Résumé
The goal of this paper is to propose a decision support system to manage within field variability. The goal is to provide a method which allows the farmer to (1) decide whether or not the spatial variation of a field allows a reliable variable-rate application (2) to discover if a particular threshold (field segmentation) based on within field data is technically feasible according to the application machinery and (3) to make an appropriate application map. Our method aims at providing improvements as compared to a previous Opportunity index proposed by the literature (Oi and TO i). The paper proposes a Fuzzy Technical Opportunity index (FTOi). This FTOi considers (1) a fuzzy footprint model of a Variable-Rate Application Controller (VRAC), which describes the area within which the VRAC can reliably operate, (2) the location inaccuracy of the data and (3) the ability (accuracy) of the VRAC to perform distinct levels of treatments. The originality of our approach is based on the use of a fuzzy estimation process to decide if a level of treatment is reliable or not on each area over which the VRAC operates. It does not need any pre-processing of the data like kriging. Tests on theoretical fields, obtained from a simulated annealing procedure, showed that the FTOi was able to assess the technically manageability of filed variation. Tests also showed that our approach was able to consider problems of lack of data (data resolution).