We propose a modified evolutionary computation method to solve the optimization problem of additively decomposed function with constraints, ft is based on factorized distribution instead of penalty function and any transformation to a linear model or others. The feasibility and convergence of the new algorithm are given. The numerical results show that the new algorithm gives a satisfactory performance. We propose a modified evolutionary computation method to solve the optimization problem of additively decomposed function with constraints, ft is based on factorized distribution instead of penalty function and any transformation to a linear model or others. The feasibility and convergence of the new algorithm are given. The numerical results show that the new algorithm gives a satisfactory performance.