About

The vast majority of innovation and development companies worked in an environment that required constant adaptation, both in the products that they developed and the methods by which they created new designs. By far, the most vital resource to these companies was the knowledge held and developed within their design teams. These companies functioned in a hyper-competitive environment, and to survive, their design and innovation had to be maximally supported. Innovations and breakthroughs by teams needed to be captured and disseminated within the organisation. There was a vital need for maximal reuse of lessons learned in order to remain at the leading edge. Although this knowledge was the company’s most valuable resource, it was often lost because it could be difficult or impossible to articulate. The knowledge that could be articulated, in the form of reports, memos, or interviews, was the only knowledge that could be routinely captured (though evidence showed that a large amount of this knowledge was lost).

The Clockwork project was based in the dynamical systems engineering field. In common with other innovation and development areas, even when the easily articulated knowledge was captured, the work of development teams was still not well supported. This was because much of the important knowledge was not held in static documents giving abstract, decontextualised, and rationalised accounts of what was done and how. The knowledge was intertwined with the processes and tools by which they worked. The rich knowledge constructed by the dynamical systems engineer could not be abstracted away from the simulation tools with which they explored and constructed their ideas. Their tools encapsulated knowledge that could not be adequately described without them. They provided not just a description but a model of a real-world problem, a model that could be manipulated in order to investigate potential solutions.

In innovation and development fields such as dynamical systems engineering, knowledge needed to be understood as an activity of achievement and social construction expressed in the activities performed using the tools of work. To be maximally effective, organisational learning needed to move beyond the epiphenomenal documents produced in organisations and focus on the artefacts with which workers developed, shared, and expressed their knowledge and expertise. This project brought the tools of work to the centre of the learning organisation, integrated with knowledge level representations of work.

Although knowledge was the most valuable and sensitive asset within the organisation, the company necessarily needed to trade and collaborate with the outside world. This was true in commercial collaborations where each company brought particular technology and knowledge to the collective activity. Each company, while wishing for a successful and lucrative collaboration, needed to share knowledge cautiously. They needed to hide valued knowledge outside the immediate scope of the collaboration to ensure their competitive advantage. This dilemma was common in dynamical systems engineering where expertise and products tended to be contracted in as needed, as having such expertise in-house would not have been commercially viable. Similar situations could arise in customer-supplier relations when the commodity was as much knowledge of the product as the product itself.

To meet these conflicting goals, the company needed to be able to sharply divide public and private knowledge, even though they were closely related. Our solution was to support cautious knowledge sharing by developing tools to provide public knowledge sharing spaces between companies, that were divided from though integrated with private knowledge on each side of the partnership.

The Clockwork project could therefore be broken down into three distinct objectives, structured around measurable outcomes.

Objective 1: Support teams in sharing knowledge enriched simulation models of dynamical systems

Product development teams were required to collaboratively create, test, and modify simulation models of new and existing designs. These teams were often distributed, for example, between centrally located staff and those testing products in the field. There was a need for communication and collaboration between developers. Existing support mechanisms for this were often separated from the tools of work, even though knowledge could only be fully and efficiently expressed within its context. Incorporating dynamical system simulations, being the tools of their work, was of crucial importance, as they acted as holding environments for ideas that could never be articulated. This allowed knowledge to be shared that was tangible though inarticulable. Our first aim was to support team-level knowledge sharing by integrating the tools of work (in our case, dynamical system simulation tools) with formal documents, informal discussions, and knowledge models.

Our success in meeting this objective was demonstrated by:

• Methodology of how enriched simulation models could be shared within distributed teams.
• Development of the ClockTool toolkit with integrated simulation tools, annotations linked to knowledge structures, documentation, and discourse.
• Case study of how the ClockTool toolkit and methodology could support knowledge sharing within the industrial and academic partners.

Knowledge sharing with enriched simulation models was not well supported. An expected outcome of the CLOCKWORK project was an approach and tool set to support knowledge sharing as demonstrated by the effective use of the CLOCKWORK knowledge sharing tools and methodology to support collaborative research and work activities of the academic and industrial user partners within the dynamical systems domain. Observations supporting this outcome included: changes to research and work practices to take advantage of the opportunity for distributed and asynchronous collaboration; development of the CLOCKWORK archive in terms of the evolution of the knowledge model, and growth in enriched simulations and related discussion; and the development of shared simulations and knowledge level models. The measurement of baseline performance and the impact of the CLOCKWORK project against this objective included the analysis of organisational memory content, server logs, and workplace observations. At present, for example, Matlab/Simulink packages included Report Generators which automatically generated descriptions of model structures in the form of a text file. Similarly, Simpack generated ASCII files with possible user comments attached to the simulation model. However, none of these facilities were sufficient support for knowledge sharing.

