specific com pany’s share of that market).
When to use it:
Stage 4: Analyze
Stage 6: Evaluate
1123-77375_ch01_5P.indd 132 12/21/18 11:02 AM
Tools 133
Histogram
An x- y graph that uses narrow rectangles to display frequencies of occur-rence of a specific set of data.
What it does: Gives a picture of the frequency of occurrence for a range of specific data and demonstrates its normalcy (or lack thereof). In other words, if the center points of the top of each column of recorded frequen-cies were joined by a continuous line (normal distribution curve), the shape produced would be that of a bell, more or less distributed around the median (central point).
When to use it:
Stage 4: Analyze
Stage 6: Evaluate
Stage 7: Follow-up
Kano Model
The Kano model is a theory of product development and customer satisfac-tion developed in the 1980s by Professor N. Kano. The model separates the delightful, per for mance, and basic requirements. The strength of the model lies in the fact that over time it recognizes that the delightful requirements eventually become basic.
What it does: The model is used to prioritize the critical to quality char-acteristics, as defined by the voice of the customer. In essence, the model provides the insights into the dynamics of customer preferences and there-fore helps optimize the design and/or pro cess with both “must” have and “want” to have items.
When to use it:
The Kano model is used primarily in concept and design FMEA. Quite often it is used as a complement with QFD.
Stage 1: Identify
Stage 4: Analyze
Stage 6: Evaluate
1123-77375_ch01_5P.indd 133 12/21/18 11:02 AM
134 Chapter Thirteen
Operational Definitions
Terms necessary for the common understanding of a pro cess.
What it does: Contains three ele ments: (1) a set of criteria, (2) a test by which criteria are applied, and (3) a yes/no result from the test. The result must be accepted by all who use it.
When to use it:
Stage 1: Identify
Stage 2: Scope
Stage 3: Define
Stage 4: Analyze
Stage 5: Implement
Stage 6: Evaluate
Stage 7: Follow-up
Stage 8: Continually improve
Pareto Diagram
An x- y bar chart with the bars prioritized in descending order (from left to right) and distinguished by a cumulative percentage line. It is based on the 80/20 rule, which states that about 80% of improvement in an effect can be achieved by acting on 20% of the causes.
What it does: The prioritization of the inputs ( causes) indicate those that should be considered first (in other words, those on the left of the chart).
When to use it:
Stage 1: Identify
Stage 4: Analyze
Stage 6: Evaluate
Stage 8: Continually improve
1123-77375_ch01_5P.indd 134 12/21/18 11:02 AM
Tools 135
PERT or Critical Path Analy sis
Road map of interdependent ele ments within a pro cess and containing cri-teria indicating critical routes through the ele ments.
What it does: Illustrates ele ments within a pro cess and indicates earliest and latest event timing against each ele ment. Clarifies order of sequential priority within pro cess and allows a critical path through the pro cess to be identified.
When to use it:
Stage 4: Analyze
Stage 6: Evaluate
Stage 7: Follow-up
Pro cess Flowchart
A road map of the pro cess from supplier(s) to customer(s).
What it does: Illustrates/clarifies events in a pro cess and the events between them. Assists in highlighting (1) the pres ent situation, (2) differences between what should be/is thought to be happening and the actual situation, (3) the pro-posed situation, and (4) the potential prob lem areas (gaps and so on).
When to use it:
Stage 1: Identify
Stage 2: Scope
Stage 3: Define
Stage 4: Analyze
Stage 6: Evaluate
Program Decision Pro cess Chart
A program decision pro cess chart (PDPC) is a tree- type chart.
What it does: Maps conceivable events/contingencies that occur when moving from prob lem to statement to pos si ble solutions. A PDPC is used to plan pos si ble chains of events that need to occur when a problem/goal is unfamiliar.
1123-77375_ch01_5P.indd 135 12/21/18 11:02 AM
136 Chapter Thirteen
When to use it:
Stage 4: Analyze
Stage 6: Evaluate
Pugh Technique
A chart that shows alternatives on the x axis and base criteria on the y axis.
What it does: Allows comparisons between current concept/design, criteria required, and a number of alternative solutions. Each alternative is compared with the current situation, requirement by requirement, and summarized in a form of total +/− points, which indicate the alternative to use.
When to use it:
Stage 4: Analyze
Stage 6: Evaluate
Quality Function Deployment
An array that enables a comparison of customer requirements against a number of design ele ments. As a tool, QFD also allows areas of conflict to be plotted.
What it does: Acts as a broad management system that assists in translat-ing the voice of the customer into operational definitions that can be used to produce and deliver the product/ser vice desired by the customer. High-lights conflicting customer requirements so that they can be reconciled in an optimum manner.
When to use it:
Stage 1: Identify
Stage 2: Scope
Stage 3: Define
Stage 4: Analyze
Stage 5: Implement
Stage 6: Evaluate
1123-77375_ch01_5P.indd 136 12/21/18 11:02 AM
Tools 137
Stage 7: Follow-up
Stage 8: Continually improve
Regression Analy sis
A procedure for fitting a mathematical model (expressed in terms of equa-tions with variables and coefficients; e.g., y = Mx + b) to a set of data.
What it does: Allows one to explore factors in a given set of data (e.g., barometric and humidity effects on CO production from combustion). It is also used in instrument calibration, design, and pro cess analy sis.
When to use it:
Stage 3: Define
Stage 4: Analyze
Stage 6: Evaluate
Reliability Analy sis
A series of statistical formulas, tables, and graphs based on probability.
What it does: As a broad area of study, reliability analy sis is concerned with random occurrences of undesirable events/failures, during the life of a physical system.
When to use it:
Stage 4: Analyze
Stage 6: Evaluate
Run Chart
An x- y type of graph that compares a mea sure ment (%, $, and so on) on the y axis with time or sequence (days, order, and so on) on the x axis.
What it does: Monitors a pro cess to assess whether the long- range average is changing and, if it is changing, whether it is improving or deteriorating.
When to use it:
Stage 4: Analyze
Stage 5: Implement
1123-77375_ch01_5P.indd 137 12/21/18 11:02 AM
138 Chapter Thirteen
Stage 6: Evaluate
Stage 7: Follow-up
Scatter Diagram
An x- y graph that examines the possibility of a relationship between two variables.
What it does: Checks for pos si ble cause- and- effect relationships. The diagram cannot prove one variable causes another, but it makes clear whether a relationship exists and the strength of the relationship.
When to use it:
Stage 4: Analyze
Stage 6: Evaluate
Shared and Interlocking Objectives Matrix
Chart that lists vari ous departments on both the x and y axis.
What it does: Enables department heads to specify their requirements from each of the other departments in order to achieve a certain goal (in other words, each customer can specify requirements from each supplier).
When to use it:
Stage 2: Scope
Stem and Leaf Plot
A vertical line with data to the left of the line— known as the stem and indi-vidual criteria as stem ends. Criteria to right of the line are known as the leaf. The stem and leaf plot is an alternative to the histogram.
What it does: Because it is quicker to produce than a histogram, it allows data to be viewed in traditional column format.
When to use it:
Stage 4: Analyze
Stage 6: Evaluate
1123-77375_ch01_5P.indd 138 12/21/18 11:02 AM
Tools 139
Surveys
A survey is an investigative questioning technique.
What it does: Through programmed questioning of the supplier and customer, a picture is formed of (1) prob lems encountered, (2) customer desires, and (3) the shape of the pro cess, and so on.
When to use it:
Stage 1: Identify
Stage 2: Scope
Stage 7: Follow-up
Stage 8: Continually improve
Time Series Forecasting
A series of statistical formulas, tables, and graphs.
What it does: As a broad area of study, it takes data mea sured at discrete, equispaced time intervals and constructs mathematical models for forecast-ing over a given period of time (lead time).
When to use it:
Stage 4: Analyze
Stage 6: Evaluate
Stage 7: Follow-up
Stage 8: Continually improve
Tree Analy sis
Also called a systematic diagram, an analytical tree, and a hierarchy dia-gram, tree analy sis is a functional tool that helps in communicating details to others when analyzing pro cesses in detail and/or evaluating several potential solutions.
