ECCS'06: A One-Day Satellite Workshop:
Thursday 28th September 2006
Complexity and Dynamics: Volatility & Stability in City & Regional Systems

Background to Workshop:
Some Key Questions

To set the context to this workshop, we will define a number of key questions.

When we look at cities, there are always questions which pose conundrums. The classic question of why cities remains largely ordered without any extensive top down planning is of course a key question of complexity theory and has been pondered for many years. But one of the key issues is 'why do cities appear volatile and fast moving at small scales while at the large scales they appear persistent and unchanging?' The City of London has been rebuilt four or five times since the Second World War and everything that goes on in its buildings are very different from 50 years ago, but in one sense it is unchanged. Moreover when we look at ordered patterns in cities in the way land uses are laid out and the way transport networks serve these uses in an efficient space-filling manner we see great regularity whose signatures are captured by distributions which scale. These are two obvious questions that complexity theory is seeking to address in terms of cities but there are many others. One particularly important question, for example, is how public authorities and policy makers can intervene successfully in such systems if there is no clear understanding of the multi-level stability and instabilities that characterize the urban system. Indeed, it is perhaps important to make public policy act in a guiding manner "with" the natural forces of evolution of the system, arising from the interconnected actions of multiple agents, or is public policy about either attempting provoke change that otherwise will not happen, or about trying to thwart changes that are seen as potentially harmful. This shows us the absolutely basic importance for policy making of trying to understand these questions. The workshop will focus on theories and models that attempt to develop simulations that are potentially capable of addressing such questions and we will entice speakers and discussants to frame their presentations in these terms as well as some policy representatives setting out the underlying questions that they need help in addressing. Temporal dynamics of course is central to all such questions and we will focus the meeting through this lens.

There is, in fact, a long tradition of treating cities and the wider systems of regions in which they exist using a systems approach and simulation models of their patterning and functioning have existed since the late 1950s. In some senses, this area is unique in that large scale empirical models which are used routinely in practice have been a feature of these developments for many years and this makes the area somewhat different from other application of complexity theory and simulation in the social sciences. Nevertheless, practical models of city and regions usually do not embrace complexity theory in that it is the norm that these models simulate urban land use and structure at single cross sections in time assuming an equilibrium that to all intents and purposes is absent. There are good reasons for this, related to the need for operationality but as part of the quest to develop better models for policy purposes, a complex systems perspective has emerged in this field which stresses the need for much more disaggregate dynamics models where equilibrium is the exception rather than the rule.

There is now a substantial body of work in this area which falls into several distinct categories which we define as themes below. Essentially the workshop will be about temporal dynamics, equilibrium and related approaches to city and regional systems which are represented as land use, transport flows and infrastructure and eco-demographic activities. The focus will be on various theories and models which are and have been developed for both academic and policy purposes. Many of these themes of course merge into one another while the focus of substantive interest is on development in physical terms such as through land use location and/or on activity allocation in which activity is treated as individuals in the population performing tasks such as employment of various sorts of residential location and so on. The interface between new dynamic, multi-scale modelling and the policy and public sector needs is of particular importance if the full potential of complexity science is to be used successfully.

The topics likely to be covered will comprise the following:

• Aggregate models based on catastrophe and bifurcation theory, dealing with the evolution of systems following the work of Prigogine, Haken, and others.
  
• Disaggregate 'agent-based' models following various traditions in the social sciences, some coming from agricultural and land cover analysis, others from market based analysis.
  
• Land development models in which the focus is on representation urban growth using cellular automata and focussing on emergent structures. These models fuse with agent-based although CA models tend to be more physically based.
  
• Micro-simulation techniques generalise samples of spatial behaviour to the wider system and follow the tradition of Orcutt and others.
  
• Urban economic models associated with new development in location theory based on trade and development following Fujita and Krugman.
  
• Social and statistical physics approaches which deal with aggregate behaviours often from random structured processes.
  
• The needs for integrated, multi-level understanding for policy and public sector decision making.