Hybrid evolutionary algorithm for mixed-integer programming problems

A hybrid evolutionary algorithm based on orthogonal experimental design is developed to deal with the mixed-integer programming problems. This hybrid evolutionary algorithm contains the heuristic mutation operator, the crossover operator as which the orthogonal experimental design serves, and the migration operator to keep the population's diversity. Numerical results show that, compared with other available algorithms, the proposed algorithm can converge at optimal solutions of all test problems with higher speed, and it has a smaller number of function evaluations and yields more accurate optimal function values.