What it does: Helps the team to or ga nize the discussion from generali-ties to specifics. In other words, it is used to break down broad categories into finer and finer levels of detail. It is an excellent tool to be used as a follow-up to an affinity diagram or relations diagram to focus on the newly discovered key issues.
1123-77375_ch01_5P.indd 139 12/21/18 11:02 AM
140 Chapter Thirteen
When to use it:
Stage 1: Identify
Stage 4: Analyze
Stage 6: Evaluate
1123-77375_ch01_5P.indd 140 12/21/18 11:02 AM
141
AFTER FMEA
Most FMEA teams think that once the FMEA form is complete they are “done.” That is the furthest from the truth. In order for an FMEA to be complete the team must review and trou ble shoot the FMEA for any errors, inconsistencies and or correct evaluations of severity, root causes and rec-ommended actions. Of paramount importance is to evaluate the correct relations between failure mode, effect, and root cause.
1. Review the FMEA.
2. Highlight the high- risk areas based on the RPN.
3. Identify the critical and/or major characteristics based on your classification criteria.
4. Ensure that a CP exists and is being followed.
5. Conduct capability studies.
6. Work on pro cesses that have Cpk or Ppk of less or equal to 1.33.
7. Work on pro cesses that have Cpk or Ppk greater than 1.33 to reduce variation and reach a Cpk or Ppk of greater or equal to 2.0.
Header
• Use correct FMEA form.
• Include accurate dates, including revisions.
• Identify pro cess step/component name/system name.
• Identify all team members.
14Troubleshooting an FMEA
1123-77375_ch01_5P.indd 141 12/21/18 11:02 AM
142 Chapter Fourteen
• Include the FMEA number.
• Identify who prepared the FMEA.
• Leave no blank fields.
Function/Purpose
• Write this description as a verb- noun- measurable construction.
• Include technical specifications and engineering requirements, with pro cess requirements stated.
• Describe any special conditions.
• Identify all functions.
Potential Failure Mode
• Failure modes address no failures, partial/over function/degraded over time failures, intermittent failures, and unintended failures.
• PFMEA describes a failure mode as what the product would be rejected for.
• No causes or effects should be listed as failure modes.
• Each failure should be associated with a function and special condi-tions identified for similar failure modes.
• Question any function with only one failure mode.
Potential Failure Effect(s)
• Are all effects considered, including customer, government, and product?
• Does the effect relate to a failure mode?
• Are any causes incorrectly listed?
• Have you noted the severity behind each effect or behind the worst, for historic purposes?
• Are you using “correct” effect phrasing that is clear, concise, and accurate?
1123-77375_ch01_5P.indd 142 12/21/18 11:02 AM
Troubleshooting an FMEA 143
Severity
• Government regulations or safety are rated 9 and 10.
• Question any rating of 1.
• Only the highest severity ranking is entered for each failure mode.
Classification
• Classification marks for each special characteristic:
– Are they correct?
• DFMEA— use potential critical (YC) and potential significant (YS) only.
• PFMEA— confirm critical characteristics (CCs), significant charac-teristics (SCs), operator safety (OS), and high impact (HI).
• Verify special controls for each CC and SC.
Potential Cause(s)/Mechanism(s) of Failure
• Are causes with government/safety effects (9–10) at the root level? Note part characteristic if appropriate.
• Do failure modes have multiple causes listed?
• Are both assumptions used in development of causes?
• Are any effects incorrectly listed?
• DFMEA—no “operator error” or “machine malfunction” should be listed.
• PFMEA—no “operator error” or “machine malfunction” should be listed, especially for significant or critical characteristics or OS.
• Have noise factors been considered?
Occurrence
• Question any rating of 1.
• Was the occurrence table followed?
• Question any rating of 10. Does a rating of 10 have an action?
1123-77375_ch01_5P.indd 143 12/21/18 11:02 AM
144 Chapter Fourteen
• Set controls to determine the occurrence rating.
• Is there one rating per cause?
Prevention Controls
• Are there prevention controls in place?
Appropriate Controls Applied
• Are the prevention controls effective?
• Are there up- front DFMEA and design concept FMEA (D- CFMEA) controls (i.e., early test strategies, not downstream production controls)?
• Have you made sure there are no pro cess controls on DFMEA?
• Are detection and prevention mechanisms identified in correct columns?
• Was careful consideration given to identifying specific controls?
• Are these actual, planned controls, not “wished for” controls?
• Are any prevention methods not rated as detection?
Detection
• Question any rating of 1.
• Verify that visual inspection is not rated less than 4.
• Verify that sampling is not overrated.
• Ensure that the best detection rating is used for more than one control.
• Is there one (best) rating per control set?
Risk Priority Number
• There should be one RPN per cause.
• Some organ izations do not recommend a threshold value for RPNs.
1123-77375_ch01_5P.indd 144 12/21/18 11:02 AM
Troubleshooting an FMEA 145
Recommended Action
• Recommended action taken should be listed by priority for each S, or S × O or RPN.
• DFMEA— recommended actions are not pro cess actions or controls.
• DFMEA— all recommended actions point at design itself.
• PFMEA— recommended actions for special characteristics should list the special controls to be put in place.
• Use “none” or “none at this time” instead of leaving blanks. The classification column may be blank only and only if there are no special characteristics identified.
• Is there a recommended action on all “classified” items?
Responsibility/Target Completion Date
• Name and date should be specified for each recommended action.
• Do not use “TBD” or “Ongoing” as a date.
• Responsibility for completion should be assigned to a team member.
Actions Taken/Revised Ratings
• Action taken should be listed only after the action is completed.
• List the action results; do not just say “completed” or done.
• Dates completed and document reference numbers may be helpful for historic purposes.
• Enter revised ratings when actions taken are listed— even if there is no rating change.
• Consider additional actions if the first action was not successful.
1123-77375_ch01_5P.indd 145 12/21/18 11:02 AM
1123-77375_ch01_5P.indd 146 12/21/18 11:02 AM
147
COMMON TEAM PROB LEMS
• Poor team composition (not cross- functional or multidisciplinary):
– Low expertise in FMEA
– Not multilevel
– Low experience/expertise in the product
– One- person FMEA
• Lack of management support
• Not enough time
• Too detailed (pro cess could go on forever)
• Arguments between team members (base opinions on facts and data)
• Lack of team enthusiasm/motivation
• Getting team to start and stay with the pro cess
• Proactive vs. reactive (a “before the event” not “ after the fact” exercise)
• Doing it for the wrong reason
COMMON PROCEDURAL PROB LEMS
• Confusion about, poorly defined, or incomplete functions/failure modes/effects or causes.
15Typical Concerns When Conducting an FMEA
1123-77375_ch01_5P.indd 147 12/21/18 11:02 AM
148 Chapter Fifteen
• Subgroup discussion.
• Using symptoms or superficial causes instead of root causes.
• Confusion about ratings as estimates and not “absolutes.” It will take time to be consistent.
• Confusion about the relationship between causes, failure modes, and effects.
• Using “customer dissatisfied” as a failure effect.
• Shifting design concerns to manufacturing and vice versa.
• Doing FMEAs by hand:
– Dependent on the engineers’ “printing skills”
– RPNs or criticality cannot be ranked easily
– Hard to update
– Complicated FMEAs take up too much space
– Time- consuming
– No one wants to be the “recorder” when it’s done manually
– Inefficient means of storing and retrieving information
• Note: With FMEA software, all the above- listed prob lems are eliminated.
• Working nonsystematically on the form. (It is suggested that the failure analy sis should pro gress from left to right, with each column being completed before the next one begins.)
• Unwillingness to assume responsibility for recommended actions.
• Doing a “Reactive FMEA” as opposed to a “Proactive FMEA.” (An FMEA is best applied as a prob lem prevention tool, not a problem- solving tool, although one may use it for both. However, the value out of a Reactive FMEA is much less.)
• Not having “robust” FMEA terminology: A robust communication pro cess is one that delivers its “function” (imparting knowledge and understanding) without being affected by “noise factors” (varying degrees of training). Simply stated, the pro cess should be as clear as pos si ble with minimum possibility for misunderstanding.
