Soft reliability in new product development : an ontological approach for utilizing field feedback to dynamically sense and adapt to evolving global markets

(1)

Soft reliability in new product development : an ontological

approach for utilizing field feedback to dynamically sense and

adapt to evolving global markets

Citation for published version (APA):

Koca, A. (2010). Soft reliability in new product development : an ontological approach for utilizing field feedback to dynamically sense and adapt to evolving global markets. Technische Universiteit Eindhoven.

https://doi.org/10.6100/IR674211

DOI:

10.6100/IR674211

Document status and date: Published: 01/01/2010

Document Version:

Publisher’s PDF, also known as Version of Record (includes final page, issue and volume numbers)

Please check the document version of this publication:

• A submitted manuscript is the version of the article upon submission and before peer-review. There can be important differences between the submitted version and the official published version of record. People interested in the research are advised to contact the author for the final version of the publication, or visit the DOI to the publisher's website.

• The final author version and the galley proof are versions of the publication after peer review.

• The final published version features the final layout of the paper including the volume, issue and page numbers.

Link to publication

General rights

Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. • Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain

• You may freely distribute the URL identifying the publication in the public portal.

If the publication is distributed under the terms of Article 25fa of the Dutch Copyright Act, indicated by the “Taverne” license above, please follow below link for the End User Agreement:

www.tue.nl/taverne Take down policy

If you believe that this document breaches copyright please contact us at: openaccess@tue.nl

providing details and we will investigate your claim.

(2)

An Ontological Approach for Utilizing Field Feedback to

Dynamically Sense and Adapt to Evolving Global Markets

(3)

Chairman:

prof.dr. A.G.L. Romme, Eindhoven University of Technology, NL

Supervisor :

prof.dr.ir. A.C. Brombacher, Eindhoven University of Technology, NL

Supervisor :

prof. F. Mistree PhD, Georgia Institute of Technology, Atlanta, USA

Members:

prof.dr.ir. J.P.M. Geraedts, Delft University of Technology, NL prof.dr. E.J. Hultink, Delft University of Technology, NL

prof.dr.ir. J.-B.O.S. Martens, Eindhoven University of Technology, NL dr. A.J.M.M. Weijters, Eindhoven University of Technology, NL

The work presented in this dissertation has been performed in the Faculty of Industrial Design and the Faculty of Technology Management at the Eindhoven University of Technology, and has been sponsored by the Dutch Ministry of Economic Aﬀairs within the IOP-IPCR program, through the “Soft Reliability” project (www.softreliability.org), under the auspices of Philips and Oc´e. A catalogue record is available from the Eindhoven University of Technology Library ISBN: 978-90-386-2242-2

NUR: 964 c

� Aylin Koca, 2009. All rights reserved. No part of this book may be reproduced, stored in a re-trieval system, or transmitted, in any form, or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior permission of the author.

M.C. Escher‘s “Metamorphosis II” c� 2009 The M.C. Escher Company B.V.-Baarn- The

(4)

An Ontological Approach for Utilizing Field Feedback to

Dynamically Sense and Adapt to Evolving Global Markets

PROEFSCHRIFT

ter verkrijging van de graad van doctor aan de Technische Universiteit Eindhoven, op gezag van de rector magnificus, prof.dr.ir. C.J. van Duijn, voor een

commissie aangewezen door het College voor Promoties in het openbaar te verdedigen

op woensdag 19 mei 2010 om 16.00 uur

door

Aylin Koca

(5)

prof.dr.ir. A.C. Brombacher en

(6)

most responsive to change.”

(7)

(8)

Acknowledgements

About four years ago, at the end of my yearlong research in England, I made an important and diﬃcult decision that would impact not only my career, but my whole life alike. In contrast to where the flow of my studies in computational linguistics was taking me, I decided to defer the long-awaited oﬀer to relocate to Boston for my PhD. Instead, I chose to do my PhD in what appeared to be a very promising multidisciplinary project on “Soft Reliability” that aimed to address an increasingly relevant real-life problem, and with strong industrial prospects. Convinced by the great potential of this project in academic, industrial, as well as in social terms, I relocated to the Netherlands to start my PhD at TU/e. It must be a blessing that I never had to reconsider my decision in retrospect, and I am more than grateful to all those who have kindly contributed to this. As the French proverb goes, “Gratitude is the heart’s memory.”

