Applied Reliability for Industry 1 E-Book
135,99 €
Mehr erfahren.
- Herausgeber: John Wiley & Sons
- Kategorie: Wissenschaft und neue Technologien
- Sprache: Englisch
Applied Reliability for Industry 1 illustrates the multidisciplinary state-of-the-art science of predictive reliability. Many experts are now convinced that reliability is not limited to statistical sciences. In fact, many different disciplines interact in order to bring a product to its highest possible level of reliability, made available through today's technologies, developments and production methods. These three books, of which this is the first, propose new methods for analyzing the lifecycle of a system, enabling us to record the development phases according to development time and levels of complexity for its integration. Predictive reliability, as particularly focused on in Applied Reliability for Industry 1, examines all the engineering activities used to estimate or predict the reliability performance of the final mechatronic system.
Sie lesen das E-Book in den Legimi-Apps auf:
Seitenzahl: 291
Veröffentlichungsjahr: 2023
Ähnliche
Table of Contents
Cover
Title Page
Copyright Page
Foreword
Preface
1 FIDES, a Method for Assessing and Building the Reliability of Electronic Systems
1.1. The inadequacy of existing methods
1.2. The ambition of FIDES
1.3. General presentation of the FIDES method
1.4. Validity of reliability studies with FIDES
1.5. Conclusion
1.6. References
2 Reliability in Maritime Transport: Choosing a Container Handling System
2.1. Introduction
2.2. Proposed case study
2.3. Inputs of the RAMS approach
2.4. Assessment of the system’s RAMS
2.5. Conclusion
2.6. General conclusion
2.7. References
3 Generation of a Failure Model through Probabilistic “
Stress–Strength”
Interaction in a Context of Poor Information
3.1. Introduction
3.2. Aims and objectives
3.3. Choosing types of legislation
3.4. Probability of failure
3.5. Safety factor
3.6. Validation and applications
3.7. Conclusion and extensions
3.8. References
4 Reliable Optimization of Dental Implants Using the Generalized Polynomial Chaos Method
4.1. Introduction
4.2. Stochastic approach
4.3. Deterministic design optimization
4.4. Reliability-based design optimization
4.5. Numerical result
4.6. Conclusion
4.7. References
5 Multi-objective Reliability Optimization Based on Substitution Models. Applied Case Study of a Hip Prosthesis
5.1. Introduction
5.2. Description of metamodeling methods
5.3. Optimization of multi-objective design
5.4. RBMDO based on hip prosthesis surrogate models
5.5. Conclusion
5.6. References
6 CMA-ES Assisted by the Kriging Metamodel for the Optimization of Thermomechanical Performances of Mechatronic Packaging
6.1. Introduction
6.2. Presentation of the system under study
6.3. Thermal fatigue models of solder joints
6.4. Modeling and finite element analysis of the PQFP housing
6.5. Evolutionary strategies
6.6. Global optimization of the PQFP housing solder joints
6.7. Conclusion
6.8. References
7 Reliable Optimization of Vibro-acoustic Problems in the Presence of Uncertainties via Polynomial Chaos
7.1. Introduction
7.2. Robust approaches to uncertainty propagation
7.3. Structural optimization
7.4. OSF method coupled with GPC applied to vibro-acoustic systems in the presence of uncertainties
7.5. Conclusion
7.6. References
List of Authors
Index
Summary of Volume 2
Summary of Volume 3
Other titles from ISTE in Mechanical Engineering and Solid Mechanics
End User License Agreement
List of Tables
Chapter 1
Table 1.1. Reference tests used to define baseline failure rates
Table 1.2. Example of a life profile for a bay computer of a medium-haul civi...
Table 1.3. Comparison of predicted MTBF MIL HDBK and FIDES versus MTBF REX
Chapter 2
Table 2.1. System mission profile
Table 2.2. Component reliability data
Table 2.3. Failure test results
Table 2.4. Strength distribution parameters for a 75% confidence level
Table 2.5. Summary of reliability results for a 75% confidence level
Table 2.6. Example of a risk criticality/acceptability table
Table 2.7. Distribution by mode of failure
Table 2.8. FMEA
Table 2.9. Modeling with Arbre Analyst V2.3.2
Table 2.10. Minimum bridge cuts with 1 service brake and type A carabiners
Table 2.11. Minimum bridge cuts with 1 service brake and type B carabiners
Table 2.12. FMEA supplement
Table 2.13. Minimum bridge cuts with one service brake and one safety brake a...
