Examinando por Autor "Velez-Gallego M.C."

Mostrando 1 - 2 de 2
  • No hay miniatura disponible
    Ítem
    An approach for printing the maximum possible number of images
    (Institute of Industrial Engineers, 2017-01-01) Rojas-Santiago M.; Barbosa R.C.; Muthuswamy S.; Hulett M.; Velez-Gallego M.C.; Universidad EAFIT. Departamento de Ingeniería de Producción; Gestión de Producción y Logística
    This paper considers a real life application of a printing process in a lithographic company that manufactures food packaging bags. The focus area of this problem is to print a variety of 2D images of bags on a surface area for maximum press layouts, with a constraint that the orientation of all bags is fixed and they should be placed parallel to the edges of the impression material. The problem under study is NP-hard and is an extension of the classical knapsack problem. An algorithm to maximize the number of rectangular images of bags that can be placed within the printing material is proposed. The results were compared with a Particle Swarm Optimization (PSO) metaheuristic. Several numerical experiments show that our procedure outperforms the PSO algorithm and improves the lithography's performance.
  • No hay miniatura disponible
    Ítem
    A variable block insertion heuristic for permutation flowshops with makespan criterion
    (Institute of Electrical and Electronics Engineers Inc., 2017-01-01) Tasgetiren M.F.; Pan Q.-K.; Kizilay D.; Velez-Gallego M.C.; Universidad EAFIT. Departamento de Ingeniería de Producción; Gestión de Producción y Logística
    This paper proposes a populated variable block insertion heuristic (PVBIH) algorithm for solving the permutation flowshop scheduling problem with the makespan criterion. The PVBIH algorithm starts with a minimum block size being equal to one. It removes a block from the current solution and inserts it into the partial solution randomly with a predetermined move size. A local search is applied to the solution found after several block moves. If the new solution generated after the local search is better than the current solution, it replaces the current solution. It retains the same block size as long as it improves. Otherwise, the block size is incremented by one and a simulated annealing-type of acceptance criterion is used to accept the new solution. This process is repeated until the block size reaches at the maximum block size. In addition, we present a randomized profile fitting heuristic with excellent results. Extensive computational results on the Taillard's well-known benchmark suite show that the proposed PVBIH algorithm substantially outperforms the differential evolution algorithm (NS-SGDE) recently proposed in the literature. © 2017 IEEE.