A new multi-objective evolutionary algorithm based on weighted gradient

In this paper, a new multi-objective evolutionary algorithm is proposed. In this algorithm, a mutation operator is presented by the weighted gradient direction to find Pareto solutions, and the weights are defined based on the information of objective function values, the fitness function is formulated based on a possibility degree matrix for the selection operator. The algorithm is implemented on five classical test functions, and compared with other algorithms by using the statistical comparison technique. Numerical experiments show that the proposed algorithm is efficient and feasible.