An improved genetic algorithm for the packing of rectangles

Aiming at the optimal layout problem of rectangular parts with dynamic constraints, a novel heuristic rectangular optimal layout method is proposed based on the improved genetic algorithm. In order to enhance operation speed of the algorithm, the algorithm of population initialization is improved by using the elitist preservation strategy; In order to enhance the global convergence performance of the algorithm, the crossover and mutation operator are improved by using the self-adaptive strategy; in order to enhance the precision of the algorithm, the decoding algorithm is improved by using the height adjustment algorithm. The proposed approach could achieve the layout optimization for both single standard and multi-standard plates. The solution of numerical example indicates that the proposed algorithm is a feasible approach to rectangular parts optimal layout.