Table 2.14. Minimum bridge cuts with one service brake and one safety brake a...
Table 2.15. Minimum bridge cuts with four brakes and type B carabiners
Chapter 3
Table 3.1. Parametric validity intervals
Table 3.2. Comparative analyses. For a color version of this table, see: http...
Chapter 4
Table 4.1. Dimensions and properties of the system
Table 4.2. Characteristics of uncertain parameters
Table 4.3. DDO using GPC results
Table 4.4. OSF using GPC results
Table 4.5. Comparison of the optimal point between DDO and OSF
Chapter 5
Table 5.1. Properties of the model studied
Table 5.2. Maximum von-Mises stresses of the different layers
Table 5.3. Characteristics of design variables
Table 5.4. Comparing the accuracy of substitution models
Table 5.5. Comparison between the initial model, DMOO and RBMDO models
Chapter 6
Table 6.1. Coefficient of thermal expansion of materials used in microelectro...
Table 6.2. Parameters for the Coffin–Manson model
Table 6.3. Parameters of the Anand model [GRI 10]
Table 6.4. Material properties
Table 6.5. Upper and lower limits of the optimization variables
Table 6.6. Performance comparison of the two algorithms CMA-ES and KA-CMA-ES
Chapter 7
Table 7.1. Correspondence between the type of distribution and the type of th...
Table 7.2. Sizing and properties of the system
Table 7.3. Natural frequencies of the flexible plate
Table 7.4. Natural frequencies of the rectangular acoustic cavity
Table 7.5. Characteristics of uncertain parameters
Table 7.6. DDO result
Table 7.7. Results of the OSF coupled with the GPC
Table 7.8. Comparison of the optimal solution between DDO and OSF
Table 7.9. Comparison of the relative error between GPC and MC
List of Illustrations
Chapter 1
Figure 1.1. History of electronic predictive reliability collections
Figure 1.2. Illustration of the electronic items covered by the FIDES guide
Figure 1.3. Illustration of the FIDES methodology with its three contributors
Figure 1.4. Distribution of contributors for gelled tantalum capacitors
Figure 1.5. Breakdown for the origins of failures
Figure 1.6. Breakdown of the lifecycle
Figure 1.7. Influence of life profiles (FIT: number of failures per 10
9
hours)
Chapter 2
Figure 2.1. Representation of the handling system with one service brake
Figure 2.2. Stress–strength method
Figure 2.3. Stress cycle during testing
Figure 2.4. Representation of a handling system with a service brake and a saf...
Figure 2.5. Modeling failure of safety encoders without any common cause
Figure 2.6. Modeling safety encoder failures with a common cause
Figure 2.7. Bridge shaft head with one service brake
Figure 2.8. Hanging container (first part) type A carabiner
Figure 2.9. Hanging container (second part)
Figure 2.10. Bridge shaft head with two brakes
Figure 2.11. No stop of the bridge load with two brakes
Figure 2.12. No stopping the bridge load with two brakes
Figure 2.13. Typical fault tree model of a “secure” system
Chapter 4
Figure 4.1. Flowchart of the OSF function using the GPC metamodel
Figure 4.2. (a) Geometric model of the dental implant; (b) boundary conditions...
Figure 4.3. (a) Geometric model of the dental implant; (b) boundary conditions...
Figure 4.4. Iterative curve of the dental implant volume by the DDO method
Figure 4.5. Iterative curve of the dental implant volume through the OSF metho...
Chapter 5
Figure 5.1. Flowchart showing the developmental process of the substitution mo...
Figure 5.2. Transformation from a parameter space to an objective function spa...
Figure 5.3. Flowchart showing the evolutionary multi-objective optimization pr...
Figure 5.4. Diagram showing a) 3D shaft model and b) 2D shaft with different l...
Figure 5.5. Adduction boundary conditions
Figure 5.6. Cross-validation for maximum cortical layer stress with a) QRS and...
Figure 5.7. Cross-validation for the maximum stress of the cancellous layer wi...
Figure 5.8. Cross-validation for maximum shaft stress with a) QRS and b) Krigi...
