APPLIED MATHEMATICS/MASCOS SEMINAR

Speaker: Professor Ian Wilkinson
School of Marketing
UNSW

Date: Wednesday October 20
Time: 4pm
Room: Red Centre 3085

Title: Complexity and Modeling the Dynamics and Evolution of Business
Networks

Market systems comprise economic institutions involved in creating and delivering products and services to end-users which operate within networks of relations that "organize" the flows of activities.

Historically, an implicit assumption underlying research into these networks is that the
patterns of relations are controlled by the organizations involved -- the so-called top down view. More recently, we have come to recognize that the relations among organizations may be (possibly unintentional) consequences of the ongoing process of interaction. That is, network structure can emerge in a bottom up self organizing way.

Market systems may be studied in the framework of complex adaptive system theory. "A 'complex system' is a system consisting of many agents that interact with each other in various ways. Such a system is 'adaptive' if these agents change their actions as a result of the events in the process of interaction" (Vriend, 1995, p. 205). Viewed from the perspective of adaptive systems, market interactions depend in a crucial way on local knowledge of the identity of some potential trading partners.

How self-organized market systems emerge in decentralized economies is a question that formal analysis of such systems seeks to answer. The research I describe aims to develop a methodology for studying the processes by which institutional and network structures evolve and the factors driving these processes. The role of agent based simulations and artificial life models are discussed, including in particular the NK Boolean
Models developed by Stuart Kauffman at the Santa Fe Institute. The need for some form of productive linkage between experts in handling the modelling and mathematics of complex systems and those seeking to apply these concepts to
developing grounded models of business networks is considered and the Danish Research Unit for the study of Industrial Dynamics (DRUID) is used as a possible exemplar.