Chapter 4 - Methodological approach and research journey

In this chapter, we present the methodology deployed over the three years of the PhD thesis. The methodology relies on an intervention-research approach (Hatchuel and David 2008) facilitated by the CIFRE contract (Convention Industrielle de Formation par la REcherche - Industrial Convention of Training through Research).

Principles of intervention research in management

The Intervention Research in Management (IRM) has been elaborated over several decades of research partnerships at Centre de Gestion Scientifique (Hatchuel and David 2008).This approach has proven efficient in revealing the intricacies of complex managerial situations specially in the innovation management field with the multiplicity and novelty of interactions, decisions, knowledge flows in project and strategy management. IRM relies on two pillars: a collaborative protocol and theoretical perspective.

The aim is not to solve a definite problem proposed by an organization. The theoretical setting can actually be rather dull at the beginning despite bearing a certain potential. Nonetheless, a dialectic between established-theories-in use and contextual-established-in-use allows improving models of action. The point is nor to perform change management, but rather to gain an understanding of theories-in-use, their performativity and how models of thought can evolve within the business. This continuous monitoring between academia and industry through the PhD candidate enables uncovering a real research issue supported by:

• Pre-conditions

  • partner with a pioneering logic

  • open management issue free interviews

  • contractual commitment to a research issue

  • support of an academic team with research potential (PhD directors/supervisors and research lab)

• Monitoring principles

  • Free interviews

  • Warranted isonomy¹

  • Confidentiality of all individual interviews

  • Capacity to create new empirical material

  • Controlled design

  • Management innovations evaluated as rational myths

Given this set of conditions and practices, the outputs emerge naturally following the dynamic and symmetric feedback supported by the PhD candidate through his meetings, steering committees with academia and industrial colleagues. It implies testing models of thought, frameworks drawn from the literature, and models grounded on field data based on the case studies conducted.

Initial management issue formulation

Originally, the PhD research program was designed to address the topic of economic performance of design at Airbus Aircraft. At that time, the PhD candidate was already working within Zodiac Aerospace.² The program discussed how design activities of project and program management could be valued by alternative means to project objectives. For instance, it raised that traditional learning mechanisms between projects were hard to track such as spill-overs (Cohen and Levinthal 1989), and how learning could be managed from a project to another in an efficient manner. The performance dimension, its definition and underlying value management was central for the topic. Today, as the PhD project was not launched in July 2015, it was put on hold and is currently being dealt by a PhD fellow at CGS (Agathe Gillain).

As this PhD subject would not come to life in its original form by 2015, we proposed to reallocate the project to Zodiac Aerospace. With the support of the Group HR management, a new sponsor was to be found: Thierry Rouge Carrassat, Chief Technical and Innovation Officer for Zodiac Aerospace. Several meetings were organized between the PhD Directors and the CTO to provide information on the company, its history, its context, innovation management practices and how the PhD program would be framed by IRM methodology.

The topic was adapted, it discarded part of the economic performance axis: the topic recentred around the old dilemma of market-pull vs. techno-push. ZA presented several cases trying to overcome the growing rationalization of market-pull by aircraft manufacturers and airlines. For instance, they had two development teams close enough to market prescribers in order to try to anticipate market-pull trends and formulate proposal of new products/services that could federate several BUs for more added value.

Furthermore, as business units were highly profitable in their respective niche-markets covered over the history of acquisitions, corporate management had been sponsoring techno-push initiatives. These were steered through a Multi-BU committee showcasing a mix of top-down and bottom-up approaches.

Another salient feature was the history of acquisitions and expansion since the establishment of the company in 1896. Zodiac Aerospace was a conglomerate of SMEs with individual responsibility for each BUs grouped based on market segmentation and technology proximity into so-called: business lines, divisions and branches. For further details on the how the PhD program was formulated, you can refer to the Appendix. It is a clean copy of the submission made to the ANRT (Agence Nationale pour la Recherche et la Technologie) - the agency sponsoring the CIFRE PhD Programs nationwide.

Phases of IRM

Intervention research can be split into different phases that do not necessarily represent a linear story in the research journey. The phases are the following: feeling of discomfort, rational myth, intervention and research, portraying a set of logics and change process. Starting with this feeling of discomfort with respect to the open management issues identified, the researcher in an abductive fashion starts generating hypothesis on raw data collected through open interviews with a wide variety of actors across business units.

