Towards controlled innovation of complex objects.

Abstract

Within the ship design industry, and in particular in the development of large, complex
and innovative vessels experienced ship designers play an important role in organizing
and structuring the design process. How this actually happens within projects that
develop such large, complex and innovative vessels for a single client is not clear. This
observation lead to a broad, initial research question as the starting point for further
research.
How are innovative, large and complex vessels developed in practice?
Because of the broad, explorative nature of this research question it was not possible
to apply a research strategy aimed on justification and validation, an approach more
common in ship design. After the evaluation of several alternatives a Deweyan inquiry
was selected as most promising for this research, as it allowed for the parallel development
of theory and practice in a pragmatic approach.
One of the first steps was to provisionally pinpoint and frame the initial doubtful situation
based on observations in both theory and practice. The current state-of-the-art ship design
literature was analysed to determine the creative elements in each design strategy. To
enable this analysis CK theory was applied. CK is an approach introduced as a unifying
theory to describe creative design, it differentiates between two distinct ‘spaces’ with
their own rules and logic: the Knowledge space, which is based on propositions that are
either true or false and the Concept space, where propositions are still undetermined and
can either be true or false. To describe creative design, a design process should describe
steps from the knowledge space into the concept space and vice versa.
The analysis showed that the majority of design processes do not describe the full scope
of creative activities, but only concentrate on parts of the ship design process. Creative
ship design, system based design and requirement elucidation explore a broader scope
of the ship design process, concentrating full creative design of the overall ship design,
or in the case of requirement elucidation of both the requirements and the ship design.
To analyse the four case studies CK theory was complemented by the system thinking
perspective observed in literature. Both were applied to analyse the development of
four vessels: the SWind1000, the Greenstream, the Bourbon Orca and the Pioneering
Spirit. In each of these case studies the project did not concentrate on a single level of
decomposition (a focus on either the business case, the ship design, the system design
or component design) but developed multiple levels of decomposition in parallel. For
example, in the development of the Bourbon Orca, the project team worked on the hull
design, the deck equipment and the propulsion system (system level) in parallel with the
overall layout of the vessel (ship design level).
This phenomena, called ‘coevolution’ between solutions on different levels of
decomposition appears to be a key part in innovative ship design, allowing creative
solutions to develop and influence each other. These creative developments often occur
within different companies or different departments of the same company, resulting in
considerable interaction between the involved actors.
During the analysis of the development of the four vessels, complemented with a
series of interviews and an additional reference case it became clear that one of the
most important roles of the experienced ship designers was to handle and manage the
interaction between different actors caused by coevolving solutions. There were two
dimensions to this interaction: the technical dimension illustrates the content of the
interaction, the social dimension concentrates on the way the information is transferred.
Based on these observations and an initial choice to explore the technical dimension of
the interaction the following constructive hypothesis was developed, as a starting point
to evaluate the observations made in both practice and theory:
To improve the development of coevolving innovative solutions a design process
should focus on two levels of decomposition, allow for weak system boundaries and
take into account the technical dimension (the content) of the interaction.
In practice, experienced naval architects are more than capable of developing innovative
solutions. However, to improve such processes more control would be preferable without
constraining the initiation of creative solutions. Based on the initial review a more
controlled process should include the following elements: the definition of potentially
innovative solutions, the involvement of technology partners and other actors, support
the integration and coevolution within weak boundaries and manage the timing of
creativity.
To develop a design strategy capable of taking the technical dimension of the interaction
into account, a model was necessary that describes cohesion within a system of systems.
This model could in a later stages be used as a template for a design strategy.
The model to determine the cohesion in a system-of-systems is based on the individual
system description using the aspects Form, Characteristics, Performance and Function,
which is the same on each level of decomposition. Five relations were determined outside
the system boundaries: decomposing and integrating a system (the form of system X
– the form of system X-1 and vice versa), the contextual influence (form of system X
– characteristics of system X-1) and the performance – function cohesion (function of
system X – performance of system X-1 and vice versa). This results in the overall model
of the cohesion shown in figure 1.
The model of cohesion was used as a template to develop a design strategy taking the
technical dimension of the interaction into account. The strategy developed in this
research starts from a required ship functionality (the objective), which is developed into
different sets of system performance parameters, which lead to the individual system
developments. In a later stage, these individual system developments are integrated into
the overall ship design and evaluated (figure 2). This is a departure from the conventional
approach within the ship design industry, which often starts from an existing general
arrangement.
The design strategy was applied during the development of two Ulstein Design &
Solutions B.V. (formerly known as Ulstein Sea of Solutions, USoS) vessels: the Ulstein
xi
Performance/
Characteristics
(parameters)
Form Form
Characteristics
Performance
Objective
(function)
Ship Design System Design
A-B-C... n
(5)
(2)
(3)
(4)
(6)
time
Existing
business-case
Existing
components
Function
Function
System X
System X+1
Form
Form
Performance
Characteristics
Performance
Characteristics
Function
Design X+2
Form
Performance
Characteristics
System X-1
AXDS, an arctic drillship developed for Statoil and the Bravenes, a subsea rock installation
vessel developed for Van Oord. The first project was developed with active influence
of the researcher, as I was involved as a naval architect and design process guardian.
The second project was provided with the initial documentation, but was not actively
influenced by my involvement.
The application of an explicit design strategy had a considerable effect on both projects.
Both designs were considered very innovative without major showstoppers. Even though
there are only two projects available, the design strategy still appears to have a positive
influence on defining innovative solutions, the structured involvement of technology
Figure 1. The model of cohesion
Figure 2. The design strategy, based on the interaction between system and ship design
xii
partners and the integration and coevolution of new solutions as well as the timing of
creativity in the process. For this particular purpose, the design strategy was well suited,
relevant and provided a workable and sufficient flexible solution. This confirmed the
idea that design strategies that take coevolving solutions and the technical dimension
of interaction into account provide a better description of the innovative ship design
process.
During the hypothesis and model development in this research the social dimension
of the interaction was not taken into account. During the experiments the role of this
social dimension became more evident, as the explicit design strategy required different
actors to think and discuss their role within the project. For further research, a proposal is
developed to explore the social dimension of the interaction between actors in the design
process, as a new starting point or ‘doubtful situation’ in further research. Such a research
could for example concentrate on boundary objects, process models or synchronizing
activities.
This research explored the way experienced ship designers develop innovative solutions,
it revealed that the current ship design methods insufficiently describe the process of
coevolving solutions, including the interaction between the different actors that result
from this. This dissertation does not result in a complete prescriptive method to control
innovative design, but it explores, identifies and evaluates the key parameters that
should be taken into account in such approaches.
xiii