Could we compute, at the end of the simulation the amount of households that use more cars than they possess ?