The multiobjective optimization technique based on the desirability estimation of several interrelated responses (MOOP-DESIRE) has been recently applied to quantitative structure-activity relationship (QSAR) studies. However, the advantage of applying this new methodology to the study of selectivity and affinity to competitive targets has been little explored. We used the MOOP-DESIRE methodology and a variation of this, to study the arylpiperazine derivates that could interact with 5-HT(1A) and 5-HT(2A), serotonin receptor subtypes with the objective of designing more selective molecules for the 5-HT(1A) receptor. We did show that the model results are in agreement with the available pharmacophore descriptions, guaranteeing an appropriate structural correlation and proving the methodology, as a useful tool for the important problem of selective drug design.