Supervisors and Doctorate Committee

First and foremost, my supervisors and mentors Prof. Aarnout Brombacher and Prof. Farrokh Mistree. It would probably never have occurred to me earlier on that doing multidisciplinary research with a team of researchers and industrial partners would be so gratifying, if at all even possible. Aarnout was the one to convince me about the potential of the “Soft Reliability” project, a product of his incredible vision back in 2005. With the concrete project infrastructure that was already set up then, he has been the actual reason of the radical choice I made to move to the Netherlands instead of to Boston and do my PhD in the faculties of Technology Management and Industrial Design, respectively, instead of in Computer Science. Along the way, but even more now in retrospect, I feel most privileged to have witnessed at first hand the spirit and unequaled skills with which Aarnout founded, steered and brought this grand project to a successful culmination as the project leader. With the opportunity to closely be exposed to his unique way of working, his always constructive approach to identifying and resolving relevant problems, and his most encouraging enthusiasm that unmistakably helps bring out the best in his

(9)

his unreserved praise for and endless support of my work right from the beginning have been essential for my quick development within the new research domain. For all of that, words could never convey my gratitude (I hope to explore further if baklava helps at all with this). Aarnout, I am convinced that if we continue probing my endurance during the marathon trainings, I can keep on learning even more from you...

My PhD studies would not have been as complete without Farrokh’s invaluable con-tributions in all aspects, which helped me sharpen details of my work, while also emphasizing the global picture, specifically in a multidisciplinary research context. I cannot thank him enough for his most dedicated and generous support especially during the part of my studies at Georgia Tech in Atlanta. My experience there would never possibly have been as enriching, particularly in establishing a wide-ranging topnotch multidisciplinary network (even beyond the borders of Georgia Tech, be-yond Atlanta, and bebe-yond the U.S.), if it were not for his focused guidance, his priceless references, and his enabling collaborations “without borders.” The what-almost-felt-like unconditional trust he put in me (I would never have foreseen driving on my own in the car of my supervisor, for a business trip from Atlanta to South Carolina and back), and the always-substantiated encouragement he has given, have been a true source of inspiration for me to this day. His meticulousness and rigor in research, writing, presenting, educating, collaborating and networking will keep on most pleasantly overwhelming and motivating me in more than many ways as time goes by. Farrokh, it has not been easy to establish my first contact with you; and I would hereby like to thank you very much for allowing me the possibility to work with and to learn so much from you. Your mentoring perfectly harmonized with, and almost entirely complemented any other I have had so far. I am sincerely grate-ful for and honored by your presence at my graduation and at all prior milestone meetings, despite the physical distances involved. You truly set a real-life example of why it is worth going that extra mile.

Valued members of my doctorate committee, Prof. Jo Geraedts, Prof. Erik Jan Hultink, Prof. Jean-Bernard Martens and Assoc. Prof. Ton Weijters. I would like to thank you wholeheartedly for your kind consideration of my dissertation during its preparation and your most constructive feedback that helped greatly in the process. I feel privileged and honored to have benefited from your wisdom regarding your respective fields of expertise. It would be my hope that this special occasion leading to my graduation demarcates our possible collaborations in the future. I am also thankful for the circumstance that allowed the opportunity to ask Prof. Sjoerd Romme to join my doctorate committee as chairman, who has very kindly accepted to be on board.

(10)

The dream team of the “Soft Reliability” project comprised some great minds, Anne Rozinat, Evangelos Karapanos, and Mathias Funk. The outcome of this project would probably not have been the same if it were not for this unique composition of perspectives, backgrounds, and personalities. Despite all the diﬀerences, and maybe even fair to say that due to all the diﬀerences, we did accomplish a lot together. And not just research output, or collaborations with industrial partners, or even founding new high-tech start-up businesses; but also culinary get-togethers, other social meetings, or sometimes even exploring unknown cities abroad... I hope whatever it was that brought us together in the first place is forever there to keep us connected no matter what!

