A multiobjective optimization formulation is presented to track a desired particle-size distribution (PSD) trajectory in semi-batch emulsion polymerization. This is done in a hierarchical framework, controlling the individual processes of nucleation, growth, and coagulation that determine the PSD. An evolutionary technique, the Nondominated Sorting Genetic Algorithm, is employed as the optimization tool. Experimental implementation of these open-loop optimal recipes yield promising results, but simultaneously highlight the importance of feedback corrective action. Preliminary state/parameter estimation studies are performed, which show an ability to estimate the full distribution and related states from the available measurements (relative distribution latex density, and feed rates), and also elucidate the requirements in the strategy to be employed for estimation. The potential for on-line feedback is examined using experimental data, in view of the irreversibilities and process limitations.