An improved evolutionary multi-objective optimization method is introduced to solve the general multi-objective optimization problem with the constraint functions and the different important degree of objective functions. In this method the genetic algorithm is used to search the global optimum solutions. The group is divided into unfeasible domain and feasible domain. Unfeasible solutions are sorted in order by the fitness. Feasible solutions are sorted in order by fuzzy method. Elitist preservation mechanism and the crowding distance are adopted to enhance the efficiency of searching. An example is given to demonstrate this algorithm. The results showed that the improved algorithm is an efficient method to solve the preference in optimization. The algorithm operated fast convergence speed of solution process, high accuracy, and can obtain the global optimum solutions.