The strong industrial prospects of the project would never have been realized with-out the dedicated support and the guidance of our industrial partners. In this regard, especially Fred de Jong from Océ, and Mariëlle Meuffels and Nico van der Gaarden from Philips have been pivotal. It is with their vision and unique exper-tise in their respective domains that many of the industrial case studies could be realized throughout the four years, which involved various industrial products and product stakeholders to collaborate with, and to explore or test the ideas presented in this dissertation. The added value of being provided with such an opportunity is incalculable whereas the possibility of it in the context of PhD studies is most likely quite rare if not nonexistent. Therefore, I would herewith like to thank Fred, Mariëlle and Nico greatly for all their invaluable contributions; and Guus Lambregts as well, for his helpful feedback during the regular project meetings. I hope we get many other opportunities to continue our collaborations. There have additionally been many industrial associates, who have been indispensable to parts of my work. I try to acknowledge their individual involvements in the respective chapter-end “Acknowledgements” sections of this dissertation.

In conclusion, the honorable mention of the involved professors, and the IOP-IPCR board facilitating the generous sponsoring of the project. The professors have been, in particular, Prof. Aarnout Brombacher as “Soft Reliability” project leader and as professor in Industrial Design, Prof. Wil van der Aalst as professor in Mathematics and Computer Science, Prof. Henk Corporaal as professor in Electrical Engineering, Prof. Jean-Bernard Martens as professor in Industrial Design, Assoc. Prof. Piet van der Putten as professor in Electrical Engineering and Assoc. Prof. Ton Weijters as professor in Industrial Engineering and Innovation Sciences. Their most insightful feedback while reviewing my written work, during regular project meetings, and at various meetings with the industrial partners/associates, helped significantly in im-proving my dissertation and especially in strengthening its multidisciplinary nature.

(11)

ing the IOP-IPCR board, who have closely been involved in the progress of work in the most benevolent way to ensure that the best possible infrastructure is pro-vided to boost the overall performance. Their kind sensibility combined with the numerous possibilities they have generously enabled over time (e.g., my internships in Singapore and in the US, the grants I got for valorization of my research and in support of establishing a start-up company, numerous international conference visits and business trips for the firsthand dissemination of my work and the project in general, other national networking/publicity events, invited talks, workshops, ca-reer/personal development training courses) helped dramatically to develop myself further in the context of my research, the project, and beyond.

Paranymphs

Mathias and Evan, I am proud and honored to have you both by my side at my defense. I hope we keep sharing many wonderful occasions together, side by side.

Colleagues at TU/e and at Georgia Tech

We have literally come a long way together (from the Paviljoen to the Hoofdge-bouw) with my colleagues in the Business Process Design group of Industrial Design, initially setting off as the Quality and Reliability Engineering group of Technology Management: Aarnout Brombacher, Elke den Ouden, Lu Yuan, Jan Rouvroye, Dim-itrios Karydas, Ilse Luyk, Peter Sonnemans, Hanneke Driessen, Valia Petkova, Mau-rits Houben, Kostas Kevrekidis, Aravindan Balasubramanian, Joël Luyk, Jeroen Keijzers, Sander Mulder, Christelle Harkema, Josephine Sari, Clèment Magniez, Wim Geudens, Girish Thiruvenkadam, and Xin Yan, and more recently, Alex, Jeff and Eva. I would like to thank you dearly for all the time we spent together working, lunching/dining, (marathon) running, traveling, and enjoying countless discussions and time together to the fullest extent wherever we are. Also thanks to you Ana Karla, for the crash course on how to do semantic process mining and our enriching collaborations. It turns out that migrating to a new department halfway through PhD studies can be very worthwhile indeed, in terms of expanding the available resources and assets. In particular, I am thankful for not missing the opportunity to meet you Aga, Wouter, Greg, and Berke. You are wonderful colleagues and friends. Also my colleagues from the Woodruff School of Mechanical Engineering at Georgia Tech have been amazing in helping me make the most of my stay in Atlanta. Thanks especially to Farrokh Mistree, Janet Allen, Jitesh Panchal, Chris Paredis, Rich Malak, John Reap, Jamal Wilson, Sarah Engelbrecht, Roxanne Moore, Stephanie Thompson, Julie Bankston, Greg Graf, Fei Ding, Jeff Olson, and Stephanie Merrick.

(12)

Experiencing the industrial environment at firsthand, especially right at the begin-ning of my PhD has proved extremely valuable to quickly grasp the industrial Soft Reliability domain and define my own approach. I am indebted to Elke den Ouden for kindly welcoming me in her group Industry Consulting. In this group, I learned a great deal regarding various product development processes not just from expe-rienced colleagues, but also at the weekly seminars, and from in depth discussions with invited speakers. I would like to thank especially Mariëlle Meuffels, Jaring Boersma, Herman Hartmann, Vincent Ronteltap and Simon Minderhoud for pro-viding feedback on various parts of this dissertation. Spanning the period of my affiliation, Industry Consulting relocated to the new High Tech Campus Eindhoven. Without the administrative support of Hennie van den Berg, the relocation would have been much harder for me.

