مدیریت دانش سازماندهی و اشاعه دانش در توسعه محصول جدید : درس هایی از 12 شرکت جهانی

New Product Development is one of the most knowledge intensive processes in business and is itself constantly creating new knowledge. As NPD relies heavily on collaboration within cross-functional teams, the question of how such knowledge, which to a large extent is tacit, should best be managed and disseminated is crucial. Studying this problem with data from the multi-project NPD environments of 12 large global manufacturing companies, we identify three typical organisational structures for knowledge management, one as a central strategic function, a second where KM is internal to individual projects and a third where it is the affair of specialized functional departments. We assess the strength of each option in terms of clarity of mission, how supportive they are in promoting the transfer and sharing of knowledge, and what sort of frictions may accompany their use. A focused area of the research is the alternative formats for job rotation that each structural style promotes, and the prospect of some ‘ideal type’ of knowledge management structure as a hybrid of these styles is examined.

With a steady record of more than 500 new products launched per year since the mid 1990s, the successful growth and development of Euro Supplier 1, a European-based global supplier of automotive components, is in a large part due to its systematic focus on managing the product development process as a continuous learning process. Since taking office 5 years ago, Jack D. (the Divisional Head of R&D), has consciously led all activities, processes and players involved in the product development process with three key objectives: activating the knowledge embedded in existing product solutions, extracting as much as possible of the tacit knowledge held by people, and capturing as much as possible of the new knowledge created in each development project. Measured by steadily improving R&D productivity in terms of cost and lead-time reductions for an increasing functional complexity and growing numbers of new products, his mission has been largely successful. Jack D. attributes this success to a mix of clear support structures, effective operating procedures and a climate that favours the sharing of knowledge during operational problem solving as natural, all under the umbrella of a project intelligence unit and his personal supervision as division head. Few product development managers are as fortunate as Jack D. While product development is widely recognised as one of the most knowledge-intensive processes in business, as a `place' – a ba, - where knowledge resides, expands and continuously changes shape as a result of action and problem-solving practice,1 there have been no in-depth analyses and few recommendations as to how Knowledge Management (KM) should best be organized and formally structured in development-intensive firms engaged in a multiple parallel development projects. This issue becomes even more intriguing when taking into account the complexity of large-scale product development projects, which Clark & Fujomoto's landmark study of New Product Development (NPD) performance has characterized as similar to solving a huge equation system.2 Starting from framing the 'problem' as the product to be developed, the ‘equations’ address a numerous series of detailed and highly volatile technical and organisational questions. These questions take time to answer, and the ‘solution’ will depend on compromises between a large array of requirements represented by different internal and external players, each with different priorities, professional backgrounds and cognitive frameworks. The complexity of the NPD process is reflected in the complexity of the related knowledge flows, which are both crucial for advancing the process but at the same time difficult to manage.4 More specifically, part of the knowledge needed for developing a new product already exists within the organisations involved, while fresh knowledge is created as the process unfolds, problems are analysed, and the product is developed, refined and ultimately launched on the market.5 The existing knowledge, stored or embedded in the minds of people, in archives, in existing products and in procedures and equipment, needs to be recognised, retrieved and made available to engineers and other project participants. The newly created knowledge developed by practical problem solving needs to be analysed, shared and integrated with previous knowledge, to ensure a spiral of continuous expansion and development/refinement of knowledge for future use in the NPD process.6 There is a consensus in literature and practice alike that the transfer and sharing of knowledge are critical components for enabling this integration of newly created and previously existing knowledge, and thus for efficient NPD.7 In spite of the apparent importance of how to organize for Knowledge Management in NPD, the literature on the subject is almost nonexistent. Project management research assumes that effective KM will be a positive side-effect of an optimised project structure, while research focusing on NPD practices is mainly preoccupied with description or prescription of what KM should be about, addressing the question of how only superficially. R&D Directors and Project Managers are thus left much to their own judgement and imagination in deciding how to organise KM to facilitate the transfer and sharing of this continuously evolving, and often tacit, body of knowledge among a large number of individuals and groups. Rather than a systematic, strategically aligned and operationally grounded organisation for KM, the result is more often a patchwork of ad-hoc initiatives of limited usefulness and weak sustainability.8 Hence, the broad questions we address in this research are How do large product development-intensive corporations organise their Knowledge Management activities? and How does KM organisation affect knowledge transfer and sharing? Our aim was to identify the different structural patterns used in practice and to analyse their perceived efficiency, strengths and weaknesses so as to progress the understanding of how KM structure influences knowledge dissemination, behaviour and, ultimately, NPD performance. In particular, we wanted to search for basic types of structures, and analyse whether any particular organisational configuration could be identified as an ‘ideal type’ for Knowledge Management organization in NPD. The article is organised as follows. We first conduct a focused literature review of organizational aspects of KM in multi-project new product development environments, specifying also the notions of knowledge transfer and sharing. Section three presents the methodology of the study. Our research is based on the study of nine global OEMs and three global systems suppliers in the vehicle, appliances and electronics industries where data was collected on large-scale and internationally conducted product development projects. Section four presents and analyses the study results, focusing on the organisation for Knowledge Management and the impact organization has on the processes of knowledge transfer and sharing. Finally, we summarise and discuss the findings, and present implications for practice and research.