The investigation is followed by the definition and gradual elaboration of a rational myth. This myth reflects as a potential model, and reference to rely on to explain and coordinate collective action. As exposed by Bachelard in his thesis “Essai sur la connaissance approchée” (Essay on approximated knowledge) (Bachelard 1927) to explain scientific knowledge development and constructivist epistemology:

The order of sensibilities does not necessarily follow the salient features, what we could call the dynamic topography of our experience. We indeed have little agreement with the simplification of the prior phenomenon of every scientific development. We think the primordial features stand out by themselves and establish themselves forcibly to our study through their generality. Nevertheless, sometimes it is the accident and not the general that is explanation principle to the point where problems should be overturned to restore the generality to the phenomenal features the spirit had represented in special meaning.³

It should not be mistaken for serendipitous discoveries but his work stresses the importance of experimentation, seen as the contact between Thought and Reality which is the only way of specifying the mode of existence. The knowledge is thus approximated by a movement of conjunction from Thought to Reality, which are supported by means of intervention and interaction. The gradual and often non-linear shaping of knowledge allows the researcher to portray a set of logics. Up to this phasing, the purpose of the researcher is mainly to explain and bring to the surface the rationality mobilized in the collective action. In that regard, we adopt an ordinary perspective of rationality (Boudon 2012). We avoid the potential patchwork and associated burst of the notion of rationality, and stunning relativism that social sciences can sometimes produce. Of course, we cannot pretend to discard existing theories that explain collective action through according to different research programs (Lakatos 1980): utility, materiality, mechanistic or cognitive.

The methodological individualism required to reveal the set of logics, and rational myth, tends to adopt a cognitive approach to understand social phenomena which are the implications of the individual actions inspired by reasons. It is a means of avoiding pitfalls as explained by R.Boudon (Boudon 2012): solipsism, psychologism, procedural knowledge and programmatic approaches, dispositional variables, restrictive axiomatic such as rational choice theory and expressive rationality (Frega 2010). So, we tend to look for how these reasons are shaped within and without context. We also identify the research program Max Weber’s proposition in his treatise on the Sociology of Religion (Weber 1946, 280):

Not ideas, but material and ideal interests directly govern men’s conduct. Yet very frequently the ‘world-images’ that have been created by ‘ideas’ have, like switchmen, determined the tracks along which action has been pushed by the dynamic of interest

The researcher in his generative search will then test his evolving understanding with the field through different intervention and interaction devices, acting as measuring devices and testing: interviews, questions with given semantics, presentations of models, steering committees, discussions with supervisors. And finally, there is a need for predictability in the new understanding built by the researcher which comes in the form of a change process. This change occurs mainly from the understanding of Reality of the studied domain. It does not necessarily imply the implementation of new management tool, or rationality organizing collective action, but rather a learning process due to the scientific understanding. We will come back to this in the following section, as we propose a methodology of anomaly detection.

Data collection

The vast majority of data was collected in the group Zodiac Aerospace through the variety of interactions with actors of the firm. Given the nature of our research, a diverse sample of data type was sourced across the research ground.

From the beginning of the PhD journey until the end, a large sample of resources: meeting minutes, presentations, TRL assessments, technology/product roadmaps, project statements, project & program documentation stored on local networks, intranet, etc.

Another set of data was gathered through active participation to workshop, seminars, meetings where sometimes where the researcher was invited for observation and we must add data collection with varying degrees of intervention: writing workshops, working groups, interviews, project meetings and seminars.

Anomaly detection and modelling

As we will present in the following sections and following chapters ( Chapter 5 ) we have a small-sample research contained in a large conglomerate of SMEs. The different phases of IRM tend to overlap and provide feedback between themselves. Overall, we have the three main categories of logics interacting with each other(Dana and Dumez 2015):

1. Deduction

2. Inference

3. Abduction

Given the sample size and in order to provide meaningful results for academia and practice, demonstrating a case of anomaly is key from a statistical standpoint (Siggelkow 2007). The purpose is to carefully observe a counter-example to the models extracted from literature (see Part 3), then propose a new model explaining the phenomenon and extending existing theories (see Part 4).

In this section, we present how we mobilize the idea of comprehensive research (Dana and Dumez 2015; Dumez 2016) to elicit counter-examples of existing models in the literature and design a new one. There are three main risks associated highlighted by the method:

1. Abstract actors or being of reasons

2. Circularity

3. Equifinality

These three major risks will serve as a baseline to promote counter-factual reasoning with the elucidated anomalies by putting the effort on creating a typology, stress mechanisms between levels of analysis through a set unit of analysis, and proposition of a new concept redefining and extending existing theories.

Detecting an anomaly in management

One of the movements of a comprehensive research (Dana and Dumez 2015) can be seen as a modulation of deductive research. The selected literature and associated models tell us what to observe when scouting the research ground, with a risk of being trapped with circularity. The first step consisted in analysing theoretical frameworks relating to ambidexterity, exploration project management, strategy management and organizational design/studies in face of the unknown. It is also a way of rephrasing the elaboration of the rational myth and set of logics portrait. Based on the literature review (see Part 1), we were able to select several of these frameworks which are key to identify a phenomenon unknown to academic literature in contrast with the synthesised model of non-mutual conditioning of exploration/exploitation.

