Chapter 10 - Test cases: the Lower Deck and the Connected Cabin

Everyone who designs devises a course of action aimed at changing existing situations into preferred ones. Herbert A. Simon

We have established a new model to reconcile the original construct of ambidexterity with the evolutions innovation management. By revisiting the canonical model of decision-making with design theory, decisional ambidexterity allows us to endogenize the unknown within a technology of organizing collective action to metabolize innovation across the firm. The performance is supported by the interactions between decision-designers revealing interdependencies and their reconfiguration. The value management unfolds as the project trajectory may be manoeuvred away from technical, organizational and environmental fixations. Exploitation dimensions are used to sustain a generative and grounded exploration.

The model was first built on a simple agent-based foundation, and since we tasked ourselves several requirements defined in contrast with anomalies, we made use of the mirroring hypothesis to discuss the necessary organizational change. This transformational process we propose to link with organizational metabolism, is an embodiment of the recombination of routines or regenerative dynamic capabilities.

Therefore, it reflects the generative learning expected by (temporary) project-based organization and its boundaries with permanent organization. Otherwise, project will float across the organization. Consequently, we propose to test our model of mutual conditioning between exploration and exploitation. To do so, we use two case studies conducted at Zodiac Aerospace. We extend the preliminary testing and understanding gained in the two previous chapter when defining our model. The anomalies became understandable as the model of decisional ambidexterity gives a larger framework to make sense of decisions and concepts that are otherwise labelled as absurd or irrational.

The first project stems from the Design Case studies hosted by ADT we have analysed in the anomaly detection chapter. As explained, in the case presentation, the project was launched after ADT manager decided to further conceptualize the recurrent idea of using aircraft’s lower deck. It was also notified by the researcher in 2016 when presenting his first analysis of the design thinking cases.

The second project originates from two members of the ZSTC with a history of entrepreneurship and innovation. They had reasons to launch a project on aircraft connectivity given their experience on telecommunications, satellites, antennae and onboard computers. It was set as a proper MBU project justifying its value proposal to the ExCom (see case presentation). Both cases are comparable in the sense they were sponsored by the ExCom, offered budget and internal/external network support to find answers, contour their ill-defined problems and test the concepts (on slides, digital mock-ups and physical proof of concepts). We use the same analytical lens and descriptors introduced in our methodological chapters enhanced with the logics of decisional ambidexterity (decision categories, learning, organization design). Comparing these two cases will allow testing further our model by gaining an understanding that is missed with models of non-mutual conditioning between exploration and exploitation.

In the first section, we present the Lower Deck project, its trajectory and present the several patterns phased with decisional ambidexterity. In the second section, we discuss the Connected Cabin project, which, despite benefitting from an equivalent context and support, fails to sustain a decisional ambidexterity. The project unfortunately regresses towards biases of traditional models of ambidexterity, hence killing its innovation potential of attraction. These two case studies give us a richer understanding of what decisional ambidexterity can achieve in reconciliating logics of exploration project management with a renewed organizational learning and adaptation rooted in generative logics.

---

Lower Deck: from design feature to value space

Trajectory of the project

As presented in our case presentation (see section), the project of Lower Deck started due to the accumulation of concepts generated in several projects over two years at ADT. Among them, many had the recurrent idea of using the aircraft’s lower deck. A first provisional patent application process was started mid-2014 (see patent no. US20170233058A1). Late 2015, the patent was made public and made quite an impact with numerous critics coming from specialized blogs, magazines and newspapers¹. The Design Thinking case studies started in 2015 and what continued over 2016 also made use of the lower deck in various ways.

The researcher had identified the pivotal role of this a priori underused aircraft area to enable concepts that cross over BU boundaries. In parallel, ZA’s CEO, during an interview conducted in May 2018 by the PhD candidate, told the concept had already been discussed during ExCom meetings previous to the project formalization. The case seemed justified to them as they had insights on the declining freight market: an increasing empty space in the lower deck was then available. Finally, a project was kicked off and took the concept on its own mid-2016 and directly managed by ADT manager.