Colleagues and Mentors at Bilkent University, Computer Engineering The period I spent at Bilkent University’s Computer Engineering department, first as an undergraduate student and then as a research assistant during my MSc studies, has a special place in my development as a scientific researcher. Hereby, I would like to pay homage to the great educators, researchers, mentors, and peers I met there, who have inspired me in more than many ways to this day. In particular, I would like to convey my deepest gratitude to Prof. Varol Akman, my supervisor during my MSc studies.

Family

Grandpa. You are my role model. Your overwhelming wisdom, countless accom-plishments to this day, and your perseverance and steadiness in life have always managed to astonish and greatly motivate me to try to do better and better. I am here today, where you once happened to be as a young researcher abroad. It is an honor for me now to dedicate this dissertation to you and grandma.

Mom, Dad and Ilkay. You have been my glowing pride ever since my childhood, and I will strive to be yours forever. Words can never convey my gratitude to you. I deeply treasure your tremendous eﬀorts and devotion through thick and thin, and your endless support of my choices and my endeavors in life, which on this occasion bring us together in Eindhoven. Thank you for being there for me today and always.

G¨urcan. The warm glow of contentment you make me feel inspires me to surmount

any diﬃculty, and makes me impatient to explore far and beyond with you.

(13)

(14)

Abstract

Soft Reliability in New Product Development

An Ontological Approach for Utilizing Field Feedback to

Dynamically Sense and Adapt to Evolving Global Markets

Today’s global businesses are prone to being both empowered and overpowered by the accessibility of various technological and market opportunities. For instance, the merging of digital technologies allows for the development of very innovative, multifunctional and adaptive products for use within rich socio-cultural contexts such as the high-end oﬃce, the digital home, and professional or personal healthcare. However, an important consequence is the growing market uncertainty regarding if, how, and when users can and will adopt such products. In fact, this uncertainty has already become an evident concern in the field where a large and rapidly increasing share of user complaints are reported that cannot be attributed to any violation of products’ technical specifications. Accordingly, the major portion of product rejections today tend to be not due to traditional (i.e., hard) reliability problems that can be resolved by repair/replacement of defective parts, but rather due to Soft Reliability (SR) problems that require instructional guidance for the user and adaptive redesign of the product. Nonetheless, current operational quality analysis and evaluation methods do not employ a SR perspective, and as such, market feedback from the field (about situational or contextual use factors) is not eﬀectively utilized in New Product Development (NPD) processes to collaboratively improve the quality of products and processes. Consequently, unforeseen user experiences relating to functional, emotional, and social aspects of product use in various geographical, cultural, or situational contexts remain as a growing and uncontrollable problem. In this dissertation, it is argued that a whole new NPD approach to evaluation of product design is needed, which is based on the continuous utilization of field feed-back from real customers, in order to dynamically sense and adapt to the (rapidly) evolving needs and expectations of global markets. In particular, two techniques are proposed; (i) continuous summative user experience evaluation based on the

(15)

tions, and the user reviews collected on Internet resources (e.g., public forums), and (ii) continuous formative user experience evaluation based on in situ feedback from successive field studies early on in NPD, involving functional prototypes. Both tech-niques aim at meaningful and structured longitudinal user experience data collection and analysis from the field that is scalable, standardized to allow comparisons and benchmarking, and that can be used to manage the field uncertainty that is growing due to the recently emerging market conditions. As a means to enable these tech-niques, an ontological model is developed to capture and analyze field data, which is motivated by the findings from extensive field explorations, and which is based on concepts from the related literature.

