Sector-Driven Co-Evolution of Regional Networks and Agent Locations

The spatial economic properties of regional systems are increasingly determined not by geographic distance but by human-built networks of spatial technologies -- the networks of roads, tracks, shipping services, and communications networks that mediate our opportunities and interactions. Agents select locations in part based on the access to other agents provided by such networks. In turn, development of network infrastructure responds to agent agglomerations. This study develops agent-based computational laboratory experiments to systematically explore the co-evolution of regional networks and agent locations as a function of the proportions of economic sectors within an economy, where sectors are differentiated by the capabilities and access objectives of their constituent agents. Regional networks are modelled and studied as synthetic "Geographic SmallWorlds," building on the concept of a "small-world network" recently developed by D. Watts and S. Strogatz. Agent-based experiments explore the cumulative effects of dominant economic sectors and path-dependent co-evolution in regional networks.