Figure 5.9. DMMO Pareto optimal front
Figure 5.10. RBMDO Pareto optimal front
Figure 5.11. von-Mises distribution: a) Initial b) DMOO and c) RBMDO
Chapter 6
Figure 6.1. Diagram of the components in the mechatronic system
Figure 6.2. PQFP housing structure (type of solder joints)
Figure 6.3. Detachment of the wire bonding
Figure 6.4. Example of a solder joint failure
Figure 6.5. Diagram showing the stresses on the soldered joints due to the war...
Figure 6.6. Stress–strain evolution of a shear torsion specimen of a SAC305 al...
Figure 6.7. Global model of solder joints and the adopted mesh size
Figure 6.8. Local model of solder joints
Figure 6.9. Thermal load cycle
Figure 6.10. Boundary conditions applied to the sub-model
Figure 6.11. Plastic deformation history
Figure 6.12. Side view of the plastic strain distribution in the solder joint ...
Figure 6.13. Description of the optimization parameters
Figure 6.14. Inelastic strain distribution before (a) and after (b) the optimi...
Figure 6.15. Average curve of Δεin as a function of the number of numerical si...
Figure 6.16. Flow chart showing the global optimization methodology
Chapter 7
Figure 7.1. Implementation of structure optimization
Figure 7.2. Minimization of a one-variable objective function
Figure 7.3. Deterministic design optimization process
Figure 7.4. The reliability-based design optimization process
Figure 7.5. Iso-probabilistic transformation of the physical space to the stan...
Figure 7.6. Classical approach algorithm
Figure 7.7. Algorithm for the OSF method
Figure 7.8. Simply supported plate coupled to the end of a rectangular cavity ...
Figure 7.9. Flowchart showing the OSF combined with the GPC
Figure 7.10. Acoustic pressure inside the cavity
Figure 7.11. Displacement of the plate at the location of the force
Figure 7.12. Cavity acoustic pressure distribution (left: MC, right: GPC)
Figure 7.13. Acoustic pressure probability distribution for the optimal soluti...
Guide
Cover Page
Title Page
Copyright Page
Foreword
Preface
Table of Contents
Begin Reading
List of Authors
Index
Summary of Volume 2
Summary of Volume 3
Other titles from ISTE in Mechanical Engineering and Solid Mechanics
WILEY END USER LICENSE AGREEMENT
Pages
iii
iv
xi
xii
xiii
xiv
xv
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
205
206
207
208
209
211
212
213
214
215
216
217
218
219
221
222
223
224
225
226
227
228
229
230
Reliability of Multiphysical Systems Setcoordinated byAbdelkhalak El Hami
Volume 16
Applied Reliability for Industry 1
Predictive Reliability for the Automobile, Aeronautics, Defense, Medical, Marine and Space Industries
Edited by
Abdelkhalak El HamiDavid DelauxHenri Grzeskowiak
First published 2023 in Great Britain and the United States by ISTE Ltd and John Wiley & Sons, Inc.
Apart from any fair dealing for the purposes of research or private study, or criticism or review, as permitted under the Copyright, Designs and Patents Act 1988, this publication may only be reproduced, stored or transmitted, in any form or by any means, with the prior permission in writing of the publishers, or in the case of reprographic reproduction in accordance with the terms and licenses issued by the CLA. Enquiries concerning reproduction outside these terms should be sent to the publishers at the undermentioned address:
ISTE Ltd
John Wiley & Sons, Inc.
27-37 St George’s Road
111 River Street
London SW19 4EU
Hoboken, NJ 07030
UK
USA
www.iste.co.uk
www.wiley.com
© ISTE Ltd 2023The rights of Abdelkhalak El Hami, David Delaux and Henri Grzeskowiak to be identified as the authors of this work have been asserted by them in accordance with the Copyright, Designs and Patents Act 1988.
Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors, contributors or editors and do not necessarily reflect the views of ISTE Group.
Library of Congress Control Number: 2022944183
British Library Cataloguing-in-Publication DataA CIP record for this book is available from the British LibraryISBN 978-1-78630-691-3
Foreword
Predicting and then guaranteeing the reliability of an industrial system is a major challenge for manufacturers in the automotive, aeronautics and defense industries, as well as for those in the rail, telecommunications, nuclear and medical sectors, among many others. But above all, it is a major challenge for us, who use this equipment on a day to day basis, who must have absolute confidence in the information transmitted and the decisions made in real time. The increasing development of connected objects (autonomous cars, home automation, etc.) will lead to a drastic reduction of human intervention in favor of the intervention of mechatronic systems. All these systems can only be deployed if users have absolute confidence in the reliability of the equipment.