The first part of the dissertation is about understanding the SR problem domain. The point of departure is described as originating from the Quality and Reliability Engineering domain, and the operational context of SR is presented by exploring the available SR information resources and flows in practice. The second part of the dissertation is about designing an ontological approach to modeling SR. The ex-clusive utilization of the ontological approach, as well as its combined utilization with other techniques such as the decision support problem technique for informa-tion deployment, remote product observainforma-tion technique for capturing the relevant user interactions with the product, or semantic process mining technique for a rich and possibly combined analysis of user experience feedback in the context of user actions, are all demonstrated in industrial case studies in the last part of the dis-sertation, which is about operationalization of the proposed approach. As a result, meaningful and structured longitudinal user experience data collection and analysis becomes possible for both summative product evaluation after products’ release to the market, as well as for formative product evaluation during product development. There is a steadily growing body of detailed knowledge in the human-computer interaction domain about not just the usability of interactive products, but also -and as of more recently- about the holistic view of the psychological and social impact of products in people’s lives, i.e., the user experience of products. Nevertheless, operational models about the pleasure, fun, aesthetics, and hedonic qualities in the use of interactive products have not yet matured to be consistently used in practice. In fact, even the research on methods for the assessment of users’ experiences is only at its infancy, despite the wealth of methods and techniques that are available for assessing the usability of interactive products. Therefore, to help extracting the actual in situ user experiences from field usage of products, the proposed ontological approach proves as an operationalizable model that enables product developers to be aware of and to be systematically responsive to the growing and uncontrollable problem of unforeseen user experiences in the field.

(16)

List of Abbreviations

CAO Consumer Appraisal Ontology

CAO-v1 Consumer Appraisal Ontology, version 1 CAO-v2 Consumer Appraisal Ontology, version 2 CAO-v3 Consumer Appraisal Ontology, version 3 CAO-v4 Consumer Appraisal Ontology, version 4

CC Call Center

cDSP compromise Decision Support Problem

CE Consumer Electronics

CFS Customer Feedback System

CRM Customer Relationship Management

EPG Electronic Program Guide

IRIS International Repair Information System

IT Internet

ITV Internet on TV

IWS Initial Repair Workshop

MIR Maturity Index on Reliability

NDA Non-disclosure Agreement

NFF No Fault Found

NPD New Product Development

(23)

PRQ Primary Research Question

QFD Quality Function Deployment

RQ Research Question

SC Service Center

SR Soft Reliability

TTP Time Till Problem

UT User Test

UX User eXperience

(24)

Chapter 1 Foundations for Soft Reliability in

New Product Development

1.1 Emerging Challenges in New Product

Devel-opment

New Product Development (NPD) practices have evolved at a rapid pace in the last decade as product cost, quality, and time-to-market have each become progressively important. Consequently, product life cycles for many product categories decreased significantly [13]. Whereas in previous years, the technology development, product development and transfer to volume production for a common product might have occurred in several phases spread over 10-15 years, now the same processes take no more than 2-5 years [76]. Accordingly, a study on the changing market trends impacting reliability of technical systems in the consumer electronics industry has revealed four major trends [12]:

• Product Complexity and Price Erosion

Products are becoming increasingly more complex1 _{(i.e., as implied by Moore’s}

law), due to new technology becoming available at lower prices ever faster. Consequently, product specifications become increasingly more complex, often incomplete, and hence an unreliable source to verify the quality and reliability of commercial products with.

1_{Many electronic products today are multifunctional, or adaptive to their users, or}

context-aware, or networked with other electronics products, or embedded into the environment, and hence resemble (ambient) intelligence in some way [1].

(25)

2 Chapter 1. Foundations for Soft Reliability in New Product Development

Moreover, as technologies and product functionalities become available for the mass market at a higher pace and at lower prices, product adoption cycles are getting shorter [76]. In Figure 1.1, the quantities of products sold on the European markets for VCRs, DVD players, and DVD recorders, and their so-called “street price” are compared.

77"

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

Figure 1.1: Market dynamics for three kinds of consumer electronics products [76]

As a result of shortened product adoption cycles, the consumer profile and hence the requirements of consumers purchasing and using the same products change much faster over time. As Rogers [91] indicated, the characteristics of consumers can be classified into five adopter categories: innovators, early adopters, early majority, late majority, and laggards. Although individual consumers do not fit into these categories as such, the main diﬀerences between the categories are relevant for understanding the consumer expectations over the adoption cycle. Since adoption cycles are shortening, a much bigger spread of consumers buy and use products than before. In Figure 1.2, the example of the VCR versus the DVD recorder is shown, where the second generation DVD recorder is already reaching the early majority.!"#"$%&'"()*%+*,*!"-./(*0(,$1-.-*2%3"$*+%4*5%(-6'"4*5%'&$,.()-*

78*

*

9./64"*8:;*:*03%&).%(*<1<$"=*4"36<"3*).'":)%:<%''%3.)1**

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{(A%<)(-BC#%,$>$)"#*(>.D} EFEGHD%I4 FJHD%K3L

(26)

• Competition and Rapid Development

There is strong pressure on time-to-market due to increasing competition in the global arena. Consequently, development times decrease, whereas field feedback operations cannot keep up due to their need to handle increasingly complex information.