The analysis requires to extract typologies and mechanisms of phenomena observed in the selected literature that shed light on the anomaly. The descriptors are introduced to stress issues and limitations already identified in the literature. It will also facilitate the instrumentation to describe the purpose of our research questions, hence the anomalies and their implications (see Part 3).

We have selected the common descriptors: model of coordination and collective action, innovation potential of attraction (generative processes, environment cognition and organization design). The literature review aimed at stress testing major models of ambidexterity in order to see that the construct had drifted overtime and locally delaminated as innovation management had evolve to embrace the unknown specially with exploration project management. These were analysed with inputs from design theories and reasoning embodying generative processes. Several assumptions and limitations were identified with the help of the descriptors as presented in the research questions (see chapter 3.

Data analysis

We must mobilize a unit of analysis that opposes inference and abductive logics. The first calls for explanations, typologies and mechanisms the selected literature will allow us to observe. The second is more independent from the literature and tends to extract what the research ground can “tell us” through the variations observations and interactions/interventions. The two are far from being independent in practice, there is a constant confrontation bore by the researcher himself echoing a declination of the Hawthorne Effect.

The two different sets of data allow us to confront the two logics separately over the collected documentation and the variety of field notes (interviews, meetings and workshops participation). The unit of analysis is crucial to elicit the articulation between the levels of analysis (from the individual to the organization/ecosystem) and associated mechanisms evolving through the timeline of the case studies.

So, we would tend to have a ‘mesoscopic’ approach as the typologies drawn from the literature review, in association with event-system theory invite us to play on different levels and temporalities (Griffin 1993; Griffin and Ragin 1994; Morgeson, Mitchell, and Liu 2015). The other benefit is that it forces the researcher to disentangle black-boxes that may have been contoured in the literature such as abstract actors and beings of reason.

We propose then to explicit the selected descriptors (see chapter 3 over time as the decision-making and action is engaged by actors (designers, engineers, managers, etc.), within projects and within committees. We then suggest to look at the time-sequence of the following derived descriptors:

• decision/problem space design: decision and problem formulation is extracted from documentation and conducted interviews for the different surveyed selected projects. The input from design theory from a methodological standpoint also helps to envision a large decision/problem space which was discussed with interviewees (as the possible course of action).

• actual made decisions: recorded from project history, documentation and interviews

• engaged course of action: recorded from project history, documentation and interviews

• organization design: supporting (temporary) organizations were found in project documentations and interviews and also directly observed.

As an example, if we propose to represent a project management for a standardized product development, the descriptors allow us to show the linear and supposed stability that is adapted through risk management procedures. The figure 1 below gives an idea of how this would be done.

This analytical lens remains very useful to draw the project trajectory, its coordination mechanisms and collective action. However, we must stress that it may also become chaotic as real life management, and interaction with the environment and organizations is not without obstacles. So we must highlight these obstacles and specify what the literature models of non-mutual conditioning between exploration and exploitation would have predicted compared to the actual course action depicted with our descriptors.

Modelling: Naming, explaining and testing

Finally, the anomaly detection encourages the researcher to formalize a new model mitigating some perspectives addressed by the research questions and the observed phenomena. The novelty invites naming the phenomenon based on traces left by academics or practitioners to weave a model from it.

The new model is introduced in Part 4, it should be free-standing: plausible on its own due to the showcased logics of the conceptual argument (Siggelkow 2007, 21). Despite being backed by the cases, it also requires elevating the model from the idiosyncrasy of the cases themselves.

Our model will support a novel explanation of the anomalies presented in the chapters 8 and 9 in Part 4.

The validity of the model is presented in chapter [10]https://mleglatin.github.io/phd/part4/chapter10/) and in Part 5 through the different ways in which the researcher intervened in the research ground: contributions to projects and management procedures. For IRM phasing, it corresponds to the establishment of a new rational myth proposing a new way of organizing collective action. The support and reaction to this experimentation is detailed in the two chapters where the model was experimented and validated by peers. Moreover, this experimentation of the model validates the status of the model as a building block providing an understanding of complex mechanisms, laying a path along the research questions perspectives for a future theory with exploratory hypotheses in Part 6.

Now, we have specified the different descriptors to detect the potential anomalies, we need to be able to compare models derived from literature with the case studies. We propose then in the following paragraphs the adaptive, interactive and encapsulated models. They represent different models of ambidexterity as we have seen throughout our literature review along the chosen descriptors.

The adaptive model is mainly concerned with responding to the environment by organizing the firm in a way that it is a priori capable to answer to potential changes. The generative processes such as radical innovation are then structured by the organization design and environment cognition.