These forms of equipment are divided into two main functionalities: a hardware part mainly composed of electronic boards (coupled with mechanical systems) and a real time software part enabling the implementation of the equipment and the realization of the expected tasks.
Predicting and guaranteeing the reliability of electronic equipment is a huge and endless task. On the one hand, the number of component types used to build these boards is very high and on the other hand, the new functionalities in innovative equipment require numerous reliability and robustness tests.
Thanks to the support of the Astech and MOV’EO competitiveness clusters, the NAE aeronautics industry, the Normandy and Ile de France regions, and the Rouen and Versailles Chambers of Commerce, programs have been implemented and have produced exceptional results (AUDACE, FIRST-MFP, PISTIS, CRIOS, SYMSTECK, etc.).
Given the richness of the results of these three books that have been published, I would like to thank very warmly Abdelkhalak El Hami, David Delaux and Henri Grzeskowiak for their remarkable work in the implementation of six volumes in total as well as all the authors who have devoted many hours to shaping all their results. This has meant that this essential information does not remain the property of few but shared by the largest number of engineers, technicians, researchers and students.
The first volume is devoted to Predictive Reliability, for estimating or predicting the reliability performance of systems.
The second volume is dedicated to Experimental Reliability using tools and test methods to demonstrate the reliability of systems.
The third volume presents Operational Reliability; it verifies the reliability performance of systems in real life by way of an analysis of the field data.
I would like to thank all those involved in this program as well as the financiers (State, Regions) without whom these projects could not have succeeded and I wish that the members of the hard core set up in these programs (AUDACE, FIRST-MFP, PISTIS, CRIOS, SYMSTECK) continue their association because their skills will be essential for guaranteeing the reliability of the very many technologies which are in the process of emerging and which will equip the mechatronics systems of tomorrow.
Philippe EUDELINE
President of Normandie AeroEspace (NAE)
October 2022
Preface
March 2020 was a turning point in our lives and for industries worldwide when Covid-19 struck the world! This pandemic has highlighted the fragility of our industries while exacerbating the paradigms of development, innovation, commercial competition and the acceleration of research phases.
More than ever, we need to be agile in our American, European, Asian and even global business plans, while at the same time cementing the use of our products in their operational use. It is clear that the challenges of combining the concepts of variability and stability, standardization and acceleration of product development are enormous. The “post-Covid” world is pushing all industries to the very limits of reliability engineering.
It is true that the word “reliability” is often used as a marketing cosmetic without any real care for its underlying scientific substantiation! Nonetheless, behind this word lies an applied science, validated by analysis, research and talented people.
Which type of the reliability approach can a designer, an engineer and a manager use in their professional environment? Whatever the industrial field (aeronautics, space, defense, automotive, home automation, etc.), how is it possible to design a reliable product more quickly, all the while upholding safety measures?
This book illustrates, in great detail, the multidisciplinary, state-of-the-art science of predictive reliability. Many experts are convinced that reliability is not limited to the statistical sciences. Various disciplines interact in order to bring a product to its highest possible level of reliability, as available through today’s technologies, developments and production methods.
In the context of perpetual research, in order to improve industrial competitiveness, the evolution of product design methods and tools appears to be a strategic necessity with regard to the need to reduce costs. Nevertheless, the reduction of design costs must not be done at the expense of reliability for the new systems being proposed, which must also make significant advancements.
To increase the competitiveness of their mechatronic devices, automotive and aerospace suppliers need to find innovations, for both the design and assembly processes, in order to reduce product development times. In addition, these innovative products must combine excellent functional and operational performance, which includes reliability, to fully meet the expectations of the global market. For car manufacturers, these reliability expectations will certainly increase with the intensified market penetration of electric or hybrid vehicles by 2020–2025, estimated at 10% of the market. In addition to these expectations for operational reliability, there is also the need to quickly eliminate the risks of immaturity associated with product innovations. This need is strongly linked to reducing vehicle development times to a strict minimum.
In the field of aeronautics, the needs are mainly related to the forecasting and control of costs induced by failures occurring at the commissioning phase, during the warranty period and during the operation of the aircraft, with the knowledge that, in the future, new aeronautics equipment sales contracts will be increasingly oriented toward sales by the operating hour. Although the aeronautical sector has relatively low production volumes when compared to the automotive industry (in terms of the number of units per type of product), the financial stakes are high and, as a result, aeronautical manufacturers are over-sizing mechatronic components to counteract problems they are currently unaware of. Reliability is defined by the international standard ISO/TR 12489:20131 as: “The probability for an item to perform a required function under given conditions over a given time interval”.
