The skyline operator is a powerfulmeans in multi-criteria decision-making since it retrieves the most interesting objects according to a set of attributes. On the other hand, uncertainty is inherent in many real applications. One of the most powerful approaches used to model uncertainty is the evidence theory. Databases that manage such type of data are called evidential databases. In this paper, we tackle the problem of skyline analysis on evidential databases. We introduce a skyline model that is appropriate to the evidential data nature and we present an experimental evaluation.