The interactive model tends to deepen the contingencies of generative processes by stressing the importance of relationships, emergent new ones so that new routines can be recombined through the open-ended network dynamics. The organization design are shaped to the complexity of relationships tied through generative interactions. The encapsulated model brings a more ‘organic’ view by framing generativity with the scope of a project supported by a temporary organization also tied functionally to resource-providing organizations. The change dynamics of the (temporary) organization are the mirror’s image of the net of projects engaged with the environment and program office/governance’s strategic input.

Adaptive model: coping with the environment’s dynamics

In the adaptive model, what is critical is the way in which the adaptive process rooted in problem-solving, and by-extension decision-making (probabilities on solutions), moved from organizational learning with a concern for exploration and exploitation regime to being a key construct to think the firm’s innovation performance in addition to having competitive advantage. Mutations of the organization structure and process are the answer to environment changes. Variation and selection should be then managed to ensure an efficient exploration. The other salient feature is that it tends to prescribe organization design with different level of analysis that should be the responsibility of top management as they are considered among the few to bring system thinking and thus conduct strategy. The underlying foundations such as decision-making and dynamic capabilities building remain quite obscure (Birkinshaw and Gupta 2013) and may lead to complicated managerial action.

Interactive model: enabling an ecology of generativity

The interactive model approaches the issue in a far more contingent way. The circulation of concepts and interactions with artefacts contribute to the exploration regime and simultaneously to the exploitation effort as stakeholders take part in the process. The decision-making has a tendency to be left on the back seat, as the engaging action with others will allow addressing uncertainties and the unknown. Yet, organizing open innovation, or freeing innovation raises the question of problem formulation and deciding of its decomposition in the network of interactions. The organization boundaries can be easily blurred by the numerous engagement of agents and artefacts with the environment. It naturally reveals numerous question on the governance, if not control, of such fractal organizations as the identity and purpose of the products being developed could bring heterogeneity and incoherence.

Encapsulated model: challenging organization design and learning

The encapsulation of generative process into projects and the management of a portfolio with a program elevate the interactive perspective by putting the emphasis on strategic management, an enclose collective action, as well as a potential issue of governance and temporary organizations.

The temporary organization supporting the project-based management also relies on functional departments providing key resources and knowledge management. The organizational learning and organization ties of projects floating across these organizations can induce difficulties that encourage to think of change management with the scope of a given project, instead of a separate activity.

Research Journey

In this section, we present the research journey over the three years corresponding to the IRM phasing and its modalities. The researcher took part in several Steering Committees with the firm and PhD directors, as well as numerous Academic confrontations which have participated to the evolution of perception of literature frameworks and gradual specification of a new rational myth. The figure below gives a global picture of the journey over several dimensions:

Steering committees

In order to elaborate the different phases of IRM, several steering committees were organized on a quarterly basis with the academic and industrial supervisors, in addition to weekly interactions with both separately. The first set of meetings gave a frame for an open discussion (isonomy) around a proposed model of action, framework projected over the data collected on several case studies. Presentations and discussions allowed testing hypothesis, different narratives, and semantics taken from literature models. Two major meetings were organized with the CTO, R&T, engineering and business development managers, in addition to PhD Directors. Otherwise 10 other steering committees were held with the CTO and PhD supervisors to keep track of the open management issue. The second set of meetings were perhaps less formal as the challenge was to update PhD supervisors as frequently as possible on the collected data and analysis made. In a more personal way, it was the occasion for the research to test extreme hypothesis or simple extrapolations in order to explore the boundaries of the domain of the models taken from the literature or grounded in the data. Practically, it implied testing narratives, semantics from different literature streams to test how the supervisors would project themselves in the understanding built by the researcher. It is a dual learning process for both sides.

Academic confrontations

The research had the occasion to participate to a range of academic activities: thesis presentation at the laboratory, presentation for the annual PhD candidates meetings of the CNRS research unit (UMR 9237), presentation at seminars, conferences and journal submissions with peer reviews.

During the three years, the researcher engaged discussions with several researchers to learn more from their publications and research interests. It was the occasion to present the research project, results of case studies and trying to find echoes with their works. The table below offers an overview in a chronological order:

Evolution of the rational myth contributors

In this subsection, we propose to cover the literature journey of the researcher. We briefly present how the modelling effort came around over which literature was mobilized, in addition which articles were presented at conferences and sent to journals. The elaboration of the rational myth and set of logics portrait is not a linear process. So the following paragraphs are an ex-post viewpoint of the PhD journey. We have taken the liberty to make three categories for which articles were written and presentations made on various occasions to test literature models on case studies.

As the reader will see, they do not follow the literature review chapters 1 and 2, instead they follow the PhD candidate through the three years and how he tested several frameworks to engage with academia. The researcher made a first effort regarding the literature around the construct of “decision-making” “design thinking” before going into the field “organizational ambidexterity” and ramifications.