This book aims at proposing new methods to analyze the lifecycle of a system. The lifecycle enables us to record the development phases according to the development time and the levels of complexity for its integration.
The three main families of activities governing the development of reliability for a mechatronic system are predictive reliability, experimental reliability and operational reliability.
Predictive reliability includes all engineering activities used to estimate or predict the reliability performance of the final mechatronic system. Predictive reliability is particularly focused on in Applied Reliability for Industry 1.
Experimental reliability includes all the tools and testing methods used to demonstrate the reliability of the mechatronic system. Experimental reliability is advanced in the Applied Reliability for Industry 2.
Operational reliability verifies the reliability performance of the mechatronic system in its real life through an analysis of the field data. Operational reliability is presented in the Applied Reliability for Industry 3.
The first chapter in Applied Reliability for Industry 1 presents FIDES, a method for evaluating and constructing the reliability of electronic systems. The second chapter presents “Reliability in Maritime Transport: Choosing a Container Handling System”. Chapter 3 presents the “Generation of a Failure Model Through Probabilistic ‘Stress–Strength’ Interaction in a Context of Poor Information”. Chapter 4 is devoted to “Reliable Optimization of Dental Implants Using the Generalized Polynomial Chaos Method”. Chapter 5 presents “Multi-Objective Reliability Optimization Based on Substitution Models. Applied Case Study of a Hip Prosthesis”. Chapter 6 proposes an adaptation of the CMA-ES method that is assisted by the kriging metamodel for the global optimization of a problem. The implementation of this proposed algorithm has been carried out for the global optimization of solder joints for a PQFP housing. The results of numerical studies show that the proposed KA-CMA-ES algorithm is more efficient and effective than the standard CMA-ES algorithm. Chapter 7 presents the “Reliable Optimization of Vibro-Acoustic Problems in the Presence of Uncertainties via Polynomial Chaos”.
To conclude, there is a before and an after Covid. The reliability challenges for our industries to address in this new period are significant. The global context but also the green, sustainable and carbon-neutral goals are the next steps in a fierce economic competition that will span the next decade. The authors hope that Applied Reliability for Industry 1 on predictive reliability will impart meaningful reflections and help to guide readers toward the best practices to encourage efficient design and effective decision-making.
This book is dedicated to all those who have lost a loved one during the Covid-19 pandemic worldwide.
Abdelkhalak EL HAMI
David DELAUX
Henri GRZESKOWIAK
October 2022
Note
1
[ISO 13] ISO/TR 12489:2013, Petroleum, petrochemical and natural gas industries – Reliability modelling and calculation of safety systems, ISO/TC 67, edition 1, 2013.
1FIDES, a Method for Assessing and Building the Reliability of Electronic Systems
Reliability has become a key performance of current developments. It is now, more often than not, a contractual performance in its own right, generally associated with strong commitments, monitored on-the-ground and subject to severe penalties. Apart from these contractual aspects, reliability remains a crucial parameter for the overall cost of ownership and availability of a product.
On the other hand, predictive reliability is an essential input for most quantified operational safety studies, such as the sizing of the support system: FMEA/FMECA (failure mode, effects and criticality analysis), probability of correct functionality, maintainability, testability and availability.
Lack of confidence in predictive reliability taints all analyses that depend on it with uncertainty. This can have profound negative consequences on product development (e.g. by leading to the erroneous rejection of one technology in favor of another, or by dictating inappropriate architecture choices), but also on costs. It is therefore imperative to have realistic predictive reliability assessments. The FIDES reliability assessment and construction method [FID 09] (from the Latin “confidence”) meets these requirements for electronic systems.
1.1. The inadequacy of existing methods
Since 1965, MIL-HDBK-217 [MIL 95] has been one of the historical standards long used by the vast majority of military equipment manufacturers (but not only by these) to predict the reliability of electronic equipment. Today, this method has become inadequate as a result of the evolution of components and the usage of systems, forcing many users to implement adjustments (more or less precise and justified) in order to make the prediction as accurate as possible.
In addition to the FIDES methodology, there are three main predictive reliability methods that are commonly used.