1123-77375_ch01_5P.indd 148 12/21/18 11:02 AM
Typical Concerns When Conducting an FMEA 149
INSTITUTIONALIZING FMEA IN YOUR COM PANY
• Institutionalizing FMEA is challenging, and its success largely depends on the culture in the organ ization as well as why it is being utilized. Among the main considerations:
– Selecting pi lot proj ects (start small and build successes)
– Identifying team participants
– Developing and promoting FMEA successes
– Developing “templates” (databases of failure modes, functions, controls, etc.)
– Addressing training needs
1123-77375_ch01_5P.indd 149 12/21/18 11:02 AM
1123-77375_ch01_5P.indd 150 12/21/18 11:02 AM
151
FMEAs were first introduced in the 1940s in the US military. However, the methodology grew when the manned space missions started in the 1960s. After this introduction the FMEA as a risk- mitigating methodology took hold in many industries, to the point where special standards have been developed over the years to make sure the appropriate risks are defined and their control is appropriate and applicable.
This chapter is not going to address all applications in all industries. Instead, we are going to address selected industries with quite diverse needs as well as expectations. Our focus is to demonstrate the flexibility of the FMEA and its application in any industry, no matter what the product or pro cess is.
AUTOMOTIVE
The auto industry did not widely adopt FMEAs until the late 1970s. Ford Motor Com pany introduced them for safety and regulatory items to improve automotive designs and manufacturing, specifically in response to the Ford Pinto fuel tank issues. The success that Ford had with the FMEA spread throughout the industry and now is part of the specific requirements of all automotive companies, both domestic and international. Typical FMEAs in the automotive world are the CFMEA, DFMEA, and PFMEA. Because of the many OEMs and their dif fer ent requirements, the AIAG was formed to standardize the FMEA. Today the AIAG has published its fourth edition with generic guidelines for all to follow.
16FMEAs Used in Selected
Specific Industries
1123-77375_ch01_5P.indd 151 12/21/18 11:02 AM
152 Chapter Sixteen
AEROSPACE
In the aerospace industry the risk factors are many and quite often cata-strophic. As such, for automobile references the SAE J1739 standard is used and for all practices for nonautomobile applications the standard ARP5580 is used.
Both standards recommend the FMEA, which encompasses func-tional, interface, and detailed FMEA, as well as certain preanalysis activi-ties (FMEA planning and functional requirements analy sis), postanalysis activities (failure latency analy sis, FMEA verification, and documentation), and applications to hardware, software, and pro cess design.
The focus of the aerospace FMEA is for organ izations to assess safety and reliability of system ele ments as part of their product improvement pro-cesses. The general approach is the traditional FMEA, although quite often the FMECA is used to account for criticality.
SOFTWARE
The focus of the software FMEA (SWFMEA) is to evaluate individual risks by differentiating between high- risk and low- risk components, modules, and functions. This approach makes risk- oriented development of software- intensive systems pos si ble.
The actual FMEA may be performed at dif fer ent phases of software development such as analy sis, design, coding, module test, system test, and final test (field test). The actual approach is the same as the generic one. However, in most cases the SWFMEA follows the rationale of the design FMEA.
A typical SWFMEA is used for architecture or design review during software development, which means before the implementation of the soft-ware. It may not be executed on software source code.
CHEMICAL/PHARMACEUTICAL
In both chemical and phar ma ceu ti cal industries risk management is of par-amount concern. The risk is mitigated through a formal risk- management methodology that includes planning, design, and the pro cess of a par tic u-lar proj ect. This methodology is followed in order to avoid proj ect failure owing to anticipated or unanticipated events. Whereas the FMEA is indeed a flexible yet power ful tool and is used as a good option for identifying and
1123-77375_ch01_5P.indd 152 12/21/18 11:02 AM
FMEAs Used in Selected Specific Industries 153
controlling adverse risk, quite often in both industries the FMECA is fol-lowed because of it criticality nature. Typical concerns in both industries are:
• Understanding the risk
• Defining the specific risk
• Forming the appropriate and applicable team
• Drawing a flowchart of the pro cess
• Analyzing and evaluating the pro cess
• Selecting the appropriate outcome
• Taking action on the most likely outcome
• Mea sur ing the outcome and comparing with the original
• Instituting improvement in the pro cess and similar pro cesses
1123-77375_ch01_5P.indd 153 12/21/18 11:02 AM
1123-77375_ch01_5P.indd 154 12/21/18 11:02 AM
155
AUTOMOTIVE PERCEPTION OF WARRANTY
Warranty may be thought as a customer dissatisfaction. It is a multibillion- dollar a year opportunity for organ izations. It is an opportunity to know about warranty, understand it, and apply its princi ples so that the organ-ization can have less warranty, which will result in improved profits.
Obviously, not every thing may have a warranty. For example, items that are generally excluded are:
• Normal wear and tear
• Maintenance (oil and fluid changes, tune- ups, wiper blades, oil or air filters, clutch linings, etc.)
• Damage from environment or weather
• Inappropriate or poor maintenance (incorrect fuel, fluids, etc.)
• Damage from accidents
Within the automotive industry one may come across two distinct types of warranty. However, even though they are very similar, they are occurring at dif fer ent times. The two types are:
1. Warranty cost: This is an estimate of the lifetime warranty obli-gation on a new vehicle at the time of sale. The cost estimate is accrued, increasing the warranty reserve, which of course affects com pany profits as reported in the income statement. Another way of thinking about warranty cost is to think of putting money in a savings account to pay for future warranty.
17Warranty, Six Sigma,
Lean, and FMEA
1123-77375_ch01_5P.indd 155 12/21/18 11:02 AM
156 Chapter Seventeen
2. Warranty spend: This is the actual occurrence of warranty obliga-tions as they are claimed. The warranty spend is paid, decreasing the warranty reserve, which of course affects com pany cash flow as reported on the balance sheet. Another way of thinking about it is to think of withdrawals from the savings account when bills are due.
Typical Terms Used in the Auto Industry
• Months in Ser vice (MIS) or Time in Ser vice (TIS): Both terms are interchangeable, indicating the number of months that the vehicle has been in the customer’s hands. This generally starts with the date of sale of the vehicle to the end- customer (not the time it was sitting on the dealership lot).
• Repair per 1000 Vehicles (R/1000): This refers to the number of repairs made on a vehicle divided by the number of vehicles made at a given TIS, multiplied by 1000.
• Cost per Unit (CPU): This financial metric gives the total cost of repairs for a given part, or parts, per the number of vehicles at a given TIS. It is approximately equal to R/1000 × Cost per repair.
• Analytical Warranty System (AWS): This is a general term that automakers use to gather and analyze and report warranty data. Dif fer ent companies may have dif fer ent names for it. In some organ-izations, one may find a very specific system that tracks warranty items and the history of the specific “fix” of a warranty item. For example, Ford Motor Com pany uses a system called Balanced Sin-gle Agenda for Quality (BSAQ), which quickly analyzes low time in ser vice warranty (usually less than three MIS) to action items.
Warranty Spend Reduction Tools
When you are in the pro cess of defining what works for spend reduction, you should consider methodologies and tools specific to your organ ization’s product. At the very least consider some of the following:
1. Claims pro cessing
2. Warranty counseling pro cess
3. Supplier recovery
4. Manufacturing
5. Early detection
1123-77375_ch01_5P.indd 156 12/21/18 11:02 AM
Warranty, Six Sigma, Lean, and FMEA 157
6. Special ser vice diagnostics/repair tool
7. Optimized diagnostic/repair procedure
8. Prior approval
9. Field ser vice actions
10. Communications
11. Remanufacturing
12. Ser vice part kit pricing
13. Ser vice part level
For maximum benefit consider developing a formal systemic pro cess(es) for recovery, such as:
• Warranty reduction program (WRP)
• Parts return analy sis and supplier chargeback
• Spike recovery pro cess
• Field ser vice action (FSA)
• Warranty special contracts
Perhaps the best warranty spend reduction is a methodology for early detection. This methodology may be specific to the organ ization or product, or it may be generic but pres ent opportunities to increase tool use, optimize tool use, and create additional tools to ensure identification of concerns as early as pos si ble. Some considerations:
• Maintain an overall awareness of how warranty lags other metrics.
