Face plastic surgery (PS) plays a major role in today medicine. Both for reconstructive and cosmetic surgery, achieving harmony of facial features is an important, if not the major goal. Several systems have been proposed for presenting to patient and surgeon possible outcomes of the surgical procedure. In this paper, we present a new 3-D system able to automatically suggest, for selected facial features as nose, chin, etc., shapes that aesthetically match the patient's face. The basic idea is suggesting shape changes aimed to approach similar but more harmonious faces. To this goal, our system compares the 3-D scan of the patient with a database of scans of harmonious faces, excluding the feature to be corrected. Then, the corresponding features of the k most similar harmonious faces, as well as their average, are suitably pasted onto the patient's face, producing k+1 aesthetically effective surgery simulations. The system has been fully implemented and tested. To demonstrate the system, a 3-D database of harmonious faces has been collected and a number of PS treatments have been simulated. The ratings of the outcomes of the simulations, provided by panels of human judges, show that the system and the underlying idea are effective.