A mid-course correction (MCC) policy has been shown as a practical approach to compensate the effects of the new disturbances on the final particle size distribution (PSD) in the emulsion polymerization process. However, the development of control model remains immature in the practical sense. In this paper, a control strategy to control the bi-modal particle size distribution in an emulsion copolymerization process within the MCC approach is presented. The strategy proposes a controlled adaptive perturbation (CAP) to develop the MCC control model while minimizing the off-specification batches generation during the perturbation. The scheme utilizes the particle size polydispersion index (PSPI) as the control objective to simplify the development process. In addition, it is also shown that the final PSD can be estimated by PLS model with an adequate accuracy by using on-line calorimetry-based conversion measurements as predictor. This eliminates the requirement of off-line analysis thus faster predictions can be made. © Copyright 2007 IFAC.