Decision-making

Through the case study at ZSSM, we had the opportunity to study the decision theories in order to explain the project management. Data was presented in a way that it encouraged the decision-making study: the story was told as sequences of events with commitment to decisions, beliefs and their representation in several documents.

Rational theories of choice

As we will see later, we proposed to test the validity of rational models of choice, where rationality is seen as instrumental(Boudon 2012). Large reviews of decision theories were executed: from the ones based on maximum expected utility to modulations made between the 1950s to early 1990s with models discussing the underlying axiomatic and formulation mathematics. The works of Allais, Kahnmenan, Tversky, Machina, Camerer, Giocolli, Quiggin, G.Klein, Slovic, Regenwetter and Gilboa were paramount in the field.

Debates on rationality, coherence between decision, beliefs and action were covered to understand how the construct of decision-making was considered in different literature domains ranging from economics, management to philosophy. It was also the occasion to make a detour through mathematics with decision models using multi-criteria approaches and imprecise probabilities.

Irrationality in decision-making

The question of not being able to expect, changing the notion of utility to deal with the way decisions are made, led the researcher to dive into the way what makes irrationality and how the decisions are constructed. So the perspective of ex-post construction of decision to justify the course of action, and deviance according instrumental rationality was addressed.

Yet, as we were looking to bridge a gap with design theories, considered as a change of paradigm, we were looking for ways in which rationality could be preserved and still extend decision-making with design theories without falling in the trap of leaving model-out-of-domain phenomena to dispositional variables such as irrationality or constructs determined by ad-hoc phenomena such as Bourdieu’s habitus. One could refer to the books of Christophe Morel on absurd decisions, and the works Nils Brunson for their perspective on irrationality and absurdity. However, the researcher was more inspired by the works of Richard Bronk on the importance of imagination and creativity in markets (Bronk 2009) as a means to propose extensions and deep revisions of outdated models of Thought explaining Reality.

Strategy management and politics of decision-making

Similarly, studying the role of strategy-making, strategy-as-practice, encouraged the researcher to find ways of formalizing decisions in a different way. The works of Crozier, Fredberg, Pettigrew, Sfez were key to test other frameworks were the role-playing is mobilized. The researcher discussed with Laurent Simon, who referred also to Laure Cabantous for further debates on decision-making and emergent strategies. The difficulty for the researcher was to make sense of the observations. Some of decisions taken were not coherent with the course of action or beliefs. Violations of rationality axiomatic could be highlighted but as the outcomes were mitigated and questioned utility and performance, a different model was to be found to add details to known phenomena or substantiate anomalies in a novel model. A turning point was reached with a detour through the literature of routines and organizational ambidexterity. Indeed, stimulating papers with a sociological perspective written by Fanny Simon and Albéric Tellier with the event system theory of Frederick Morgeson and event-structure analysis of Larry Griffin.

Design Thinking

As two case studies were based on the use of Design Thinking (DT) methodology, in-depth literature review was conducted to understand its origins, evolution and complementarity with other design perspectives in literature: design studies, design theories and creativity.

Design Thinking in organizations

The place of a creativity methods, supporting the Fuzzy Front End and a subsequent efficient NPD, has been largely studied in the past decades starting with creative problem solving, brainstorming, and more recently with Design Thinking methodology. Since it came through practice mainly, several academics starting in d.schools (Stanford and Potsdam) but also Swedish researcher centres, worked towards positioning DT with respect to engineering, creativity and innovation management and organization studies also.

However, as design in general is usually seen as a confined activity confined to engineering, or perhaps strategy, or even organizational design, several barriers were faced when trying to framing and valuing DT as part of an ecology of routines in the organization.

Design Theory and engineering

From the perspective of design theory, design science, and engineering, reaching out for their impact of the organization was not a easy task. But, as the field appeared more structured, it helped as reference to position DT and try to reach out for organizational issues.

Design-Oriented Organization

In the tradition of works of design theories and the importance of management devices in organization, embracing post-Simonian paradigm, the works of design-oriented organizations for innovative capabilities (Andy Dong, Armand Hatchuel). They place design as a generalized activity taking place at all levels of the organized, hence requiring to be carefully managed to sustain firm’s regeneration.

This literature complements existing literature on dynamic capabilities and organizing for radical innovation (O’Connor, Dougherty, Leonard) but also the contributing organizational routines needed to support these: knowledge management, valuation, power and legitimacy.

Organizational ambidexterity

As mentioned earlier, in the same way that the duality between market-pull and techno-push was first addressed by the research project, the researcher inquired on the model of ambidexterity. The motivation for such topic started when trying to formalize a portfolio of case studies, and projects observed at the intra/inter-BU level.

