A kinetic model has been proposed to describe the propylene polymerization process with long chain branching for a twin catalyst system to fit the experimental evolution of molecular weights, polydispersity index of atactic polypropylene, isotactic polypropylene and the grafting density at different catalyst and cocatalyst concentrations. Kinetic parameters are estimated by real coded genetic algorithm (an evolutionary optimization technique) from experimental data available in open literature. The validated model has the capability of predicting the branching density as a function of catalyst addition pattern, catalyst ratios and copolymerization time. Further, the validated model has been used to calculate the 'molecular weight long chain branching distribution'. Parametric sensitivity study has been conducted to analyze the effect of kinetic parameters on the long chain branching formation and other molecular properties of the polymer. © 2014 Elsevier B.V.