For an aircraft assembly workshop with dynamic products arrival, the self-evolution problem of knowledgeable manufacturing systems is studied. The principle of self-evolution is applied to the workshop. An algorithm is proposed to solve the self-evolution problem of the workshop in the knowledgeable manufacturing environment. According to the bi-level programming theory, a general mathematical model of the static decision problem at each decision moment is established,
and a bi-level genetic algorithm based on feasible region(FR-BiGA) is designed to solve the model. Simulation results demonstrate the effectiveness and feasibility of the proposed model and algorithms. Experimental data show that the system with self-evolution operations has comparatively better production performance.