Organizational perspective

These projects were hard to label in the organizations. The belonging and having appropriate management tools appeared critical to interviewees and attendees of the PhD steering committees. They acknowledged that projects all along where not fully exploration nor fully exploitation. Consequently, the researcher looked for literature items that would touch upon this grey zone contributing to strategy and competitive advantage. Luckily, in recent years (2013-2015), several authors such as Tushmann, Birkinshaw, O’Reilly pointed towards the necessity to study decision-making and managerial capabilities to better understand the translation of ambidexterity at a micro-level, beyond top management and environment perspectives.

Exploration vs. exploitation of routines

Since ambidexterity can be approached in two main different ways. Besides the organizational one, the other stresses the nature of routines and they should be coordinated in a global way for organizational learning. In 1991, March’s proposes a more behavioural approach by considering problem-solving, theories of choice, evolutionary theory and information-processing. This echoes the decision-making issues mentioned earlier with an a priori irrationality. The cases encouraged to conceptualize the possibility of switching between modes, but also to find a middle ground avoiding the dichotomy. In other words, a new kind of rationality where absurdities and anomalies can be resolved, hence blurring the artificial separation between exploration and exploitation or even forcing them to be over a same continuum.

Project Management

Finally, during the last stretch of the PhD journey, we reached out to project management literature. Recent research on exploration project management and the role of the unknown in such situation, have amended the literature with new ways of addressing the tension between exploration and exploitation. The works of Loch and Lenfle are now a cornerstone in the field. Since the organizational studies were key for the research, naturally we were drawn towards the role of matrix organization, adhocracies, and more generally temporary-based-organization in the field of project management. It was a means to loop back with organizational design and by which means the organization supports these projects: resources, routines and dynamic capabilities.

Paper writing and case studies

The several discussions, debates and literature reviews were key to formalise reasoning into several papers. These are summarized and reformulated here in the manuscript and can be extensively consulted in Publications. The associated case studies are briefly presented in the following chapter.

In the previous methodological sections, we have explained the intent to demonstrate an anomaly. Three papers were written for this common objective with different streams of literature. They are based on three major cases representing the variety of innovation management. These are presented in the first part of table in the followin chapter.

Then a theoretical paper offers a model to reconcile with the anomaly. Compared to published version (see publication, we propose in this manuscript an extension that is then discussed in the last paper of this thesis (see second part of table in following chapter).

Role and position of the researcher - Intervention Research

In this section, we explain what were the activities and relations of the researcher but also employee of the firm. We explain the mission of the Innovation Direction and how the researcher took part in it. We also cover daily activities, main drivers of integration with respect to the management issue at stake in the research project, and peculiarities.

Being part of the Innovation Direction

The Innovation Direction was created in 2013, following the creation of the Zodiac Scientific and Technical Council (ZSTC) by Thierry Rouge-Carrassat. He had been in charge of engineering departments and the general management of business units. In parallel, several discussions with the Group Human Resources benchmarked how Research & Technology or Research & Development was organized. Among the surveys, it was decided to establish an “Experts Career” program that would value technical careers besides the traditional management one and the program management (ZA-H-5003 Expert Technical Career Path Standard). This dual ladder model (Allen and Katz 1986) was largely inspired by the car equipment manufacturer Valéo. Thierry Rouge-Carrassat had also previously established a dedicated R&T within business unit.

The CEO, Olivier Zarrouati had decided to create a Group Technical & Innovation Officer, and Thierry was offered the opportunity. With the support of a consulting firm, management tools were introduced to offer BUs means of managing the nascent dedicated R&T activity. Previously, these were part of more traditional engineering activities designing products for client demands, but induced high risk. A separate process and structure was proposed to mitigate risks between an upstream/exploratory set of activities compared to more downstream/exploitative ones taking place in client programs (new product development).

The management tools consisted in what is almost a standard for aeronautics after having being one for the Department of Defence in the United States of America: Technology Readiness Level and Technology Road maps (Mankins 2009). Airbus and Boeing, in addition to the institutions such as public funding schemes, certification authorities have been in the past decades gradually introducing the concept of technology readiness assessment in 9 levels. Some discrepancies between scales can sometimes be observed such as Boeing’s being longer one (see Boeing’s procedure PRO-5157 - Technology and Application Readiness Criteria (Issue March 21, 2007)).

Mission definition

The researcher arrived in late July 2015 in the newly created team: a CTO delegate for the Americas from the start in early 2014, the Director of Intellectual Property had been recruited in October 2014 and the Director of R&T European funded programs in May 2015. An additional member was delegated at the Institute for Technology Research in Toulouse (IRT Saint-Exupéry) in charge of the department of Electrical Systems. In the following months, an experienced project manager would be hired to take care of a national program for metal additive layer manufacturing, joined by a PhD candidate based in an university laboratory in Orléans (Ecole Nationale Supérieure de Mécanique Aéronautique).