The research has confirmed the great importance of knowledge in the New Product Development process, driving a pressing need to develop and manage effective organisational structures and support mechanisms for knowledge dissemination. KM represents a timely supporting mechanism in view of current challenges related to the globalization of NPD projects, the race for the shortest development lead-times, and the ever-increasing participation of suppliers in all development process stages. Given its strategic and operational importance, NPD managers have to decide on some way of organising the responsibility for KM. The scarce literature on this subject and the different approaches chosen by this study's companies indicate that more attention needs to be paid to this basic condition for effective Knowledge Management. Based on our findings, a template for the efficient KM organisation in NPD would combine four important pillars: • Strong strategic alignment of KM objectives, initiatives and outcomes ensured by a central KM function with a direct mandate from the CEO or the VP for R&D. This central function should develop procedures for knowledge capturing and dissemination following the operational reality of the development process, and ensure both continuous and post-project knowledge transfer across the organisation's full project portfolio; • Strong project focus ensured by allocating dedicated personnel to each development project with the mission of identifying specific project needs in terms of knowledge transfer and sharing, reporting them to a central KM function or the R&D Manager, and then implementing and following up on their execution and performance implications at the project level; • Strong integration of expertise ensured by initiatives for developing state-of-the-art knowledge in deep fields of expertise and transfer of this expertise between projects. Dedicated personnel, reporting to a central KM function or the R&D Manager, should be responsible for making the functional knowledge developed in individual projects available to the entire project community; • Improved leverage of different job rotation schemes by combining effective rotation between projects of personnel possessing valuable knowledge with rotation of personnel between departments with a view to extending individuals’ inter-functional tacit knowledge base. This holistic approach to Knowledge Management is an important finding. A key lesson for managers is that they must find the best fit between KM structure and NPD organisation, and then integrate different KM organisational structures in such a way as to leverage positive aspects and neutralise negative ones. Jack D., the R&D division head who, together with his colleagues, reported the greatest satisfaction with their KM efforts concluded: ‘From the very beginning, I have emphasised that KM cannot be superposed as an “extra” requirement at certain points in the development process. It certainly requires specific routines, systems and methods, but needs to be integrated with other managerial imperatives of strategic, technology and human nature’. In terms of implications for the academic community, the most fundamental is that studies into Knowledge Management in general, and knowledge dissemination effectiveness in particular, should control for KM structure, to identify if a formal structure exists and what it looks like. This study has shown that in the NPD field: • Strategic alignment of KM efforts is likely to be poor unless KM efforts are defined and implemented through a central structure; • No extension of common cognitive grounds is likely to take place unless KM is driven by the pragmatics of the development projects and organised from within the project structure; • Functionally oriented KM efforts tend to have little impact on project performance unless they are actively integrated into the project management structure through formal reporting schemes. The analysis of the different approaches to job rotation points to the fact that, as suggested by the literature, there is a strong preference for relying on the individual as the means for knowledge transfer. Managers also had a tendency to mistake knowledge transfer (explicit) for knowledge sharing (tacit), while in fact the transfer of personnel from finishing to new projects is unlikely to result in shared knowledge. There is thus a need for more research into the actual outcomes of different job rotation initiatives, with careful specification of what was intended to be achieved – `simply' knowledge transfer or knowledge sharing and ultimately a continuously growing base of shared knowledge. Further, the importance of integrating different approaches to KM structure calls for an effort of developing integrated research agendas. The current dichotomy of planned vs. emerging KM, with the former emphasising the role of formal structures and IT support and the latter focusing on social networks and interactions, must be bridged in order for research to support practice more relevantly. The study has also opened up interesting avenues for future empirical research. Survey research would be needed to specify in more detail the three identified organisational Knowledge Management forms, search for and specify other possible forms, and relate organisational forms to performance outcomes in terms of formal NPD performance measures. This could lead to the development of a contingency dependent typology of KM organisations in NPD. Moreover, it would be important to research the possible process of convergence, either towards one of the basic types identified, or towards some kind of integrative scheme, approaching an `ideal type' as discussed above. Our study has shown that Knowledge Management is taking root as a subject at the top of the product development strategic agenda in large manufacturing firms, and was considered equally important across the companies and geographical regions that were studied. The sample was far too small to enable any conclusions to be drawn concerning regional differences. We could simply observe that the European company interviewees provided the richest information about their KM practices. On the other hand, American managers indicated a stronger strategic intent to develop and integrate their practices to take them forward, while the Japanese interviewees gave the impression that knowledge transfer and sharing were not seen as something new, but as already embedded in their current NPD processes.