Several dimensions were addressed simultaneously. Firstly, an underlying mechanical stress study was conducted: the CTO contracted an engineering consultancy firm to evaluate the technical and safety feasibility of using the lower deck for passengers seating and other airline ancillary revenues. Secondly, the support of group business development directors in charge of airlines accounts offered a way to engage talks with key airline representatives (interpreters (Verganti and Dell’Era 2014)). It was the occasion to probe airlines interest and gradually contour client value. Thirdly, later 2016/early 2017, the VP of Strategy (ADT manager’s director) contracted another consultancy firm to dig into the ticket pricing model (yield management) and airlines operation cycles in order to economically value lower deck’s use cases but also find time frames when the aircraft lower deck could be modified without disrupting aircraft’s operations. Finally, a last study was contracted internally with two business units familiar with tailored aircraft modification (VIP, movies, medical care, etc.) for a preliminary design and engineering specification to adapt the lower deck for the different use cases. Early 2018, a new patent application was made with more critical claims specifying the value of the concept refined over time (see patent no. WO2018037268A9). And the global concept was finally revealed to the public with Airbus as a sponsor and co-designer during Aircraft Interior Exposition in April 2018. A partnership agreement was signed between Airbus and Zodiac Aerospace to develop the product.

What is rather surprising in this project trajectory is the fact that the lower deck feature was first used in several concepts and systematically discarded by BUs who participated to previous projects (including those supported by Design Thinking practice). Concepts were discarded despite having justified user and client value, and having raised some technical difficulties through complimentary meetings with engineers and designers. However, once the lower deck was turned into a proper value space looking for use cases, it gained much more momentum, specially from business unit managers. They indeed requested to host the project in their branch of activity. The figure 1 below gives a synthetic view of the project trajectory including the previous Design Thinking cases. Given the decisional ambidexterity model we can specify the nature of decisions designed.

Problem inversion of Design Thinking cases

Firstly, we can see that some decisions started from an enhanced optimization, moved towards hacking several choices extending concepts in between or outside BUs boundaries. Some of these were triggered by user input and What If? scenarios. The Lower Deck feature was one of them ($d^*$). However, once the concepts are selected, ranked, sponsored and promoted to BUs, the generative effort is reduced to an optimization: finding the best fit for BUs without changing engineering capabilities, or at best marginally by incrementally modifying existing products (new requirement). The issue is also that this optimization and biased adaptation is mirrored by underlying interdependencies and fixations.

For instance, using a waste chute and sorting mechanisms in the lower deck could be developed by Zodiac Inserts who developed the trash compactor (a trolley locked into the galley area). The latter corresponds to a given certification and qualification process. Zodiac Galley could also have considered the product development given its specialty on panels and fixtures withstanding stress and loads, as well as water and electrical circuitry integration (also another certification/qualification process). Finally Zodiac Cargo, who develops containers would be appropriate to develop the relevant container lodged in the lower deck providing the sorting and extraction of waste. It is a potential multi-BU project which always calls for another requirement partly mastered by another BU. It could be developed by each BU individually as technologies are similar or at reach. The incentive to work together would only come if they had a real client demand and/or a change in certification and qualification. Moreover, the concept requires reconsidering airline operations and purchasing policy, so training and change of practice is required in the sales pitch.

Secondly, the induced organizational and environmental phenomena were not really addressed in the Lower Deck project premises. The problem was turned inside out to make the choice hacking more visible. So, instead of forcing the BU congruence, it gave a whole new flavour to the innovation potential of attraction as described in Table [table:LWDtab] below. Meetings held with the ExCom insisted heavily on the technical advantage that should be designed, since they were rather confident on the client value and market attractivity. The discussions were oriented to ensure the development of a turnkey solution as well as competitive advantage against the footprint of aircraft manufacturers. In other words, they shifted the concept towards having an empty shell. The value space guaranteed its technical feasibility, with the prospect of a separate qualification and certification process. Consequently, BUs could articulate their products in this new space, without worrying about their engineering and organization design fixations, and focus on regenerating their products for this new value space.

Regenerative capability through design formulation

The Lower Deck became rather modular and generic for several BUs who needed to develop new design capability to fit requirements evolution. It encouraged the adaptation to the newly designed environment. Instead of forcing the fit to each specific BUs, the problem inversion, allowed attracting them to provide use cases.

Decisional ambidexterity reveals the importance of project management and corporate entrepreneurship interactions steering the project trajectory where the novel value is not forced onto an a priori independent BU, but rather on formulating a concept design that attracts them to grow. It moves them away from their fixations by creating generative fit (Avital and Te’eni 2009), thus supporting the regenerative capability expected from them. The organization design was not clearly discussed during project meetings, but the decisional ambidexterity allows us to formulate the idea that product engineering embedded how interdependencies could be unlocked. Otherwise, a non-mutual conditioning would focus only on numerous other explanations: leader as champion, acculturation to the lower deck prospect. Perhaps yes, but it does not make it manageable and does not give much directions to guide collective action and engineering design practice.

