We present an improved GA-based tool that calibrates Agent-Based Models (ABMs). The GA searches through a user-defined set of input parameters to an ABM, delivering values for those parameters so that the output time series of an ABM match the real system's time series to certain precision. The improvements are focused on shortening the computational time that the GA needs to find good solutions. Additionally, one of the new mechanisms prevents the new GA from reaching a certain condition under which the original GA does not converge. The results of experiments show that the improved GA fulfills those goals.