The time-to-market pressure yields diﬀerent challenges for the first generation and the second generation products in the same product line. For the first generation products, since NPD projects have less time available for quality management, consumer testing and quality testing is done less vigorously, while the implementation of the results is under pressure [76]. On top of that, for the second generation products the field feedback from the first generation is not available in time, and hence learning from the field is hampered.

(better functionality against lower costs) it becomes interesting for all types of business processes at least to consider the use of new, not yet fully proven and therefore often not quite mature technology. For example, in Military industry (originally a typical class C business processes) the functionality and costs of products generated with class A business processes are such that the use of commercially of the shelf (COTS) products is currently seriously being considered. As a consequence it can be expected that reliability problems will shift from B3/C4 type of problems to A1/B2 type of problems. It is the question whether the current industrial quality and reliability management methods are able handle this challenge.

To answer this question two parameters will be used: the speed7of the quality control loop and the granularity8of the quality control loop. Speed is used since for innovative products unexpected, new, reliability problems may appear. If the rate in which new technology is introduced is much faster than the time that is required to learn the actual field reliability performance of these products, companies will, sooner or later, have insufficient time to really understand the root causes of field problems. Due to the increasing product complexity it will be also necessary to have far more detailed field feedback; understanding failure mech-anisms in complex products may require far better skills and analysis techniques than are currently available with the service people who act, for may companies, as the main source of product reliability data.

3.8. The increasing time gap (processing time and structure for data feedback)

One of the key-problems in fast, strongly innovative product development processes is the difference between the time that is required to develop a product and the time needed to learn about the actual product performance in the field. Applying new technology in new products and submitting them to, for this product, new customers will always involve a high degree of uncertainty; uncertainty about the performance of the new technology and uncertainty about the way customers will apply this new technology. Over the last decades the speed to bring new technology to the market has increased considerably. Unfortunately the time required to learn about the actual performance and perception of this new technology has not been reduced at equal pace (seeFig. 4).

Although currently many companies are able to develop products such as high-end consumer products and complex storage products (optical and magnetic) in timeframes of less than half a year the main paradigm used for feedback based on learning and managing quality and reliability is based on class 3 and class 4 failures. In order to manage random failures in components (class 3 failures) by metrics such as Mean Time To Failure, component failure or field call rate [8] are used. For class 4 failures metrics like average operating life (AOL) are very common. Not only it requires a comparatively long time to get this data, given a certain population of products and a corresponding market penetration rate, but these metrics are also fundamentally incompatible as a figure of merit for class 1 or 2 failures. Therefore substantial research will be required to develop metrics that are adequate for class 1 and 2 failures. A prerequisite for these metrics is also that the time to acquire these metrics should clearly fit within the timeframe of current and future product development processes. Research in this area will not only deal with the metrics themselves but also with methods to (fast and efficiently) generate reliability data (during the process and/or at the end of the process), gather reliability data, process reliability data

Fig. 4. Development time versus feedback time for high-volume consumer electronics[9].

7_{In this document speed is defined as the time between the moment a} product is released to the market and the moment adequate information is available to understand causes of reliability problems in this product to the level where they can be avoided in future products.

8_{In this document granularity of feedback is defined as the level of detail} that is available on reliability problems. The level of details is defined as the amount of information required to determine the cause of a failures in terms of the earlier mentioned parameters failure type (what was the technical nature of the failure (physical, functional) and what was its cause), time (when did the failure happen) and statistics (at which customer(s) did the failure happen).

A.C. Brombacher et al. / Reliability Engineering and System Safety 88 (2005) 137–146 144

Figure 1.3: Development and feedback times [13]

As seen in Figure 1.3, the development time of consumer electronics products is around 6 to 9 months, whereas the feedback time is still over a year for obtaining statistically valid and consolidated information from the field. As a result, feedback from the first generation products is only available for the third generation products (Figure 1.4).

Fig. 3. Information transfer in an industrial product roadmap.

