In the present study, a CO2/NH3 cascade refrigeration system is modeled and analyzed from exergetic, economic and environmental points of view. In the next step, multi-objective optimization using genetic algorithm method is employed to achieve the optimal design parameters of the system. The exergetic efficiency and the total cost rate of the system (including the capital and maintenance costs, operational cost, and social cost of CO2 emission) are taken into account as objective functions. A set of optimal solutions is achieved and a final optimum point is then chosen using TOPSIS decision-making method. A sensitivity analysis has been performed to study the effect of variation of the unit cost of electricity on the achieved optimal solutions. While operating at the achieved optimum design, the variation of the exergy destruction of each component and the whole system due to the change in the cooling load has also been investigated. The optimization results demonstrate that for the considered plant with cooling capacity of 50 kW, the optimum design selected from the Pareto front results in exergetic efficiency of 45.89%, while it leads to the total cost rate of 0.01099 US$/s.