• Ensure that warranty is taken into account while making sourcing evaluations/decisions.
• Use a Supplier Warranty Reporting (SWR) tool to familiarize your-self with all pertinent warranty per for mance.
• Review warranty improvement plans when on- site with appropriate personnel.
• Drive closure of open warranty issues and ensure that there is link-age to the supplier’s Quality Operating System (QOS).
• Seek input from the customer’s representative on your per for mance and ensure all issues are addressed systemically via the Manufac-turing Site Assessment.
1123-77375_ch01_5P.indd 157 12/21/18 11:02 AM
158 Chapter Seventeen
• Ask your customers or suppliers to provide examples of their warranty per for mance and actions they have taken to improve per for mance.
A typical starting point for developing a standardized warranty model is to start with a specific commodity and include at least three items:
1. Current warranty per for mance.
2. Commodity characteristic questions (e.g., difficulty to determine root cause, supplier design responsibility, and system interac-tion issues). Use P- diagram, interface diagram, and FMEAs; if mechanical item, use finite ele ment analy sis.
3. The ability to determine warranty per for mance targets at sourc-ing and a supplier’s cooperation level. The intent of this item is to develop a practical (doable) warranty reduction program.
SIX SIGMA
FMEA represents a technique aimed at averting future issues in proj ect pro cesses and eliminating risks that may hamper a solution. Therefore, it fits the Six Sigma methodology in identifying and evaluating defects that could potentially result in reducing the quality of a product. Defects within the methodology are defined as anything that reduces the speed or quality at which a product or ser vice is delivered to happy customers. While Six Sigma techniques are implemented to discover and reduce the variables in pro cesses which cause nonrandom fluctuations, FMEA is used to discover and prioritize aspects of the pro cess that demand improvement and also to statistically analyze the success of a preemptive solution.
In the Define- Measure- Analyze- Improve- Control (DMAIC) model we use the FMEA in the mea sure and control phases. In the mea sure phase we make sure that a closer look at the causes and customer impacts of potential pro cess/product failure modes are considered and addressed. In the control phase we make sure that the appropriate controls are implemented so that failures can be stopped before they reach the customer.
In the Design for Six Sigma (DFSS), FMEA is used to anticipate prob-lems in design, pro cesses, and products in order to reduce costly and embar-rassing risks. In other words, the FMEA is used to find areas that need design/pro cess improvement and to mea sure the success of the later fix. Specifically, it is used in the optimize phase of the Define- Characterize- Optimize- Verify (DCOV) model.
1123-77375_ch01_5P.indd 158 12/21/18 11:02 AM
Warranty, Six Sigma, Lean, and FMEA 159
LEAN
By definition lean methodology is a method of identifying and removing waste. Waste is defined as variation. Therefore, the FMEA is used in a vari-ety of ways to identify failures and remove those failures from the system or pro cess under consideration. Specifically, the FMEA in a lean environment acts as a proactive tool to prevent the solutions from going wrong.
AFTER IMPROVEMENTS ARE MADE
Whether the FMEA is used under the ISO, IATF 16949, Six Sigma, or lean criteria, when completed it is reviewed with the intention of making sure (1) all failures have been identified, (2) actions have been recommended for the causes, and (3) appropriate strategies have been identified to control and monitor these strategies to prevent recurrence.
In some cases lists are also made, providing step- by- step guidelines to follow when evaluating pro cess maps and CPs as a result of the PFMEA. These guidelines may include the following:
• Key pro cess stages
• Probable failure modes for each stage
• Effects of this failure mode
• Severity of effect on a scale of 1–10
• Identification of failure mode causes
• Identification of controls being used to detect prob lems
• Statistical analy sis of data collected
• Allocation of necessary actions to responsible individuals
• Reevaluation of the design and/or pro cess
Concerns
One of the biggest complaints about conducting an FMEA when it is used to assess a design or a pro cess is that it is stored away after completion and no longer referred to when additional prob lems occur in the course of devel-oping supplementary proj ects. In other words, most of the time it ceases to be a dynamic (living) and useful document. Furthermore, the complaint
1123-77375_ch01_5P.indd 159 12/21/18 11:02 AM
160 Chapter Seventeen
that it is too time- consuming discourages management to be serious about conducting an FMEA as it should be.
To be sure, the FMEA is considered both useful and dynamic in all areas of improvement in quality initiatives such as total quality manage-ment (TQM), ISO, Six Sigma, and lean methodologies and provides valu-able information that contains beneficial implications regarding the design or pro cess of the product. This explicit benefit should be further utilized when similar issues occur in future proj ects, which could save time, money, and useless expenditures of energy and manpower.
Indeed, the FMEA and all its derivatives take time. However, if the organ ization is committed to continual improvement and reduction in waste, all FMEAs will contribute to this improvement. For this to happen, there are three fundamentals that must be followed:
1. Evolve and adopt. Use the TGW to improve, but do not forget the things gone right (TGR). Both are impor tant. TGW focuses on past failures and lessons learned, but the TGR should be empha-sized to remind you of the good things that happened and need to be repeated.
2. Nourish and grow. Since the FMEA is a living document, it must be remembered always that it takes encouragement to identify prob lems without the fear of intimidation. It is appropriate to recall the words of Henry Ford: “Coming together is a beginning. Keeping together is a pro cess. Working together is a success.”
3. Devote time. Winemaker Paul Masson once famously advertised that “We will sell no wine before its time.” The implication is that it takes as much time as it is necessary to develop a wine’s character and quality: its color, taste, and bouquet. Storage and aging in oak barrels are very impor tant. Consequently, the price is adjusted to reflect these par ameters for higher quality. If it is good enough for a winery it should be good enough for every one concerned with qual-ity and reduction of risk to appropriate enough time to have the opti-mum result. Certainly, we cannot know or control all the unknowns, but we can identify and mitigate the risk with some level of success by building a strong foundation to respond to potential failures. That response is the methodology of an FMEA and its derivatives.
1123-77375_ch01_5P.indd 160 12/21/18 11:02 AM
161
References
Dorfman, M. 2007. Introduction to Risk Management and Insurance. 9th ed. Englewood Cliffs, NJ: Prentice Hall.
Kokcharov, I. 2015. “What Is Proj ect Management?” June 2. https:// www . slideshare . net / igorkokcharov / what - is - project - risk - management.
PMBOK (Proj ect Management Body of Knowledge). 2000, 2017. PMBOK Guide. Newton, PA: Proj ect Management Institute.
Roehrig, P. 2006. “Bet on Governance to Manage Outsourcing Risk.” Business Trends Quarterly. January. http:// www . objectivasoftware . com / docs / Business TrendsQuarterly _ Jan _ 2007 . pdf.
Stamatis, D. 1998. Advanced Quality Planning: A Commonsense Guide to AQP and APQP. New York: Quality Resources.
— — —. 2016. Quality Assurance: Applying Methodologies for Launching New Products, Ser vices, and Customer Satisfaction. Boca Raton, FL: CRC Press.
1123-77375_ch01_5P.indd 161 12/21/18 11:02 AM
1123-77375_ch01_5P.indd 162 12/21/18 11:02 AM
163
AD&P. 2013. “Quality Tools: Digital and Physical.” Automotive Design and Pro-duction (September), 40.
AIAG (Automotive Industry Action Group). 1914, 2008. Advanced Product Qual-ity Planning and Control Plan: APQP. 1st and 2nd ed. Southfield, MI: Chrysler, Ford Motor, General Motors.
— — —. 2001, 2008. Potential Failure Mode and Effects Analy sis: FMEA. 3rd and 4th ed. Southfield, MI: Chrysler, Ford Motor, General Motors.
— — —. 2005. Statistical Pro cess Control: SPC. 2nd ed. Southfield, MI: Chrysler, Ford Motor, General Motors.
