Fuzzy Logic Approach to Represent and Propagate Imprecision in Agri-Environmental Indicator Assessment.

Zainab Assaghir (1, 2) , Philippe Girardin (2) , Amedeo Napoli (1)

1 ORPAILLEUR - Knowledge representation, reasonning
2 LAE - Laboratoire Agronomie et Environnement

The indicator of groundwater contamination developed and used in agriculture is calculated from data available in the field or data estimated by an expert. The modeling of this indicator generally requires a large number of parameters whose measure is imprecise. Several information sources provide information about the same imprecise quantities which have to be combined for defining what is called an "indicator of groundwater contamination " (Igro). This indicator estimates the impact of cultivation practices on the groundwater contamination. In this paper, we explore a possibilistic information fusion method by using the notion of maximal coherent subsets to represent the imprecisions of multisource variables of the indicator. We also calculate the bounds of this indicator, and we propagate imprecision by using an interval analysis. Finally, we present the indicator's results for pesticides applied on different crops

Mots clés: Possibility theory; maximal coherent subset; interval analysis; indicator; fuzzy set; fusion

Domaines: Intelligence artificielle [cs.AI]

Zainab Assaghir, Philippe Girardin, Amedeo Napoli. Fuzzy Logic Approach to Represent and Propagate Imprecision in Agri-Environmental Indicator Assessment.. Proceedings of the Joint 2009 International Fuzzy Systems Association World Congress and 2009 European Society of Fuzzy Logic and Technology Conference, Jul 2009, Lisbonne, Portugal. pp.707-712. ⟨hal-00606318⟩