Actors, Intentions and Randomness: Evolution of Regional Innovation Networks
Wednesday 14 March 2012, 15:00
LT12, Management School
Ozge Dilaver Kalkan, University of Surrey
Abstract: It is widely accepted that innovations occur through complex interactions within and between organisations. Successful collaborations gradually form distinctive interactive structures known as innovation networks (Breschi and Malerba, 2005; Pyka, 2007). These networks consist of (i) inter-personal and inter-organisational relationships, (ii) shared habits of thought that inform the nature of these relationships and (iii) distributed intelligence and knowledge. It has also been argued, particularly in evolutionary economic geography that regional factors often play an important role in the formation of these complex networks (Frenken, 2006; Martin and Sunley, 2007; Glückler, 2007). Innovation networks facilitate and shape innovative interactions, by enabling sharing of information, expertise and workload. Being distinctive social structures, innovation networks introduce elements of path-dependency and regularity to innovation processes. At the same time innovation networks are far from static, they continuously evolve through the cumulative effects of individual and organizational behaviour.
The emergence and evolution of innovation networks are shaped by the nature and dynamics of the complex interactions they enable. The intrinsic complexity of these networks has made it difficult for traditional modelling techniques to provide results and robust explanations of their dynamics. Agent-based models, on the other hand, have been able to reproduce a number of stylised facts about processes of change in complex networks which are not well accounted for by existing analytical tools. The SKIN model (Ahrweiler, Pyka and Gilbert, 2004) is one of the agent-based models, which have been used for investigating the key factors determining the evolution of innovation networks (see, for example, Korber, 2009). The agents in the SKIN model are firms that possess different knowledge bases (kenes), firms’ kenes evolve through mutations and cross-over and the model allows investigation of the development of collaborative interactions among firms.
In this work we present fundamental extensions to the SKIN model. We employ a multi-level modelling approach incorporating not only firms as unit of analysis but also representing individual actors and their interactions within the firm. This additional level enriches the analysis by serving three purposes. First, it allows us to represent innovative processes within organisations and thereby connecting organisational learning literature to agent-based models of innovation networks. In particular, we synthesise March’s (1991) model on exploration and exploitation and the SKIN model. Second, the model incorporates intentionality in the innovation process while also allowing for unexpected results and randomness. Third, our multi-level model accommodates labour mobility, which is particularly useful to follow patterns of knowledge accumulation in a region and the role of often-ignored commercially unsuccessful innovation activities in this accumulation. The paper compares simulation results with structures that are found in well-known examples of regional innovation networks, such as industrial districts in traditional sectors and high-tech clusters.
Ahrweiler, P.; Pyka, A. and Gilbert, N. (2004) Simulating knowledge dynamics in innovation networks, in Leombruni, R. and Richiardi, M. (eds.) Industry and Labor Dynamics: The Agent-based Computational Economics Approach, 284-96, Singapore: World Scientific.
Breschi, S. and Malerba, F. (2005) Clusters, networks and innovation. Oxford: Oxford University Press.
Frenken, K. (2006) Technological innovation and complexity theory, Economics of Innovation and New Technology, 15: 2, 137 — 155
Glückler, J. (2007) Economic geography and the evolution of networks, Journal of Economic Geography, 7 (5), 619-634.
Korber, M. (2009) The Impact of Public RTI Funding in Biotechnology Innovation Systems: An Agent-Based Approach, Proceedings of DRUID-DIME Academy Winter PhD Conference 2009.
March, J.G. (1991) Exploration and Exploitation in Organizational Learning, Organization Science, 2 (1), 71-87.
Martin, R. and Sunley, P. (2007) Complexity Thinking and Evolutionary Economic Geography, Journal of Economic Geography, 7 (5), 573-601.
Pyka, A. (2007) Innovation networks, in Hanusch, H. and Pyka, A. (2007) Elgar Companion to Neo-Schumpeterian Economics. Cheltenham: Edward Elgar.