A computer aided molecular design (CAMD) approach is proposed for generating molecular structures of polymer candidates which have the potential to be effective polymer carriers in drug delivery. A group contribution method has been used to predict properties such as glass transition temperature (Tg), expansion coefficient (Of) and water absorption (W). A mixed integer nonlinear optimization technique is used to successfully generate some novel polymer structures not available in the literature. The generated polymers are ranked according to a desirability curve. A solubility parameter analysis is done to further analyze and filter the generated polymers. The proposed new polymers need to be synthesized and analyzed to make them practical for use in oral drug delivery. (C) 2014 Elsevier Ltd. All rights reserved.