A Multi-Agent System for the Simulation of Land Use Organization
Résumé
We built a multi-agent system for simulating land-use organization in farming territories in the north-east of France. Farming territories were described with soil, slope and distance features. Farming systems were described as sets of land-use categories which had needs and soil, slope or distance constraints. Our multi-agent system was inspired by the eco problem solving approach: each land-use categorie was represented by a set of agents which tried to conquer zones to satisfy its needs, fighting each other. Our system was shown convenient for our problem and provided valide solutions on real data sets.