The team was rather young in the group except for the CTO and the one based in Toulouse, we were then all discovering the intricacies of the group. We would share a lot of information and thoughts on what we would discover as we held meetings across business units. In December 14, 2015, team’s mission was defined, despite all being already being autonomous in their functions (project management, patent management, funding and R&T management support):

• Lead the development of a long term technological vision with associated tools & processes

• Manage efficiently the expertise and IP, develop relevant internal & external partnerships and secure funding to ensure a sustainable road map

• Lead the development of a long term technological vision with an associated sustainable road map

• Manage efficiently the expertise and IP, develop relevant internal & external partnerships and secure funding by applying required tools & processes

PhD Candidate in Innovation Management position

The researcher would report to the CTO as he would potentially benefit from the feedback from field work, analyses and model creation and testing. Bi-weekly meetings were organized to keep track of what the researcher was conducting, and potentially answer to his questions and test hypothesis.

On several occasions, the PhD Candidate took some of the CTO’s tasks. These were quite varied but all served the research project in some way or another. The table offers a glimpse of the different activities (see below).

First steps and gaining autonomy

From August to November 2015, the time was dedicated to gain acquaintance with several business units, understand the purpose of the Technical and Innovation Direction, as well as try to reveal potential open management issues for case studies supporting the research project.

Being introduced by the CTO

The first month was fully dedicated to open discussions with several managers (engineering, marketing, sales), technical experts and project managers part of the middle management of BUs across the group. Meetings were held in the corporate headquarters in Plaisir where several BU representatives have offices, as well as several other video-conferences.

The CTO had sent an email introducing the researcher’s PhD project to the ZSTC community, as well as key experts and managers who have a rather dense network within their business unit or across others. In total, 15 employees were consulted. Not all of them contributed equally to data collection. They were from 8 different business units all from middle management and Direction committees.

The researcher received a warm welcome from all interviewees who were curious in finding what he could bring to their daily concerns. As these first meetings were quite open, the setting was about having them tell stories about their work, their projects, difficulties, good practices and general management topics. They were in a favourable setting to disclose valuable signals feeding a feeling of discomfort guiding the rational myth sensing.

First intervention: TRL assessment tool

As the researcher was gradually being identified, recognized in the R&T community, the R&T Manager of Zodiac Seats France (ZSFR) reached out for support on the implementation of the TRL assessment, before reconsidering the TRM structuring. The request originated from a demand of funding scheme who steers carefully the progress of portfolio of projects. TRL targets had been given and several debates occurred on the understanding of their definitions, so they needed support on providing accuracy to definition baseline given in the ZA procedure and justify the TRL assessment through a proper tool.

Four workshops were organized with the support and input of the CTO for the first meetings. The researcher had benchmarked several business units beforehand following the round of open interviews, and other aeronautical firms for examples of TRL assessment use. Based on the latter, a draft created by the researcher served as a basis for discussions with R&T stakeholders at ZSFR to propose an update TRL assessment tool.

A by-product of this intervention, a new ZA procedure was re-issued to share the updated baseline definition with detailed tool to support other BUs. The TRL was seen as an evaluation and not necessarily associated with gates to be passed. Such perspective fed many debates on the maturity assessment of systems or designs since technologies would be integrated in a given product at some point. Thus, it also encouraged the revision of the complementary procedure on Technology Roadmap definition, creation and sustaining and its articulation with Product Roadmaps.

Subsequently, the relationship established with the business unit led to the identification of a case study on the engineering of a modular platform for business class seat (see case description).

Master courses for literature frameworks

In parallel to the Zodiac Aerospace integration, during the months of September-October, several courses were taken in the Design Engineering curriculum of Mines ParisTech organized by the chair of Design Theory and Methods for Innovation to confirm learnings on design theories, innovation management, epistemology of management sciences and organization studies.

These courses were taken in the perspective of gaining robustness over certain topics related to the feeling of discomfort originated from the open management issues faced during the first set of open interviews. It was also a means of trying different set of logics portraits to explain first batch of observations and further observations through intervention and interaction in the form of case studies.

Surveying BUs for topics

As explained briefly, a first intervention was the opportunity to identify a first case study at Zodiac Seats France on the engineering design of a modular platform of business class seat. The rest of case studies were mostly identified following the first interviews and validation with the CTO. He played an important role in challenging the interest of the case studies and serve as liaison with the managers at stake.