1.1.1. MIL-HDBK-217F [MIL 95]
This method, often presented as a standard but which in fact is only a handbook, still remains the flagship in the field. However, it is largely obsolete. It has not been maintained since 1995 and is generally based on user experience from the 1980s. Its content is no longer representative of current technologies. The MIL-HDBK-217 guide has been adapted to certain specific fields, such as telecommunications in the United States (Bellcore/Telcordia SR-332).
1.1.2. UTE-C-80810 (or RDF2000, or IEC 62380 TR Ed.1)
This compendium of reliability covers most of the recent technologies [UTE 00]. Nevertheless, it suffers from certain shortcomings that do not allow it to provide an all-inclusive answer. In particular, RDF 2000 is inadequate for dealing with non-benign environments. It does not enable us to distinguish the different suppliers of the same item, nor the level of manufacturer processes. The IEC 62380 standard, which is not maintained, was recently withdrawn from the international IEC standard.
1.1.3. PRISM® or 217plus
This is a commercial software product, initially developed by RAC, of which another version (217plus) is now being marketed by QUANTERION [DEN 97]. This tool proposes an approach to predictive reliability that breaks tradition with MIL-HDBK-217. However, it has several drawbacks that prevent its recognition. At the end of the day, its use is limited outside the United States. The proposed calculation models are rather poor. The coverage of recent component families is still quite limited, and the availability of a database of raw failure rates is only a meager counterpart. The 217plus method does not allow us to deal with complex life profiles (at most one operating phase) or to distinguish between different suppliers of the same item.
The existing methods are therefore inadequate to meet the needs expressed above. Moreover, none of these methods can really be used as support for a reliability engineering activity. Finally, we can see in Figure 1.1 that only FIDES has evolved in the last few years with a new version under construction and with a possible release in 2023.
Figure 1.1.History of electronic predictive reliability collections
1.2. The ambition of FIDES
In order to have a new reference system, the DGA (Direction Générale de l’Armement: “Directorate General of Armaments”) brought together a consortium of eight major aeronautical and defense companies in early 2000: AIRBUS, Eurocopter, MBDA Missiles Systèmes, Nexter Systems, Thales Research and Technology, Thales Airborne Systems, Thales Avionics and Thales Underwater Systems.
The first objective of this study was to have a new reliability assessment method for electronic components that is realistic and takes into account new technologies. The second objective was to provide a reliability engineering guide to support manufacturers in the development of new electronic systems.
The main objective of FIDES is therefore to have a reliability assessment method for electronic components that is an alternative to the MIL-HDBK-217 standard and that takes into account new technologies, in particular, the consideration of Commercial Off-The-Shelf (COTS) components that are increasingly used. One of the characteristics of FIDES is the identification and consideration of all technological and physical factors that have an impact on reliability. This guide also considers failures related to development, manufacturing, operation and maintenance processes, as well as support activities during the entire product lifecycle. It is consolidated by analyses of test data, user experience and existing models.
In this context, the purpose of the FIDES guide is to propose a new methodology that is:
– realistic and accurate;
– usable for both the estimation and construction of reliability;
– usable for any type of item (components, assemblies, sub-assemblies), including COTS, as illustrated in Figure 1.2.
Figure 1.2.Illustration of the electronic items covered by the FIDES guide
A first version of the guide was published in 2004 (FIDES 2004A) and an update was published in 2010 (FIDES 2009A) [FID 09]. The FIDES guide is available free of charge on the FIDES website (www.fides-reliability.org) and is well distributed in the French industry and throughout the rest of the world as well.
Since its publication, FIDES has aroused great interest. As a result, the Directorate General of Armaments has approved the method as a “standard applicable to French military programs”; the guide has been integrated into the Référentiel normatif des programmes d’armement français (the “normative reference framework for French armament programs”) and, since 2011, into the European Defense Agency’s (EDSTAR) reference system. In addition, the board of directors of the UTE (now AFNOR) accepted the FIDES guide as a technical document at the end of 2004, under the reference UTEC 80-811. In France, FIDES is therefore accessible to anyone as a recognized organization in the field of standardization, which allows for a wider distribution to organizations outside of the consortium (equipment manufacturers, CNES [VER 05a, VER 05b]). The extension to an international document is underway within the IEC through a group of international experts representing 22 nations. The international FIDES standard under the reference IEC 63142 was officially released in December 2021. In addition, we have noted the interest shown by many international companies, notably American companies such as Boeing [BEC 04] or Raytheon, but also major automotive manufacturers in Europe. A tangible sign for the utility of FIDES is that most of the publishers for operational safety software throughout the world now offer a FIDES module.
