close this window

[ProcessMatter]

ProcessMatter aims at translating urban processes into a digital environment, to study the dynamics that occur between hierarchical and distributed urban systems. The project finds its base in the Wild City research that looks at the mechanisms and effects of the uncontrolled transformation of the city of Belgrade. The outcome is an agent-based simulation software that brings one of the central Wild City practices – street economy – to the digital ground to get a hunch of its evolution and the impact it generates.
 

The graphic interface that surrounds the ProcessMatter simulation software; it enables to set up specific urban layouts and introduce the 'players' (agents) in the environment

[concept]

The core of this research is an interest in the spatial and organisational implications of different degrees of control in urban development, degrees of hierarchy and heterarchy in urban systems, and the effect of certain blends of regulated and non-regulated urban activities. How does an urban environment respond (configure and perform) when the dynamics between such poles are varied or even unleashed

Understanding cities as complex adaptive systems, questions whether architecture (and urbanism) can be open towards ‘unprompted’ innovations; whether it can be flexible in interaction with its environment, and shift from the top-down, unilateral and project-based response towards an open-ended approach? In our view an apparent direction for architects and planners is in influencing, steering and shifting the processes themselves - which means a change of focus from designing objects to designing processes.

The ProcessMatter simulation application is envisioned as a descriptive and investigative device of urban processes. Here, the practical interest in physical form or growth patterns is secondary to interest in the character and behaviour of the urban process itself. At present, the focus is not put on a quantitative quality of the output but rather on the qualitative landscape of the model variables and parameters.

[methodology]

ProcessMatter has been set up as an interdisciplinary research project involving architects, computer scientists, a complex systems expert and an artist. The project developed around a set of intense workshops.

Workshop 3, Rotterdam 2003


First, a modelling concept has been defined to be able to ‘port’ actual urban and architectural phenomena to a digital simulation environment. The choice has been made to create an environment, that blends a so called agent-based-modelling (MAS) with certain forms of control or intervention. As the base for the application, a ‘swarm-like’ simulation framework called RePast (JAVA based) is used.

Dependence between the ProcessMatter software (PM-grey) and RePast


For the first prototype in this line of applications, the processes of street trade has been scripted into a complex set of rules and algorithms. The behaviour of buying individuals and trading entities (shops) and characteristics of the contiguous urban environment (buildings, territories or lines of infrastructure) has been translated into a language that can be used in a simulation application

A simulation starts with an initial state that is set up manually by the user where he introduces (draws) the ‘geography’ of the world and the ‘ecology’ of the agents. The agents behave according to the procedures that describe the local transactions. The user has at any time possibilities to intervene in the setup of the environment (its ‘built’ architecture) and can influence the ‘condition’ of the world through a wide variety of parameters.

Sequence of the simulation tests

[results]

With its first spring-off ProcessMatter has arrived at a demonstration prototype of an urban simulation application – publicly available as an open source software package, open for experimentation. ProcessMatter enables to envision and set up simulation environments that deal with the mix of emergent and hierarchical processes that are typical for urban environments. As it consciously involves shifts between emergent, bottom-up modelling and the impact of more top-down control it is in this respect a novelty in the field of urban simulation.