2.2. Lack of time

(a) The total time from product idea till market release has to be very short [10]. At the same time products with a high degree of innovation have to be o!ered in order to be competitive. This implies that there is a need for reliability testing, but the tests must be done in a limited amount of calendar time. In principal testing a lot of products at the same time can do just this. But because the product is still under design, the production process is not yet ready and therefore the test products have to be made by hand what makes the products ex-tremely expensive. The "nal result is that there is pressure to skip tests, which may lead to considerable delays and high costs later in the development process.

(b) Because of the high innovation speed and the short development times, often products are in production not much longer than half a year. By the time "eld data is available, the product is already out of production. Looking from the viewpoint of product quality, one of the most unpleasant consequences of the time pressure is that developers are developing a new genera-tion of products without knowing whether the quality of the present generation is satisfying (Fig. 3). The problem will be clear: to survive it is necessary to create products of the right quality, but the information that is needed is di$cult to get in time. The high innovation speed has as a consequence that the product is

no longer in production by the time "eld in-formation comes available. If "eld inin-formation on a product of generation 1 is collected during the production of a product of generation 2, then the corrective actions can be applied in generation 3 and only to a limited extent in generation 2. Preventive actions can be applied in generation 4 and only to a limited extent in generation 3.

2.3. Lack of reliability data and models

(a) As mentioned before, products with a high degree of innovation have to be o!ered. This implies that there is a need for reliability calcu-lations and hence for the right reliability mod-els. As Ascher and Feingold [11] noticed, most reliability calculations are not realistic among other things because they are based on con-stant hazard rates. Not much has changed since then. This is becoming more and more relevant from a customers perspective as, espe-cially, for the customer the early phases of a product's life (where constant hazard rates are especially invalid) are the most important phases.

(b) Apart from the argument given in Section 2.2 there is another reason why "eld data is di$-cult to get, namely: service centres are not geared to the collection of failure data. During the warranty period the manufacturer has to pay the service centre for repairs, and this gives

40 P.C. Sander, A.C. Brombacher / Int. J. Production Economics 67 (2000) 37}52

(27)

4 Chapter 1. Foundations for Soft Reliability in New Product Development

• Dispersed Businesses

The increasingly global economy drives the collaborating business units of companies further apart, due to organizational costs-saving schemes (e.g., out-sourcing) and eﬃciency reasons (e.g., local repair centers), leading also to distributed markets. Consequently, the communication overhead among the distributed business units grows, and cultural or contextual use diﬀerences among the distributed markets surface.

78"

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

Figure 1.5: The global activities of a business group [22]

Business processes become increasingly complex, disintegrating value chains and outsourcing of certain NPD activities. An effect is that increasing demands are put on especially the quality and reliability related information flows; e.g., unexpected events occurring at different locations in the business chain need to be communicated not only among different domains or disciplines, but also via different countries and companies that all have individual cultures. In Fig-ure 1.5, the global spreading of the activities of a business group in a global consumer electronics company is shown. In this figure, the parts of the busi-ness chain that are managed on corporate level or outsourced to a third party (e.g., research, assembly factories, consumer care centers, service centers, re-pair centers) is not shown, which would complicate the picture even more. On a relevant side-note, while outsourcing the service and repair activities serves efficiency reasons (i.e., consumers can get their serviced products back as soon as possible), there is less focus on gathering root-cause information for product improvement. As a result, the disintegration of the chain significantly impacts the complexity of information networks, leading to loss of data integrity, delay in information flows, and loss of relevant information.

The increasingly global economy also entails a global market in which there are various diﬀerences amongst the needs and expectations of consumers from

(28)

diﬀerent regions. Also, due to the increased interconnectivity of today’s prod-ucts, there are many unknowns about the contextual use factors due to other systems and products (e.g., diﬀerent brands) expected to be used together. Especially for new products and technologies, where high technological uncer-tainty (i.e., regarding their field performance) is involved and where standards are not yet available, NPD processes are challenged to manage also other uncertainties with regard to meeting various regional market needs and expec-tations (i.e., market uncertainty), as well as, anticipating on manufacturing capabilities of a global range of suppliers, (i.e., industrial uncertainty).

• Extended Warranty Policies

The expectations of users from purchased products increase, while their tol-erance for perceived failures decrease, due to extended warranty coverage and duration. Consequently, the threshold for complaining or seeking instructional help has lowered.

