This paper develops an approach based on optimization to allow for the modeling of polymerization reactors even if no information about the specific monomer is available. The algorithm is robust, allowing us to deal with different kinds of polymers, and it is built with practical and easy-to-measure responses in mind. It needs only a minimum of experimental information, thus becoming readily applicable to industrial situations. This approach can be applied successfully if a few or even no parameters (kinetic and diffusional) are available. The first part of the paper will revisit diffusional effects during polymerization, proposing robust expressions for gel, glass, and cage effects, with a minimum number of parameters to be fitted. The second part will estimate kinetic and diffusional parameters at the same time, assuming that no information about the polymer system is available. Results are compared with a great quantity of experimental data.