Decisional ambidexterity analysis

Model descriptors	Non-mutual conditioning	Mutual conditioning extended model
Model of coordination and collective action	Lower deck exploration could be fully conducted separately to create a new business unit.	Lower deck business case is tightly articulated around the unlocking of underlying interdependencies that were not addressed in previous projects. It allows preparing exploitation and coordinate BUs around a host concept and value space.
Generative processes	Non applicable	The conceptualization builds upon limitations raised by BUs. Demonstrate value through actionable knowledge on how to exploit the concept (technical and use cases propositions). BUs don’t have to worry for the engineering of what could be common to other BUs, as the concept already determined those interdependencies.
Environment cognition	Non applicable	The environment is managed with concept tested among potential clients, engineering qualification and certification to ensure passenger and cabin crew safety. The environment is shaped to welcome BUs future engineering effort. ExCom sponsored the engineering effort, qualification/certification constraints, and ensured the concept would crystallize and secure the market creation (technological advantage).
Organization design	The new BU or spin-off is a possibility, but pending discussions consider encouraging the regeneration of the cargo and container BU.	The organization design is integrated in the conceptual exploration as interdependencies and fixations ingrained in BUs are unlocked. BUs do not see it as a an add-on to their existing products, but rather as a value space to project onto. The Lower Deck may be integrated by a BU (cargo and container) hosting the contributions of other BUs combined with engineering capabilities for aircraft modification BU. Relationships are now built around the knowledge of enabling the cargo area for a multitude of use cases.

---

Connected Cabin: project mode and proof of concept?

Trajectory of the project

The project was initiated by two members of the ZSTC (see case presentation) who started testing the idea of addressing the trending topic of Big Data and Aircraft Connectivity. The topic had already triggered the development of in-flight connectivity with satellite antennae and ground antennae, including the ongoing product development by Zodiac Data Systems specialized in providing tele-measurement devices for testing and monitoring. In September 2016, the first summit dedicated to aircraft connectivity, Smart Plane, was hosted in parallel as a chapter of the major World Satellite Business Week in Paris. The topics were gradually gaining importance in the industrial ecosystem. However, already in late 2015, during a ZSTC meeting, the two members pitched the idea on how Zodiac Aerospace, as a group, could set a foot on this emergent market.

As pictured in figure 2 below, they originally presented the idea of creating connectivity platform providing value-added service for predictive maintenance and airline operations. The claim was grounded on the fact that Zodiac Aerospace was a legitimate player to provide such integrated solution as the group covers the full extent of the cabin and several other key equipment in the aircraft systems. Their knowledge on tele-transmission with the products developed within Zodiac Inflight Innovations (fast growing player in the inflight entertainment market) allowed them to consider connectivity solutions between equipment so far isolated and managed by independent systems. The concept would thus overcome rigidities for airline operations and consequently passenger discontent.

Several meetings were held with the ExCom in order to have full support of top management so that middle management would allocate resources in order to demonstrate the value proposal and business case of such project. The discussions were complex and frustrating as the board of directors was asking for what would make Zodiac Aerospace legitimate beyond market footprint. They were asking for a technology differentiator, something that would ensure the market niche². For instance, the CEO would have preferred to have Zodiac Data Systems to develop a technology (antenna, computer, etc.). In the end, the board was gradually convinced that the key was to be able to manage an ecosystem of industrial players rather than the hardware by making a direct comparison to Apple iPhone’s development. They allocated budget to the project so that BUs could allocate time to their engineers and managers. The former CEO also communicated to investors on the project’s kick-off.

Soon after late 2016, a whole project team was tasked by the ExCom to build a demonstrator for the following Aircraft Interior Exposition (AIX, April 2017). The design effort was driven towards making a Minimum Viable Product: a scenario was drawn with several pieces of equipment showing the benefits of cabin connectivity for airline operations and passenger experience. The proof of concept embodied in a full scale mock-up was modelled to show integration capabilities and synergies. Numerous workshops were held in different location to stage gate the project and track the progress of sub-projects around the connectivity backbone.

During an internal show with airlines organized by of one the major cabin BU in California, they had the occasion to probe some potential clients to provide feedback to the ongoing demonstrator design. It was also used to narrow down the questionnaire used during AIX. The exposition was a success, leading airlines executives came to visit and experience the product along with ZA’s top management. Several critical comments came regarding known preoccupations in the market: data ownership, intellectual property, compatibility and standardization among industrial players.