!"#"$%&'"()*%+*,*!"-./(*0(,$1-.-*2%3"$*+%4*5%(-6'"4*5%'&$,.()-*

78*

4"-")9* &4%36:)* :%'&%("()-* ,4"* .(:%'&,).;$"9* &4%36:)* <,-* '.--.(/* +",)64"9* &4%36:)* <,-* 31-+6(:).%(,$*+",)64"9*&4%36:)*+,$$-*,&,4)*-%%(9*&4%36:)*3.++.:6$)*)%*,::"--*=364.(/*6(&,:>.(/9* 6-"*%4*-"4#.:"?@**

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

* H./64"*IJK*J*A4"(3-*.(*B,44,()1*:%#"4,/"*=;,-"3*%(*E"43"(*7FFF?*

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

Figure 1.6: Extension of warranty coverages (adapted from [6])

In the past, consumers were allowed to return products only if the product violated its technical specifications within the warranty period of usually 1 year (Figure 1.6). Currently, for many products there is an extended warranty period of up to 3 years. Region specific return policies are eventually becoming more flexible, e.g., in the USA ‘no questions asked’ policies are common, and in the EU, consumers’ rights are increasing with European Directives that indicate that warranty has to be granted in certain cases even outside the contract or product specifications [25]. The trend of extended warranty periods and coverage is expected to ensue [22].

The decreasing tolerance of consumers for quality and reliability problems with products can be attributed to their understanding with what can go wrong with the underlying systems. According to [4], a product of high quality in the eyes of the consumer is supposed to prevent a broad scope of problems such as “product does not provide suﬃcient information for use,” “product

(29)

does not provide customer with suﬃcient control,” “product needs to be con-stantly reset,” “product components are incompatible,” “product has a missing feature,” “product has a dysfunctional feature,” “product falls apart soon,” “product diﬃcult to access (during unpacking, use, or service)”.

As a consequence of all four trends, today’s consumers reject a product mostly on the basis that it does not comply with their expectations in some way or that it is unfit for their use-contexts, even if it complies with its technical specifications. While products that violate their technical specifications can be treated by the manufacturer for hard failure recovery, the remaining soft failures cannot be acted upon for product improvement due to the lack of timely and relevant use-context information. Therefore, while hard reliability of products can be managed with reference to technical specifications, soft reliability of products cannot be managed with reference to consumer feedback from the field.

Furthermore, soft reliability of products is no longer limited only to usability con-cerns, but more importantly also to so-called beyond usability concerns. Another name for this new line of research is user experience (UX) research. While tradi-tional usability research is about preventing usability problems, UX research aims for achieving more than just designing for the absence of problems, through de-signing for pleasure and hence creating outstanding quality experiences as a result [46]. This shift relates to the observation that, while usability was once seen as a “satisfier,” it is now seen as a “dissatisfier” as referred to in the marketing domain. That is, the presence of usability does not anymore add to customer satisfaction, but the lack thereof leads to customer dissatisfaction.

1.2 “No Fault Found”

Misalignments between product capabilities and user preferences damage the over-all success of a product in the market. Especiover-ally in the past few years, these misalignments increasingly lead to users rejecting or returning products after pur-chase. However, technical analyses of such products show that these products fully meet their technical specifications. This is particularly the case with highly in-novative products that bear considerable market uncertainty during their develop-ment. A wide-spreading industrial designation of such events is known as No Fault Found (NFF). NFF cases have first been recognized explicitly within modern high-volume consumer electronics industry (Figure 1.7) and more recently within the mobile phone industry: In 2006, NFF returns cost the global mobile industry $4.5 billion [80]. In 2007, NFF processing costs, only in Europe and USA, were about

Soft reliability in new product development : an ontological approach for utilizing field feedback to dynamically sense and adapt to evolving global markets

Soft reliability in new product development : an ontological

approach for utilizing field feedback to dynamically sense and

adapt to evolving global markets

An Ontological Approach for Utilizing Field Feedback to

Dynamically Sense and Adapt to Evolving Global Markets

An Ontological Approach for Utilizing Field Feedback to

Dynamically Sense and Adapt to Evolving Global Markets

PROEFSCHRIFT

Aylin Koca

Acknowledgements

Abstract

Soft Reliability in New Product Development

Contents

List of Abbreviations

Chapter 1

Foundations for Soft Reliability in

New Product Development

1.1

Emerging Challenges in New Product

Devel-opment

1.2

“No Fault Found”