— — —. 2009. Production Part Approval Pro cess: PPAP. 4th ed. Southfield, MI: Chrysler, Ford Motor, General Motors.
— — —. 2010. Mea sure ment Systems Analy sis: MSA. 4th ed. Southfield, MI: Chrysler, Ford Motor, General Motors.
AIAG/VDA (Automotive Industry Action Group/Verband der Automobilindust-rie). 2017. Failure Mode and Effect Analy sis— FMEA: Design FMEA and Pro-cess FMEA Handbook. 1st ed. Southfield, MI: AIAG.
Carlson, Carl S. 2012. Effective FMEAs: Achieving Safe, Reliable, and Eco nom ical Products and Pro cesses Using Failure Mode and Effects Analy sis. Hoboken, NJ: John Wiley & Sons.
Chadha, R. 2013. “Dig Deeper: Deploy a 5S Blitz to Create a High- Performance Work Environment.” ASQ Quality Pro gress (August), 42–49.
IATF (International Automotive Task Force). IATF 16949:2016. Quality Manage-ment System for Organ izations in the Automotive Industry.
IEC (International Electrotechnical Commission). 1985. Analy sis Techniques for Sys-tem Reliability: Procedure for Failure Mode and Effects Analy sis. Geneva: IEC.
— — —. IEC 60812:2006. Analy sis Techniques for System Reliability— Procedure for Failure Mode and Effects Analy sis (FMEA). 2nd ed. Geneva: IEC.
— — —. IEC 60601-1-2:2007. Medical Electrical Equipment— Part 1–2: Gen-eral Requirements for Basic Safety and Essential Performance— Collateral Standard: Electromagnetic Compatibility— Requirements and Tests. 3rd ed. Geneva: IEC.
Selected Bibliography
1123-77375_ch01_5P.indd 163 12/21/18 11:02 AM
164 Selected Bibliography
ISO (International Organ ization for Standardization). ISO 13485:2003. Medical Devices— Quality Management System. Geneva: ISO.
— — —. EN ISO 14971:2012. Medical Devices— Application of Risk Management to Medical Eevices. Geneva: ISO.
— — —. ISO 14001:2015. Environmental Management Systems. Geneva: ISO.— — —. ISO 9001:2015. Quality Management System. Geneva: ISO.— — —. ISO/DIS 31000:2018. Risk Management— Princi ples and Guidelines on
Implementation. Geneva: ISO.ISO/TS 16949: 2009. Quality Management Systems— Technical Specification.
Southfield, MI: ANFIA, FIEV, SMMT, VDA, Chrysler, Ford Motor, General Motors, PSA Peugeot Citroen.
Kececioglu, Dimitri. 1991. Reliability Engineering Handbook Volume 2. Engle-wood Cliffs, NJ: Prentice- Hall.
McDermott, Robin E., Raymond J. Mikulak, and Michael R. Beauregard. 1996. The Basics of FMEA. New York: Productivity Press.
SAE (Society of Automotive Engineers). 2000. Aerospace Recommended Practice ARP5580: Recommended Failure Modes and Effects Analy sis (FMEA) Prac-tices for Non- Automobile Applications. June. Warrendale, PA: SAE.
— — —. 2000. Surface Vehicle Recommended Practice J1739: (R) Potential Fail-ure Mode and Effects Analy sis in Design (Design FMEA), Potential Failure Mode and Effects Analy sis in Manufacturing and Assembly Pro cesses (Pro-cess FMEA), and Potential Failure Mode and Effects Analy sis for Machinery (Machinery FMEA). June. Warrendale, PA: SAE.
Spath, P. L. 2003. “Using Failure Mode and Effects Analy sis to Improve Patient Safety.” AORN Journal 78 (1), 16–37.
Stamatis, D. H. 2003. Failure Mode and Effect Analy sis: FMEA from Theory to Execution. Milwaukee, WI: ASQ Quality Press.
— — —. 2014. Introduction to Risk and Failures: Tools and Methodologies. Boca Raton, FL: CRC Press.
The Joint Commission. n.d. Joint Commission Requirements. http:// www . jointcommission . org / standards _ information / tjc _ requirements . aspx.
Topel, S. 2013. “A Total Transformation.” Quirk’s Marketing Research Review (October), 54–55.
US Department of Defense. 1980. MIL- STD 1629. Washington, DC: Department of Defense.
— — —. 1980. MIL- STD-1629A: Procedures for Performing a Failure Mode Effects and Criticality Analy sis. Washington, DC: DOD.
US Department of the Army. 2006. “Failure Modes, Effects, and Criticality Analy sis (FMECA) for Command, Control, Communications, Computer, Intelligence, Sur-veillance, and Reconnaissance (C41SR) Facilities.” TM-5-698-4. September 29.
1123-77375_ch01_5P.indd 164 12/21/18 11:02 AM
165
Index
AACAT (avoid-control-accept-transfer)
model, 17actions taken
definition and description of, 42–43
FMECA, 100form detailing, 81–82HFMEA, 90rankings as basis for, 54–55records of, 55timing of, 55troubleshooting, 145
aerospace industry FMEA, 152affinity diagrams, 75, 94, 112, 123–24AIAG. See Automotive Industry
Action Groupalpha, 103–4analytical warranty system (AWS),
156area charts, 53, 53fautomotive industry
dynamic control plans, 114FMEA adoption, 151FMEA form, 12, 13fgoal of QMS, 6IATF 16949 standard, 1, 6, 7rankings, 50SAE standards, 7, 50, 152VDA standard, 7–10, 11–12, 13fwarranty, 155–58
Automotive Industry Action Group (AIAG)
FMEA form, 12FMEA guidelines, 151risk assessment criteria, 11–12tools, IATF 16949 supporting, 7VDA standard approval, 8
avoid-control-accept-transfer (ACAT) model, 17
AWS (analytical warranty system), 156
Bbar chart graphs, 132basics of FMEA
benefits, 31, 36definition, 29–30need for, 30–31post-completion steps, 37–38process of conducting, 32–36
advantages of, 36starting, 34, 34f–35fsteps in, 32–34timing of, 35uses of, 35
understanding customers and their needs, 36–37
vocabulary, 38–43benchmarking, 60, 64, 66, 73, 119,
120
Note: Page numbers followed by f or t refer to figures or tables, respectively.
