Thermal and kinetic studies were carried out in a previous work using an adiabatic batch reactor. In this work a very exothermic reaction has been chosen. The reaction is performed by feeding H2O2 to a solution of Na2S2O3. Several experiments have been carried out at different operating conditions (addition flow, initial temperature, initial concentration of reagents, etc.) in semibatch mode of operation with heat transfer. An algorithm describing the mathematical model has been developed and implemented in Matlab software. The mathematical model has been validated with experiments at different addition flows of H2O2, while concentration and heat profiles have been obtained by simulation. Then, an optimization program has been developed to obtain the best combination of inlet flow and addition time, to maintain the reactor temperature at the selected temperature setpoint, and to minimize the reaction time. The optimization method used is based on genetic algorithms (GA). The GA technique is shown to be the most appropriate multivariable method that avoids falling in local maxima and shows satisfactory results in all cases tested.