Skip to Main content Skip to Navigation
Theses

Modèles Multi-Agents pour l'Aide à la Décision et la Négociation en Aménagement du Territoire

Abstract : This thesis addresses multi-agent systems (MAS) used for decision and negotiation support for land-use planning. In a first part, we discuss expertise, decision making and negotiation for linear infrastructures planning. We introduce MAS therefore with a design oriented approach. We build on various sociological references to analyze artificial social systems. In a second part, we discuss MAS for modelling, simulating and problem solving, and reactive vs. cognitive MAS. We describe the SMARRPS library dediacted to spatial problem solving. We finally give three formal models assessing convergence. In a third part, we describe two applications : SMAALA, for linear planning. This expertise support system computes a set of alternatives based on environmental sensitivity maps, on structural constraints, and on actors preferences models. We also present SANPA, an internet negotiation support system for land use planning. It uses an assistant agent community. They exchange with project support agents and SMAALA like models. SANPA uses structured messages, based on speach act theory. Nous conclude on some perspectives for environmental information systems using multi-agent approaches.
Document type :
Theses
Complete list of metadatas

Cited literature [134 references]  Display  Hide  Download

https://tel.archives-ouvertes.fr/tel-00003562
Contributor : Nils Ferrand <>
Submitted on : Tuesday, October 14, 2003 - 6:03:46 PM
Last modification on : Friday, November 6, 2020 - 4:13:13 AM
Long-term archiving on: : Friday, April 2, 2010 - 7:10:02 PM

Identifiers

  • HAL Id : tel-00003562, version 1

Collections

Citation

Nils Ferrand. Modèles Multi-Agents pour l'Aide à la Décision et la Négociation en Aménagement du Territoire. Interface homme-machine [cs.HC]. Université Joseph-Fourier - Grenoble I, 1997. Français. ⟨tel-00003562⟩

Share

Metrics

Record views

1283

Files downloads

2830