1123-77375_ch02_5P.indd 165 12/21/18 11:01 AM
166 Index
beta, 103block diagrams, 32, 46f, 66, 82boundary diagrams, 45, 46f, 46t, 120box plots, 95, 124BPEST (business, political,
economic, social, technological) analysis, 1
brainstormingcause determination, 41data analysis, 33description and use of, 124DFMEA, 64, 66EFMEA, 82FMEA team, 32, 33FMECA, 112HFMEA, 94PFMEA, 73
breadboard tests, 60business, political, economic, social,
technological (BPEST) analysis, 1
business continuity planning, 1
Ccapability studies, 38cause-and-effect analysis
data analysis, 33description and use of, 125DFMEA, 66EFMEA, 83fishbone diagrams, 41, 125FMECA, 112PFMEA, 75
causes, possibledefinition and description of,
40–41determination of, 41form detailing, 80root-cause analysis, 86–87, 87t,
101root cause of risk, 12, 52troubleshooting, 143
c-control charts, 71, 72t, 125–26CFMEA. See concept FMEAchanges
actions taken eliciting, 55to control plans, 116effective implementation of, 6
characteristicsclassification and ranking based
on, 52PFMEA assessing, 58t, 74special, control plans monitoring,
113types of FMEAs assessing, 58t
check sheets/checklists, 46f, 71, 72t, 94, 112
chemical industry FMEA, 152–53classification
definition and description of, 40form detailing, 80ranking of characteristics for, 52severity, 40, 52, 97, 103, 105–6troubleshooting, 143
common types of FMEA. See types of FMEA
computer simulation, 33, 60, 112, 125concept FMEA (CFMEA)
benefits of, 59–60characteristics assessed, 58tcontrol plan linkage with, 120, 121focus of, 58tform, 60industry-specific, 151objective of, 58tpurpose of, 59rankings, 60tools, 60use of, 59
continual improvement goals, 33–34, 38
control chart-c, 71, 72t, 125–26control chart-median and R, 126control chart-np, 71, 126control chart-p, 71, 72t, 127control chart-u, 71, 127control chart X-bar and R, 71, 72t,
127–28control chart X-bar and S, 128control plans
adherence to, 38benefits of, 114content of, 114–15deficiencies in typical, 116–17definition of, 113existence of, 38FMEA linkage with, 115–17,
119–22
1123-77375_ch02_5P.indd 166 12/21/18 11:01 AM
Index 167
purpose of, 113reaction plans and, 113, 115risk management, 16–18statement of applicability, 16tools for, 117types of, 114use of, 113–14
corporate knowledge, 60, 119, 120correlation analysis, 95cost per unit (CPU), 156critical path analysis, 135cross-functional process maps, 128customer needs and requirements
control plans to meet, 113, 114risk management project
addressing, 6understanding, 36–37
Ddata collection and analysis, 33DCOV (Define-Characterize-
Optimize-Verify) model, 158
decision making under conditions of risk and uncertainty, 2
decision trees, 95, 128–29Define-Characterize-Optimize-Verify
(DCOV) model, 158Define-Measure-Analyze-Improve-
Control (DMAIC) model, 158
Department of Defense, avoid-control-accept-transfer model, 17
designcurrent, form detailing, 81design concept input, 119design concept output, 120design input, 120design output, 121for reliability, 21for Six Sigma, 158
design FMEA (DFMEA)benefits of, 61–62characteristics assessed, 58tcontrol plan linkage with, 116,
116f, 119–22focus of, 58tform, 49, 50f, 62
industry-specific, 151objective of, 58toverview, 34fpurpose of, 61rankings, 51, 62–66, 63t–65trisk assessment criteria, 11risk reduction strategies, 62,
64, 66robustness, 46ttools for, 32, 62, 64, 66troubleshooting, 143–44use of, 61VDA standards, 9–10
design of experiments (DOE), 33, 83, 129
detection of riskactions taken affecting, 55definition and description of,
42, 97detection controls, 42, 111early, for warranty spend
reduction, 157–58FMEA results, 33, 38form detailing, 81, 111new, 43, 55rating or ranking, 51–52, 64, 65t,
66, 71t, 73–74, 94, 111reduction of, 52risk assessment for, 11–12risk priority number based on, 42,
53, 111–12troubleshooting, 144
DFMEA. See design FMEADMAIC (Define-Measure-Analyze-
Improve-Control) model, 158
DOE (design of experiments), 33, 83, 129
dot plots, 129–30
Eeffects of failure
definition and description of, 39–40, 97
form detailing potential, 80, 110–11
troubleshooting, 142EFMEA. See equipment FMEA
1123-77375_ch02_5P.indd 167 12/21/18 11:01 AM
168 Index
8 wastes analysis, 87, 88tend-of-the-line (EOL) testing, 111engineering testing, 75equipment FMEA (EFMEA)
benefits of, 76characteristics assessed, 58tcompletion of, 77discarding of, 77FMEA team, 76–77, 78focus of, 58tform, 77–82, 79fgeneral information, 76–77method of preparing, 76objective of, 58tpurpose of, 75rankings, 82timing of starting/updating, 76, 77tools for, 82–83updates to, 77use of, 75
escape point, 12, 90event tree analysis, 1
Ffacilitators, team, 24–25failure
causes of, possible, 12, 40–41, 52, 80, 86–87, 87t, 101, 143
criteria for, 20data on, 33definition of, 20, 27detection of, 97, 111 (See also
detection of risk)effects of, 39–40, 80, 97, 110–11, 142FMECA quantitative analysis of
probability of, 105function, 97mean time between, 97, 106–7mindset of minimizing, 27–28mode of, 39, 78, 80, 97, 104–6,
108, 110, 142rate of, 104, 107–8types of, 20–21understanding concept of, 20–21
failure modedefinition and description of, 39, 97FMECA quantitative analysis of,
104–6
form detailing potential, 78, 80, 108, 110
troubleshooting, 142Failure Mode Effects and Criticality
Analysis. See FMECA (Failure Mode Effects and Criticality Analysis)
Failure Mode and Effect Analysis. See FMEA (Failure Mode and Effect Analysis)
fault tree analysis (FTA)cause determination, 41description and use of, 130–31DFMEA, 62EFMEA, 82FMECA analysis and, 101, 103risk assessment, 1
fishbone diagrams, 41, 1255 Whys method, 41FMEA (Failure Mode and Effect
Analysis)advantages or benefits of, 31, 36basics of, 29–43challenges of, 58concept FMEA, 58t, 59–60, 120,
121, 151control plans, 38, 113–17, 119–22customer needs and requirements,
36–37definition of, 1, 29–30, 130design FMEA (See design FMEA)equipment FMEA, 58t, 75–83FMECA extension of, 97–112,
152form, 12, 13f, 49, 50f, 60, 62, 67,
68f, 77–82, 79f, 91, 92fhealth FMEA, 85–95industry-specific, 49, 50, 57–58,
59, 62, 151–53 (See also specific industries)
institutionalization of, 149lean methodology, 159linkages, 46t, 115–17, 119–22monitoring and system response, 12need for, 30–31post-completion steps, 37–38,
159–60prerequisites of, 23–28problems and concerns, 147–49,
159–60
1123-77375_ch02_5P.indd 168 12/21/18 11:01 AM
Index 169
process FMEA (See process FMEA)
process of conducting, 32–36quality management system, 1–6,
15–16, 155–60rankings (See rankings)reliability and, 19–21, 29, 51,
120–21risk and (See risk)robustness of, 45–48, 121service FMEA, 32, 58tSix Sigma, 158starting or beginning, 34, 34f–35fsystem FMEA, 32teams for, 23–27, 32, 33, 76–77, 78,
89, 147timing of, 35tools (See tools)troubleshooting, 141–45types of, 57–83 (See also specific
types)uses for, 35, 130VDA standard, 7–10, 11–12, 13fvocabulary for, 38–43warranty, 1, 155–58
FMEA teamsbrainstorming by, 32, 33champions/sponsors of, 24common problems, 147considerations for, 25core, 23–24, 78creating effective, 23EFMEA, 76–77, 78facilitators for, 24–25HFMEA, 89leaders of, 24motivated, 25–26potential members of, 26–27recorders for, 24selection of, 32structure of, 23–25useful information for, 25–26
FMECA (Failure Mode Effects and Criticality Analysis)
analysis of significant function, 99–108
qualitative, 101, 102fquantitative, 101, 103–8, 109f
benefits of, 112definition of, 97–98
detection of failure, 97, 111effect of failure identification, 97,
110–11failure mode analysis and
identification, 97, 104–6, 108, 110