Objective 2: Support reuse of simulation and modelling knowledge across the organisation

Although innovation in engineering design involved developing solutions to novel problems, there was great scope for the reuse of prior design knowledge. The knowledge that needed to be shared was contained in a number of formal and informal documents as well as the dynamical system simulations themselves. In many domains, including the dynamical systems domain, contextual information was vital for reuse. Developers though could not be expected to reuse from vast archives of previous design projects without support for finding the right knowledge at the right time. The designers needed to be supported by retrieval tools. Searches on surface criteria such as text phrases in the documentation or the kind of components contained in the simulation would have been inadequate, and had to be driven by a knowledge level representation of the domain to which the reusable design scenarios were linked. Our second aim was to support effective knowledge reuse by retrieving rich representations of the design process according to a knowledge level model of the domain and the design process.

Our success in meeting this objective was demonstrated by:

• A methodology of how simulation and modelling could be reused within a company.
• The development of the ClockTool toolkit to support the capture and intelligent retrieval of simulations and their design history.
• Case-study of how the ClockTool toolkit and methodology could support knowledge reuse within industrial settings.

Currently, there was insufficient support for the efficient reuse of simulation and modelling knowledge distributed across different media to support design innovation. The current tools supporting reuse employed only standard search keyword-based techniques. No knowledge-based retrieval tools were available.

An expected outcome of the Clockwork project was the development of an effective methodology and toolkit to support knowledge reuse as demonstrated within the project by the academic and industrial user partners. Observations supporting this outcome included: access to previous design cases during the innovation process, and discussion of how aspects of retrieved design cases could motivate or form part of solutions to current challenges. The measurement of baseline performance and the impact of the Clockwork project against this objective included the analysis of captured discussions, workplace observations, and surveys of reuse strategies adopted by the user partners.

Objective 3: Support cautious knowledge sharing between engineering companies

For the innovative organisation, knowledge was their most valuable asset. These companies needed methods and tools by which they could negotiate, trade, and collaborate with other organisations, giving access to knowledge they wished to share, whilst allowing them to hide knowledge they wished to keep private. The cautious, selective sharing of knowledge was vital for companies that needed to collaborate with partners, suppliers, and consumers. Facilitation of these processes also supported continuous innovation within the company. Innovation lay on the interface between an organisation and its environment, from which market needs and opportunities appeared. There were further obstacles to successful cautious knowledge sharing between companies. The physical product that was the locus of their collaboration could be captured in a shared simulation, though the shared object would have unique links to the private simulations and knowledge in each collaborating company. Our intended approach was to develop a knowledge mediation solution that could develop knowledge level and simulation level interfaces between the shared artefact and the private knowledge and products of each company. Our third aim, therefore, was to support cautious knowledge sharing through knowledge interfacing between competitive collaborating companies.

Our success in meeting this objective was demonstrated by:

• A methodology of how companies could successfully collaborate and trade with other companies within a competitive market.
• An extension of the ClockTool toolkit to support cautious knowledge sharing between companies.
• A case-study of how the ClockTool toolkit and methodology could support cautious knowledge sharing within an industrial setting.

Innovation companies were currently ill-supported in cautious collaboration with other companies. Current co-simulation tools made it possible to connect simulation sub-models from different simulation packages by compiling simulation components into the binary coded libraries. Consequently, knowledge sharing was either fully opened or completely closed. None of these extreme cases was desirable for knowledge sharing companies.

An expected outcome was the effective use of cautious knowledge sharing tools and methodology by INTEC after trial development and initial use within the academic network. Observations supporting this outcome included: improved working practices between companies in collaborations involving products and techniques incorporating valuable restricted knowledge; and effective interfaces between private knowledge and external activities. The measurement of baseline performance and the impact of the Clockwork project against this objective included an analysis of the use and development of simulation level and knowledge level interfaces, and interviews investigating the nature of cautious knowledge sharing activities.

Project Details

Start: 1/07/2000
Duration: 36 Months
Funding Body: CEC
Project Number: IST 12566

Project Partners

The Open University – Knowledge Media Institute
Dr Zdenek Zdrahal

Deutsches Forschungszentrum fuer Kuenstliche Intelligenz GmbH
Dr Harold Boley

ELOTHERM GmbH
Alex Seitzer

University of Ljubljana – Jozef Stefan Institute
Prof Ivan Bratko

Czech Technical University in Prague – Department of Mechanics
Prof Ing. Michael Valasek

INTEC GmbH, Ing. Ges. fuer neue Technologien
Dr Alex Eichberger

Loughborough University – The Mechatronics Research Group
Prof Rob Parkin