(19th ICPR) Distributed Manufacturing as Co-evolutionary System

Rob Dekkers (1)

1 Business School

Research into Distributed Manufacturing – originally focusing on control of autonomous production cells - has embraced over time the challenges of research into industrial networks and more and more identical issues are pursued in both of these fields. Existing strands of research in networks often explore social-dynamic relationships and contractual aspects, thereby ignoring the underlying dynamics based on characteristic issues: collaboration, decentralisation of decision-making and inter-organisational integration. Therefore, theories relating to the loosely connected regime of networks should account for both the instability caused by the autonomous behaviour of agents and the collaboration necessary for sustainability and inter-organisational integration (all pointing to mutual dependencies). Within evolutionary (biological) models, co-evolution has gained a prominent place in the description of mutual relationships for collaboration. Essential to the modelling of co-evolution is the combined development of agents involved, expressed by the factor for connected traits in the NK[C] model. However, in this model co-evolution happens in semi-static landscapes, which hardly exist in reality. Hence, more advanced game-theoretic applications might serve as a foundation for understanding the development of networks since these describe the interactions between agents. This paper expands on co-evolutionary models and it includes the autonomous development of agents in a network, the connectivity between agents and the dynamic forms of collaboration and communication to advance research in Distributed Manufacturing.

Mots clés: Physical Sciences

Rob Dekkers. (19th ICPR) Distributed Manufacturing as Co-evolutionary System. International Journal of Production Research, 2009, 47 (08), pp.2031-2054. ⟨10.1080/00207540802350740⟩. ⟨hal-00513044⟩