failure rate analysis, 104, 107–8FMEA distinction from, 100form, 101, 102f, 108, 109f, 110–12implementation of results, 100industry-specific, 152planning and preparation for, 99process of conducting, 99–101rankings, 101risk priority number calculation,
111–12severity classification, 97, 103,
105–6significant function analysis in,
100–108qualitative, 101, 102fquantitative, 101, 103–8, 109f
sources for identifying functions, 98–99
sustaining, 100tools for, 112vocabulary, 97
force field analysis, 60form
CFMEA, 60DFMEA, 49, 50f, 62EFMEA, 77–82, 79fFMECA, 101, 102f, 108, 109f,
110–12generic, 49, 50fHFMEA, 91, 92fPFMEA, 49, 50f, 67, 68fVDA standard compliance, 12, 13f
FTA. See fault tree analysisfunction
definition and description of, 39, 97
form detailing, 78significant, FMECA analysis of,
100–108sources for identifying, 98–99troubleshooting, 142
functional diagrams, 32, 60function failure, 97function trees, 66, 131
1123-77375_ch02_5P.indd 169 12/21/18 11:01 AM
170 Index
Ggage repeatability and reproducibility,
131gages or indicators, 111Gantt chart graphs, 132graphs-bar chart, 132graphs-Gantt chart, 132graphs-pie chart, 132
Hhazard analysis, 86–87health FMEA (HFMEA)
conducting analysis in, 90defining topic of, 88–898 wastes analysis, 87, 88tFMEA team, 89form, 91, 92fgraphical description of process,
89–90hazard analysis, 86–87identifying actions and outcome
measures for, 90organizational, 88–89, 89toverview, 85–86process of conducting, 87–91process-oriented, 88–89, 89trankings, 91, 93t–94t, 94repetition of, 91root-cause analysis comparison,
86–87, 87t6S goals for, 87, 88ttools for, 89–90, 94–95
histograms, 133
IIATF (International Automotive Task
Force) 16949 standard, 1, 6, 7
identification of risks, 12, 14, 37. See also detection of risk
industry-specific FMEAaerospace, 152automotive, 151 (See also
automotive industry)
chemical/pharmaceutical, 57–58, 152–53
rankings, 49, 50, 62software, 152types of, 57–58, 59
inspections, 75, 117institutionalization of FMEA,
149insurance, 17interface matrix, 46, 46t, 47t, 66,
82, 120internal past corporate knowledge,
60, 119, 120International Automotive Task Force
(IATF) 16949 standard, 1, 6, 7
ISO (International Organization for Standardization) standards
9000, 59001:2015, 3–5, 6–713485:2016, 214001:2015, 314971:2012, 227001, 1631000:2018, 4, 7, 15goal of, 5–6risk assessment defined, 2risk-based thinking, 2–5, 6–7,
15, 16risk defined, 1, 2, 5risk management defined, 2system-oriented, 6
Jjigs, 117The Joint Commission (TJC),
85
KKano model, 37, 133Kepner Tregoe method, 41key process input variables (KPIV),
113key process output variables (KPOV),
113
1123-77375_ch02_5P.indd 170 12/21/18 11:01 AM
Index 171
Llaboratory tests, 60leaders, team, 24lean methodology, 159linkages
design concept input, 119design concept output, 120design input, 120design output, 121FMEA-control plan, 115–17,
119–22machinery output, 122process concept input, 120process concept output, 120process input, 121process output, 122robustness, 46t, 121
logistical analysis, 101
Mmachinery output, 122mathematical modeling, 33, 60mean time between failure (MTBF),
97, 106–7metrics
FMEA, 33HFMEA, 90measurement system analysis, 117measures of central tendency and
dispersion, 2Mil-Std 1629, 91, 103MIS (months in service), 156mistake proofing, 71, 73, 75, 111monitoring and system response (MSR)
control plans for, 113–14FMEA establishment of, 12risk-based effectiveness checks, 15
months in service (MIS), 156MTBF (mean time between failure),
97, 106–7
Nnew detection of risk, 43, 55new occurrences of risk, 43, 55
new risk priority number, 43, 55, 82
new severity of risk, 43, 55np-control charts, 71, 126
Ooccurrences of risk
actions taken affecting, 55classification based on, 52definition and description of,
41–42, 97FMEA results, 33, 38form detailing, 80–81new, 43, 55rating or ranking, 51, 62, 64, 64t,
70t, 73, 91, 94t, 101reduction of, 51risk assessment of, 11risk priority number based on,
42, 53, 111–12troubleshooting, 143–44
operational definitions, 134opportunity, risk and, 2, 3, 4–5, 6
PPareto diagrams, 134p-control charts, 71, 72t, 127P-diagrams
DFMEA, 66EFMEA, 82linkages, 120, 121robustness tool, 46t, 47–48, 48f
PERT analysis, 135PESTLE (political, economic,
social, technical, legal, environmental) analysis, 2
PFMEA. See process FMEApharmaceutical industry FMEA,
57–58, 152–53pie chart graphs, 132Poisson distribution, 83,
105political, economic, social,
technical, legal, environmental (PESTLE) analysis, 2
1123-77375_ch02_5P.indd 171 12/21/18 11:01 AM
172 Index
prerequisites of FMEAFMEA teams
creating effective, 23motivated, 25–26potential members of, 26–27structure of, 23–25
mindset of minimizing failures, 27–28
prevention controls, 42, 81, 144prioritization, FMEA, 32problems and concerns
institutionalization, 149post-completion evolution,
159–60problem analysis, in risk
assessment, 14problem-solving process, 41, 123procedural, 147–48team, 147
procedural problems, common, 147–48
process concept input, 120process concept output, 120process control. See also control
plansDFMEA, 66nonstatistical, 71–72, 72tPFMEA, 70–72, 72t, 74–75statistical, 33, 71, 72t, 94, 117
process flowcharts, 32, 89–90, 94, 117, 119–22, 135
process FMEA (PFMEA)benefits of, 67characteristics assessed, 58t, 74control plan linkage with, 116,
116f, 119–22focus of, 58tform, 49, 50f, 67, 68findustry-specific, 151objective of, 58toverview, 35fprocess control matrix, 70–72, 72tprocess control system guidelines,
74–75purpose of, 66–67rankings, 51, 52, 69–70, 69t–71t,
73–74recommendations, 74risk assessment criteria, 11–12risk reduction strategies, 73–74
tools for, 32, 73, 75troubleshooting, 142use of, 67VDA standards, 10
process input, 121process output, 122program decision process charts,
135–36Pugh technique, 64, 66, 73, 74, 136
QQFD (quality function deployment),
33, 37, 60, 62, 66, 136–37QMS. See quality management systemqualitative analysis, FMECA, 101, 102fquality function deployment (QFD),
33, 37, 60, 62, 66, 136–37quality management system (QMS)
goal of, 5–6lean methodology, 159post-FMEA implementation,
159–60risk-based approach, 2, 3–5, 15–16Six Sigma, 158warranty in, 1, 155–58
quantitative analysis, FMECA, 101, 103–8, 109f
Rrankings
CFMEA, 60classification and characteristics,
52detection rating, 51–52, 64, 65t, 66,
71t, 73–74, 94, 111DFMEA, 51, 62–66, 63t–65tdriving action plans, 54–55EFMEA, 82FMECA, 101HFMEA, 91, 93t–94t, 94industry-specific, 49, 50, 62not standardized, 49occurrence rating, 51, 62, 64, 64t,
70t, 73, 91, 94t, 101PFMEA, 51, 52, 69–70, 69t–71t,
73–74
1123-77375_ch02_5P.indd 172 12/21/18 11:01 AM
Index 173
risk understanding and calculation, 53–54
severity rating, 50, 62, 63t, 64, 69t, 73, 91, 93t–94t, 94, 101
troubleshooting, 145RCM (reliability-centered
maintenance), 103reaction plans, 113, 115real option modeling, 1recommendations
definition and description of, 42form detailing, 81linkages output, 122PFMEA, 74troubleshooting, 145
recordkeeping, 24, 33, 55regression analysis, 137reliability
analysis/modeling, 33, 66, 83, 111, 112, 137
design for, 21expertise in, for occurrence
ratings, 51FMEA and, 19–21, 29, 51, 120–21linkages design, 120–21overview, 19–20reliability-centered maintenance
(RCM), 103understanding failure, 20–21
repair per 1000 vehicles (R/1000), 156
resource planning, 6risk
assessmentcost-benefit analysis, 14, 15defined, 2identification of risk options,
14–15method or process of
conducting, 10–14techniques, 1–2 (See also
FMEA)avoidance of, 16definition of, 1, 2, 5detection of (See detection of risk)escape point from, 12, 90goal of standards for, 5–6high-risk area identification, 37IATF 16949 standard, 1, 6, 7identification of, 12, 14, 37
