[Departement_IRSTEA]EA [TR1_IRSTEA]TESAD / CASYSInternational audienceA general approach to practical inference with gradual implicative rules and fuzzy inputs is presented. Gradual rules represent constraints restricting outputs of a fuzzy system for each input. They are tailored for interpolative reasoning. Our approach to inference is founded on the use of well-conditioned fuzzy systems. It is based on the computation of the fuzzy output under an interval-valued input. A double decomposition of fuzzy inputs is done in terms of @-cuts and in terms of a partitioning of these cuts according to areas where only two rules apply.Cet article propose une methode générique d'inférence avec des règles implicatives graduelles et des entrées floues....