This review paper presents an optimization technology for the computer-aided molecular design of environmentally friendly solvents. The approach combines a stochastic optimization method and group contribution methods (GC-methods) to design solvents with desirable physicochemical and environmental properties. A simulated annealing algorithm is used to investigate feasible molecular structures. The main objective method is adopted to balance the multi-objective functions. One property is chosen as the main objective function, while the other properties are considered as constraints, and thus, the multi-objective problem is transformed into a single objective one. The optimal solution is a set of molecules satisfying the formulated target. The properties of each molecule are evaluated through GC-methods, including pure component properties, mixture properties and environmentally properties. Finally, the proposed methodology is illustrated with several examples of industrial separations.