This article is a contribution to the decision-maker's requirement for quantifying and improving industrial performances. The design and use of Performance Measurement Systems (PMSs) in this sense have received considerable attention in recent years, but only a few quantitative models for PMSs have been proposed for this purpose. Among them, the one proposed by the authors, based on a fuzzy Choquet integral aggregation operator, that allows to express an overall performance according to the relative importance of criteria and interactions between them. As this model is non linear, it is useful to define pieces of information aimed at aiding the decision-maker to optimize the improvement of the overall performance with regards to the allocated resources and the associated costs. Moreover, the consequences of the non complete realization of the optimal improvement strategy have also to be quantified in terms of overall performance decrease or cost increase. The propositions are illustrated on a case study submitted by a SME manufacturing kitchens, bathrooms and storing closets