Parameterizing equations of state is often a compromise between minimizing experimental effort and maximizing accuracy of the calculated thermodynamic properties. Parameter prediction Methods can avoid experimental effort altogether. Alternatively, we need at least as many experimental property data points as the number of model parameters to parameterize the equation of state. Thus, for PCP-SAFT, we need at least three experimental data points corresponding to the three parameters required to model non-associating components. In this work, we present a systematic approach to combine predicted parameters with data points chosen by Optimal Experimental Design. This approach allows to find a trade-off between experimental effort and accuracy of the estimated thermodynamic properties. Predicted parameters yield an average RMSD (root mean square relative deviation) of 40% from vapor pressure data. Combining predicted parameters with one optimally chosen data point for vapor pressure yields already an accuracy of 4% average RMSD in vapor pressure. The presented approach allows to obtain PCP-SAFT parameters which are more accurate than the predicted parameters at a lower experimental effort than required for parameter fitting without predictive information. Our results can serve as a guide for the practitioner to evaluate the minimal experimental effort necessary in order to reach a desired accuracy. (C) 2016 Elsevier B.V. All rights reserved.