What is quite impressive is AIX aftermath. Every BUs complimented each other, went back to their businesses. And several emails exchanged between project’s members, including R&T, Marketing and Engineering managers of respective BUs, revealed how perplex they were on the project’s termination. They wouldn’t understand why the ExCom and other BUs wouldn’t continue with the project. Several other factors can be taken into account: e.g. change of priorities for ExCom concerned with Safran’s acquisition (but it didn’t slow nor stop the lower deck project), and European funded projects on small connectivity technology block at BU-level but reduced scope.

Regression to the mean

By contrast with the lower deck project, and given the similar context, the project’s gained momentum raises the question of how efficient the project’s output was. Of course, pure continuity of the project cannot be always requested, but the takeaways are scarce. The major one still keeping a system thinking on the value chain can be seen for Zodiac Catering Equipment, but they already had an agenda on passenger experience, cabin crew and airline/caterer operations before the project.

The figure 2 shows how the genericity and platform concept was originally pushed forward in a newly managed environment. Decisional ambidexterity quickly reveals that there was some regression to the mean (Kahneman 2011) or concept shifting towards a reference point (Heath, Larrick, and Wu 1999). This drift is unfortunately stuck due to organization fixation and interdependencies relatively to their product engineering and market segmentation.

Proof of concept to trigger a full-scale multi-BU project?

As we also stress in table [table:Iconnztab], the organization design was left aside and uncontrolled. So, each respective BU would work on recombining their engineering routines to think of the new platform for cabin connectivity. The proof of concept targeted the added requirement of using a communication bus and protocol (wired or wireless). The decisional ambidexterity highlights the need for an organization-design-oriented proof of concept supporting the original intent and sustaining the regeneration of capabilities among BU.

At some point, the CTO notified the researcher that the project was in “project mode”. This comment clearly criticizes the limitations of encapsulation and its tendency for not considering enough change management and organizational learning and design. Decisional ambidexterity can also be used to discuss the proof of concept activity embodied by the physical demonstrator, scenario/storytelling, and associated prototyping activities to embody the scenario/storytelling.

It allows indeed to formalise choice hacking, environment management, design of states of nature, designing by referring to BU (organizational) fixation effects. In the Connected Cabin demonstrator prototype, we can actually extrapolate what else could have been done to sustain the original concept. For instance, beyond the hardware add-one for existing products, exploiting the constraints of the connectivity topic could have been explored: bandwidth allocation and use, protocol, confidentiality and property. The hardware is of course important, but it corresponds to a retrofit activity, and it is made possible only after crossing qualification/certification/acceptation barriers. These are high-level and system constraints but they can easily shadow engineering ideas. Decisional ambidexterity proposes to generate and segregate alternatives to identify such control variables and actions to realize concepts.

Decisional ambidexterity analysis

Model descriptors	Non-mutual conditioning	Mutual conditioning extended model
Model of coordination and collective action	Exploration is conducted by a temporary team relying on delegated resources.	Exploration initially targets a new platform approach for cabin equipment exploited in ZA. The equipment connectivity and associated system is used as an input to pool data in a centralized core system.
Generative processes	The purpose of the project drifts as the project’s main output becomes a proof of concept for AIX 2017.	Hacking all choices to wish for a new industrial ecosystem organized around ZA integration capabilities. New connectivity solution, services for airline operations.
Environment cognition	The demonstrator and prototyping activities do not overcome fixations and locked interdependencies.	Original intent is to act as a platform leader for the ZA equipment and competition. Request from ExCom to have a key technological advantage to secure the new market.
Organization design	The project gradually moves away from the concern of articulating the technology with BUs engineering capabilities as the proof of concept mainly emphasizes the need for an added connectivity feature to all pieces of equipment.	Original project’s mission was to federate a service platform for connectivity with BUs benefiting from added-value equipment enhanced by data. BUs would then need to regenerate their capabilities forservitization.

---

Chapter synthesis: a reconciliatory model for ambidexterity

---

References

---

1. For instance see DailyMail Online brief on December 8, 2015: Just when you thought they couldn’t get any more seats on board: Aerospace company designs lower-deck cabin with vending machines and display screens instead of windows. ↩︎

2. this information was drawn from Multi-BU meeting minutes and CEO’s interview conducted on May 7, 2018 ↩︎