1.3. General presentation of the FIDES method
The main features of FIDES are as follows:
– models for EEE components (electrical, electronic, electromechanical) as well as for electronic boards or some sub-assemblies, and for recent technologies;
– a methodology applicable to both military or space components (so-called high reliability) and COTS and for all fields (military, industrial or civil);
– the identification and consideration of all technological and physical factors that have an identified role in reliability;
– models based on the physics of failures, which are not only based on experiential feedback but also on the exploitation of reliability tests (active) or manufacturer data (sub-assemblies);
– precise consideration of the life profile, as opposed to the use of standard predefined environments;
– consideration of accidental electrical, mechanical and thermal overloads (or overstress);
– taking into account failures related to the development, production, operation and maintenance processes;
– the possibility of distinguishing several suppliers of the same component;
– easier updating, and the possibility to deal with new technologies without waiting long for user feedback.
The FIDES reliability approach described in the publication [GUI 04] is based on the consideration of the three components (technology, process and usage) as represented in Figure 1.3, which represents the “logo” of the method. These components are considered throughout the entire lifecycle, from the product specification phase and through the operation and maintenance phases. “Technology” covers both the technology of the item itself and the equipment into which it is integrated. The “Process” considers all the practices and rules of the art, from the specification of the product to its replacement. Finally, “Usage” takes into account both the constraints of use defined by the design of the equipment manufacturer and those in operation by the end user.
Figure 1.3.Illustration of the FIDES methodology with its three contributors
The proposed models therefore consider a “Technology” subject to “Usage” constraints according to an approach that takes into account the failure mechanisms and associated contributors, and above all, they weigh the risk of failure by all the “Process” contributors that can activate, accelerate or reduce the effects of these mechanisms.
By identifying the contributors to reliability, whether technological, physical or process-based, the FIDES methodology is able to act according to definitions throughout the whole lifecycle of the products so as to improve and control the reliability.
1.3.1. Failure rate
The FIDES evaluation model assumes that the failure rate is constant, at least during the so-called service life phase. The periods of youth and old age are excluded from the forecast. Strictly speaking, the failure mechanisms do not respond to a modeling of their likely occurrence through a law whose parameter (failure rate) would be constant. However:
– many failure mechanisms, although cumulative, which therefore increase over time, have such a dispersion in their distribution that they can be assimilated to a constant over the periods being considered;
– the multiplicity and diversity of the components, even on a single Printed Circuit Board Assembly (PCBA), will lead to an accumulation that is very close to a constant;
– the differences in the lifecycle situation of the components between the equipment of a same system, or a multiple system, result in a constant rate for the observer of the system level.
For these reasons, the use of a constant failure rate remains the most relevant approach for estimating the predictive reliability of electronic systems.
The inadequacy of available predictive reliability methods, and in particular MIL-HDBK-217, has often led to the use of a constant failure rate being considered fundamentally inadequate, instead favoring deterministic life prediction approaches (e.g., the simulations proposed by CALCE, University of Maryland, USA [BAU 94]).
In reality, these two approaches are essentially complementary.
1.3.2. The structure of FIDES models
The general FIDES model allows for the calculation of failure rates for electronic equipment before any consideration of redundancy or architecture.
The overall failure rate of the electronic equipment is obtained by adding up the failure rates of each of its components, thus considering a serial system. We have the usual relationship of:
The FIDES methodology is based on four key components:
– the physical failure rate, which reflects the contribution of technologies and constraints: λPhysical;
– induced failures (accidental overloads): ΠInduced;
– the manufacturing quality of the part: ΠPart_Manufacturing;
– the level of control and construction of reliability in the lifecycle: ΠProcess.
The general equation for FIDES models is given as:
with:
where
– λ0 is the basic failure rate, and also representative of the characteristics of the technology;
– Π_acceleration is an acceleration factor reflecting the sensitivity to a physical contributor: electrical stress, temperature, thermal cycling, humidity, mechanical, chemical environment (specified nominal contributions);
– ΠInduced represents the contribution of induced factors (also called accidental overloads or overstress), typically expected in a given application (EOS, ESD, etc.).
