Intelligent optimization is a domain of evolutionary computation that emerges since a few years. All the methods within this discipline are based on mechanisms for maintaining a set of individuals and, separately, a space of knowledge linked to the individuals. The aim is to make the individuals evolve to reach better solutions generation after generation using the knowledge linked to them. The idea proposed in this paper consists in using previous experiences in order to build the knowledge referential and then accelerate the search process. A method which allows reusing knowledge gained from experience feedback is proposed. This approach has been applied to the problem of selection of project scenario in a multi-objective context. An evolutionary algorithm has been modified in order to allow the reuse of capitalized knowledge. This knowledge is gathered in an influence diagram allowing its reuse by the algorithm.