Using Multi-Agent Systems in a Companion Modelling Approach for Agroecosystem Management in South-East Asia

Abstract

This paper describes research using multi-agent systems as a companion modelling tool to address key issues related to agroecosystem management in northern Thailand and northern Vietnam. The authors illustrate an approach for the use of complex models for the accompaniment of adaptive management experiences. First, some considerations on the shifts of paradigm that underlie the research are discussed. Then two case studies are presented. The first one illustrates the iterative process of problem solving with local stakeholders, while the second emphasizes the emergence of local institutions in the context of land reforms. In both cases, the research started with an analysis of the agrarian system, which integrated multiscale biophysical and socioeconomic knowledge by means of a model. The research process then evolved towards the use of such models in participatory approaches for community-based natural resource management. Regular interactions between researchers and local stakeholders mediated by the companion modelling tools were helpful in progressing local development.