Computing action policies that insure resilience of social and ecological systems
A workshop at ECCS 2009, September 24, 2009
Resilience is generally defined in the framework of dynamic system theory (attractors, attraction domains, bifurcations): a system is resilient to a perturbation when this perturbation keeps the system in domains of attractors where the system shows the desired properties or functions. However, this definition has two strong limitations: (i) It supposes no possibility to modify the dynamics, whereas socio-ecological systems often involve stakeholders and managers who want to define efficient policies of action. (ii) It supposes that a set of the dynamics attractors show the desired properties, whereas these properties could appear only at non-equilibrium states.
This workshop will demonstrate how new methods, based on viability theory, can overcome these limitations and provide means to compute action policies favouring the recovery from perturbations, and maintain some systems in a desired non-equilibrium state.
This approach has nevertheless drawback: it is very demanding computationally. This is a strong limitation on applying it to complex models from ecology, cognitive science or sociology. This is why the workshop will also introduce methods to simplify complex models. These methods, inspired from physics, can, under certain conditions, derive simple dynamical systems (involving a small set of synthetic variables) representing the main features (patterns) of the dynamics observed on a complex individual or agent based model.
Several contributions to this workshop show case studies where these methods for simplifying an individual based model are coupled with the computation of action policies favouring resilience.
- What is resilience? Review, critical assessment, and outlook (Volker Grimm, Camille Roth and Sophie Martin)
- Viability and resilience, concepts from a toy example (Sophie Martin and Guillaume Deffuant)
- Computing viability and resilience, the bottleneck of dimensionality (Laetitia Chapel and Guillaume Deffuant)
- The macro description of individual based models (Federico Vázquez, Cristóbal López and Maxi San Miguel)
- Bridging the gap between structurally realistic models and viability theory in savanna ecosystems (Justin M. Calabrese and Volker Grimm)
Workshop buffet lunch (free to those registered with the workshop)
- Bacterial biofilms (Jean-Denis Mathias and Nabil Mabrouk)
- Viability and resilience in the dynamics of language competition (Xavier Castelló and Maxi San Miguel)
- Resilience in the dynamics of collaborative Web communities: Three case studies (Dario Taraborelli, Camille Roth and Nigel Gilbert)
- Resilience and viability problems in social dilemma modelling (David Chavalarias and Laeticia Chapel)
- Panel discussion
The speakers are all members of the PATRES: Pattern Resilience project, which is supported by the New and Emerging Science and Technology programme of the sixth Framework Programme of the European Commission.
Guillaume Deffuant, Cemagref, Groupement Clermont-Ferrand, 24, avenue des Landais, B.P. 50085, 63172 Aubière Cedex 1, France <firstname.lastname@example.org>
Nigel Gilbert, Centre for Research in Social Simulation, University of Surrey, Guildford, UK <email@example.com>