insurance against, 17ISO standards, 1–7, 15, 16management
defined, 2plans, 16–18
occurrences of (See occurrences of risk)
opportunity and, 2, 3, 4–5, 6overview, 1–3problem analysis, 14project plan for, 6–7, 15, 16–18QMS addressing, 2, 3–5, 15–16ranking based on understanding
and calculating, 53–54reduction of, 17, 62, 64, 66,
73–74retention of, 17risk-based thinking, 2–5, 6–7, 15,
16, 18risk priority number (See risk
priority number)risk register, 3–4, 15, 16trisk treatment plans, 16–18root cause of, 12, 52severity of (See severity of risk)sharing or transfer of, 17source analysis, 14uncertainty and, 1–3, 5, 18VDA standard, 7–10, 11–12, 13f
risk priority number (RPN)actions taken affecting, 55area chart of, 53, 53fcalculation of, 42, 53–54, 111–12classification based on, 52definition and description of, 42FMEA results, 33, 37–38form including, 81, 82, 111–12new, 43, 55, 82rankings in relation to, 53–54troubleshooting, 144
robustnessboundary diagrams, 45, 46f, 46tinterface matrix, 46, 46t, 47tlinkages process, 46t, 121P-diagrams, 46t, 47–48, 48f
root causeanalysis, 86–87, 87t, 101of risk, 12, 52
RPN. See risk priority numberrun charts, 137–38
1123-77375_ch02_5P.indd 173 12/21/18 11:01 AM
174 Index
SSAE. See Society of Automotive
Engineerssampling, 117scatter plots/diagrams, 95, 138scenario, FMECA, 112self-insurance, 17service FMEA, 32, 58tseverity of risk
actions taken affecting, 55classification based on, 40, 52, 97,
103, 105–6definition and description of, 40FMEA results, 33, 37–38form detailing, 80new, 43, 55rating or ranking, 50, 62, 63t, 64,
69t, 73, 91, 93t–94t, 94, 101reduction of, 50risk assessment of, 11risk priority number based on, 42,
53–54, 111–12troubleshooting, 143
shared and interlocking objectives matrix, 138
significant function, 100–108simulation, 33, 60, 112, 1256S (sort, set in order, shine/sweep,
safety, standardize, sustain) goals, 87, 88t
Six Sigma, 158Society of Automotive Engineers (SAE)
ARP5580 standard, 152J1739 standard, 7, 50, 152
software FMEA, 152source analysis, 14, 98–99SPC (statistical process control), 33,
71, 72t, 94, 117special characteristics, 113statement of applicability, 16statistical inference, 2statistical process control (SPC), 33,
71, 72t, 94, 117stem and leaf plots, 138surveys, 139SWOT (strengths-weaknesses-
opportunities-threats) analysis, 1
system FMEA, 32
Tteams. See FMEA teamsterminology. See vocabularytheory of inventive problem solving
(TRIZ), 64, 66, 73time in service (TIS), 156time series forecasting, 139timing
of actions, review of, 55EFMEA start/update, 76, 77FMEA, 35mean time between failure, 97,
106–7risk management project plan, 6severity variance based on, 50
TIS (time in service), 156TJC (The Joint Commission), 85tools
affinity diagrams, 75, 94, 112, 123–24
area charts, 53, 53fbenchmarking, 60, 64, 66, 73, 119,
120block diagrams, 32, 46f, 66, 82boundary diagrams, 45, 46f, 46t, 120box plots, 95, 124BPEST analysis, 1brainstorming, 32, 33, 41, 64, 66,
73, 82, 94, 112, 124breadboard tests, 60business continuity planning, 1cause-and-effect analysis, 33, 41,
66, 75, 83, 112, 125CFMEA, 60check sheets/checklists, 46f, 71,
72t, 94, 112control chart-c, 71, 72t, 125–26control chart-median and R, 126control chart-np, 71, 126control chart-p, 71, 72t, 127control chart-u, 71, 127control chart X-bar and R, 71, 72t,
127–28control chart X-bar and S, 128control plan, 117correlation analysis, 95critical path analysis, 135cross-functional process maps, 128data analysis, 33
1123-77375_ch02_5P.indd 174 12/21/18 11:01 AM
Index 175
decision making under conditions of risk and uncertainty, 2
decision trees, 95, 128–29design of experiments, 33, 83, 129detection controls, 42, 111DFMEA, 32, 62, 64, 66dot plots, 129–30EFMEA, 82–83end-of-the-line (EOL) testing, 111engineering testing, 75event tree analysis, 1fault tree analysis, 1, 41, 62, 82,
101, 103, 130–31fishbone diagrams, 41, 125flexible risk assessment, 15FMECA, 112force field analysis, 60functional diagrams, 32, 60function trees, 66, 131gage repeatability and
reproducibility, 131gages or indicators, 111graphs-bar chart, 132graphs-Gantt chart, 132graphs-pie chart, 132HFMEA, 89–90, 94–95histograms, 133inspections, 75, 117interface matrix, 46, 46t, 47t, 66,
82, 120internal past corporate knowledge,
60, 119, 120jigs, 117Kano model, 37, 133laboratory tests, 60logistical analysis, 101mathematical modeling, 33, 60measurement system analysis,
117measures of central tendency and
dispersion, 2mistake proofing, 71, 73, 75, 111operational definitions, 134Pareto diagrams, 134P-diagrams, 46t, 47–48, 48f, 66,
82, 120, 121PERT analysis, 135PESTLE analysis, 2PFMEA, 32, 73, 75Poisson distribution, 83, 105
process control, 33, 66, 70–72, 72t, 74–75, 94, 117
process flowcharts, 32, 89–90, 94, 117, 119–22, 135
program decision process charts, 135–36
Pugh technique, 64, 66, 73, 74, 136
quality function deployment, 33, 37, 60, 62, 66, 136–37
real option modeling, 1regression analysis, 137reliability analysis/modeling, 33,
66, 83, 111, 112, 137reliability-centered maintenance, 103robustness, 45–48root cause analysis, 86–87, 87t, 101run charts, 137–38sampling, 117scatter plots/diagrams, 95, 138scenario, 112shared and interlocking objectives
matrix, 138simulation, 33, 60, 112, 125statistical inference, 2statistical process control, 33, 71,
72t, 94, 117stem and leaf plots, 138surveys, 139SWOT analysis, 1time series forecasting, 139tree analysis, 139–40TRIZ, 64, 66, 73visual alarms, 111warranty analysis, 1, 156warranty spend reduction, 156–58whisker plots, 124
tree analysis, 139–40. See also decision trees; event tree analysis; fault tree analysis; function trees
TRIZ (theory of inventive problem solving), 64, 66, 73
troubleshooting an FMEAactions taken/revised ratings, 145appropriate controls applied, 144classification, 143detection, 144function/purpose, 142header, 141–42
1123-77375_ch02_5P.indd 175 12/21/18 11:01 AM
176 Index
occurrences, 143–44potential causes/mechanisms of
failure, 143potential failure effects, 142potential failure mode, 142prevention controls, 144recommended action, 145responsibility/target completion
date, 145risk priority number, 144severity, 143steps in, 141
types of FMEA. See also specific typeschallenges for, 58characteristics assessed, 58tcommon, 59–83concept, 58t, 59–60design, 58t, 61–66equipment, 58t, 75–83focus of, 58tindustry-specific, 57–58, 59objective of, 58tprocess, 58t, 66–75service, 58t
Uu-control charts, 71, 127uncertainty, risk and, 1–3, 5, 18
VVDA (Verband der
Automobilindustrie) standardaction priority, 10benefits of, 8flow, 9–10FMEA form in compliance with,
12, 13fprinciples of, 7–10risk assessment criteria, 11–12six-step model, 9
visual alarms, 111
vocabularyFMEA key terms, 38–43FMECA key terms, 97actions taken, 42–43causes, possible, 40–41classification, 40detection of risk, 42effect of failure, 39–40, 97failure detection, 97failure mode, 39, 97function, 39, 97function failure, 97mean time between failure, 97new detection of risk, 43new occurrences of risk, 43new risk priority number, 43new severity of risk, 43occurrences of risk, 41–42, 97prevention controls, 42recommendations, 42severity of risk, 40, 97warranty-related, 156
Wwarranty
automotive industry, 155–58cost, 155definition of, 155early detection methodology,
157–58spend, 156spend reduction tools, 156–58standardized model, 158terminology, 156warranty analysis, 1, 156
whisker plots, 124
XX-bar and R charts, 71, 72t,
127–28X-bar and S charts, 128
1123-77375_ch02_5P.indd 176 12/21/18 11:01 AM