For instance, the case identified at Zodiac Sensing and Systems Management (ZSSM, former Intertechnique, and now called Zodiac Fuel and Control Systems) came around as the CTO proposed to study a case of a technology developed from scratch, above TRL6 and about to be integrated in a Boeing aircraft program. The intent was to study a case of success despite numerous difficulties along the way. A first interview was set with the R&T Manager. The history of the project was told, but the narrative did not reflect anything special besides collaboration issues with Boeing and unfavourable conjecture due to commercial litigation concerning other ZA business units. However, the manager had mentioned another potential case study he quickly presented. The researcher looked for feedback from the CTO and concurred to the idea that this other case would perhaps reveal more stimulating management issues than the bumpy road of the other project’s trajectory. Another interview was set to dig deeper and try to grasp the manager’s motivation (see case description).

In another case, the CTO introduced the researcher to the Airbus Development Team for on-going projects and invited him to several other meetings held by the Executive Committee to follow up on the concepts, projects and several debates related to the activity domain and associated innovation management issues (see case description).

The regular attendance to Executive Committee meetings for strategic technologies and product/service concepts (see case description of Multi-BU meetings) also led to the acquaintance of other managers who had a deep interest in understanding the performance and success of their ongoing projects (see case description)

The CTO’s guidance through bi-weekly meetings focused on discussing the researcher’s analysis, modelling, and collected data. These interactions were key for the research trajectory in addition to the autonomy of the researcher having built his relations/network across business units.

End of PhD and autonomy

During the second half of the PhD, as several case studies were ended but required further analysis and modelling, the researcher had already established regular contact with several R&T Managers or Director of Engineering. Naturally, the researcher was requested for his contributions to TRL assessment and TRM definition, as well as procedures written for the Zodiac Aerospace Operational System (ZAOS) (see table).

Nonetheless, the researcher was gradually acknowledged among these managers for his input on innovation management. For instance, the case study and intervention for Zodiac Oxygen Systems Europe (ZOSE, see section) was proposed by the R&T manager. He had started an internal program to anticipate evolution and disruption of pilot oxygen masks to complement the benchmark made by the marketing department. Creativity sessions were considered and the researcher assisted the R&T team to organize them and capitalize on the design efforts made as well as with the evaluation. In parallel, an internal open innovation challenge had been organized by the ZSTC, of which one topic concerned the comfort and functions associated to the pilot protection against the risks of hypoxia. Consequently, the researched helped transcribing the open innovation outcomes into an innovative design mapping based on C-K theory.

Furthermore, the initiated program was carried on and partially translated into an answer to a call for partners in the European funded program: Clean Sky 2. Airbus was leader of the topic reviewing the future pilot oxygen mask. A team was put in place at ZOSE and started designing concepts dealing with the ill-defined problem presented in the program. The researcher was invited again to help formalizing the efforts made and explain their design process. The challenge was to gradually explain how to address highly uncertain and unknown parameters in their design process and discussions with the principal partner/client (Airbus).

The researcher was called to provide internal training, recognized by HR department, for the R&T department of Zodiac Seats France. The training consisted in an introduction to design theories, design regimes and a workshop to practice. Another example, during the PhD writing, the researcher had the opportunity to contribute to an innovative design initiative at Zodiac Electrical Power Systems. Beyond the design issue, in the same way as in the ZOSE case, the researcher intervened and tested models based on his research, as part of the process of testing the set of logics and the change process.

Advantages and risks of intervention research for the study

The position of the researcher, as an executive PhD given the CIFRE convention, exposed him to valuable insight on open management issues but also some potential contamination of data collection.

Catching CTO’s political influence and complaining

Being hosted at the Innovation Direction and studying the topic of innovation management opened many doors in business units and corporate meetings. It was quite easy to gather information during the first round of interviews and the subsequent interactions with engineers, designers and managers, as well as interventions in business units.

On some rare occasions, some interviews were hard to manage as the speech being held was very negative (complaining). The dialogue had to be redirected towards proposing solutions, alternatives or different management philosophy to deal with. And sometimes, the closeness to the CTO was tentatively tried to put a word or find backup. Of course, these are dimensions of daily life in organizations and reflect a certain situation, but for the role of the researcher, it can complicate the activity of separating a clean analysis from one that is contaminated by potential superfluous details. For instance, the question of irrationality and absurdity of decisions or lack of understanding of strategies of some managers were hard cases that could have been quickly put into the quarantine of useless collected data. Yet, the role of the researcher is to propose hypotheses, new models that can explain Reality and be predictive in some way.

Being consulted for multiple and varied topics

As mentioned, in a previous section on gaining autonomy, the researcher had new opportunities to study cases, following Jacques Girin’s “opportunistic methodology” and also introduced in the comprehensive methodology as expressed by Hervé Dumez.

For instance, as new procedures had been written by the researcher for the TRL assessment, the Internal Audit Team quite often relied on him to provide appropriate recommendations to BUs. It was the occasion for the researcher to bring additional information when interacting the concerned R&T manager and try to probe the field in search for critical events, shifts in their speech, or simply new projects catching one’s attention: new design/technology embarking a organizational change.

Chapter synthesis

References