Documents disponibles écrits par cet auteur

An Improved particle swarm optimization algorithm for the hybrid flowshop scheduling to minimize total weighted completion time in process industry / Lixin Tang in IEEE Transactions on control systems technology, Vol. 18 N° 6 (Novembre 2010) 

Public ISBD [article]  in IEEE Transactions on control systems technology > Vol. 18 N° 6 (Novembre 2010) . - pp. 1303-1314 Titre : An Improved particle swarm optimization algorithm for the hybrid flowshop scheduling to minimize total weighted completion time in process industry Type de document : texte imprimé Auteurs : Lixin Tang, Auteur ; Xianpeng Wang, Auteur Année de publication : 2011 Article en page(s) : pp. 1303-1314 Note générale : Génie Aérospatial Langues : Anglais (eng) Mots-clés : Hybrid  flowshop  scheduling  (HFS)  Hybrid  simulated  annealing  Hybrid  variable  neighborhood  search  Improved  particle  swarm  optimization  Three  level  population  update  method Index. décimale : 629.1 Résumé : In this paper, we present an improved particle swarm optimization (PSO) algorithm for the hybrid flowshop scheduling (HFS) problem to minimize total weighted completion time. This problem has a strong practical background in process industry. For example, the integrated production process of steelmaking, continuous-casting, and hot rolling in the iron and steel industry, and the short-term scheduling problem of multistage multiproduct batch plants in the chemical industry can be reduced to a HFS problem. To make PSO applicable in the HFS problem, we use a job permutation that is the processing order of jobs in the first stage to represent a solution, and construct a greedy method to transform this job permutation into a complete HFS schedule. In addition, a hybrid variable neighborhood search (VNS) incorporating variable depth search, a hybrid simulated annealing incorporating stochastic local search, and a three-level population update method are incorporated to improve the search intensification and diversification of the proposed PSO algorithm. Computational experiments on practical production data and randomly generated instances show that the proposed PSO algorithm can obtain good solutions compared to the lower bounds and other metaheuristics. DEWEY : 629.1 ISSN : 1063-6536 En ligne : http://ieeexplore.ieee.org/xpl/freeabs_all.jsp?arnumber=5357401 [article] An Improved particle swarm optimization algorithm for the hybrid flowshop scheduling to minimize total weighted completion time in process industry [texte imprimé] / Lixin Tang, Auteur ; Xianpeng Wang, Auteur . - 2011 . - pp. 1303-1314. Génie Aérospatial Langues : Anglais (eng) in IEEE Transactions on control systems technology > Vol. 18 N° 6 (Novembre 2010) . - pp. 1303-1314 Mots-clés : Hybrid  flowshop  scheduling  (HFS)  Hybrid  simulated  annealing  Hybrid  variable  neighborhood  search  Improved  particle  swarm  optimization  Three  level  population  update  method Index. décimale : 629.1 Résumé : In this paper, we present an improved particle swarm optimization (PSO) algorithm for the hybrid flowshop scheduling (HFS) problem to minimize total weighted completion time. This problem has a strong practical background in process industry. For example, the integrated production process of steelmaking, continuous-casting, and hot rolling in the iron and steel industry, and the short-term scheduling problem of multistage multiproduct batch plants in the chemical industry can be reduced to a HFS problem. To make PSO applicable in the HFS problem, we use a job permutation that is the processing order of jobs in the first stage to represent a solution, and construct a greedy method to transform this job permutation into a complete HFS schedule. In addition, a hybrid variable neighborhood search (VNS) incorporating variable depth search, a hybrid simulated annealing incorporating stochastic local search, and a three-level population update method are incorporated to improve the search intensification and diversification of the proposed PSO algorithm. Computational experiments on practical production data and randomly generated instances show that the proposed PSO algorithm can obtain good solutions compared to the lower bounds and other metaheuristics. DEWEY : 629.1 ISSN : 1063-6536 En ligne : http://ieeexplore.ieee.org/xpl/freeabs_all.jsp?arnumber=5357401

A Two-phase heuristic for the production scheduling of heavy plates in steel industry / Lixin Tang in IEEE Transactions on control systems technology, Vol. 18 N° 1 (Janvier 2010) 

Public ISBD [article]  in IEEE Transactions on control systems technology > Vol. 18 N° 1 (Janvier 2010) . - pp. 104-117 Titre : A Two-phase heuristic for the production scheduling of heavy plates in steel industry Type de document : texte imprimé Auteurs : Lixin Tang, Auteur ; Xianpeng Wang, Auteur Année de publication : 2011 Article en page(s) : pp. 104-117 Note générale : Génie Aérospatial Langues : Anglais (eng) Mots-clés : Decision  tree-based  heuristic  Heavy  plate  Hotrolling  scheduling  Reheating  furnace  scheduling  Scatter  search Index. décimale : 629.1 Résumé : In this paper, the production scheduling of heavy plates is considered. Different from general hot rolling mill, the reheating furnace scheduling shares the same importance with the hot rolling scheduling due to the special characters of heavy plates such as the large differences in weight and dimension. One major characteristic of the reheating furnace is that it can simultaneously process multiple slabs. When the furnace is full, a slab can be drawn in only if the head slab in the furnace has been drawn out. Therefore, the practical heating time of a slab is variable and besides the weight and dimension, is also dependent on not only the slab heating sequence but also the furnace capacity. Since there is no buffer between the reheating furnace and the hot rolling line, a finished slab will reside in the furnace until the hot rolling line is available. To solve this problem, a two-phase heuristic is proposed. In the first phase, only the hot rolling scheduling problem is considered and a scatter search is developed for it. In the second phase, a decision tree-based heuristic is constructed to obtain the corresponding reheating furnace schedule. The experimental results on real production data show that the two-phase heuristic outperforms the current manual scheduling system. DEWEY : 629.1 ISSN : 1063-6536 En ligne : http://ieeexplore.ieee.org/xpl/freeabs_all.jsp?arnumber=5337884 [article] A Two-phase heuristic for the production scheduling of heavy plates in steel industry [texte imprimé] / Lixin Tang, Auteur ; Xianpeng Wang, Auteur . - 2011 . - pp. 104-117. Génie Aérospatial Langues : Anglais (eng) in IEEE Transactions on control systems technology > Vol. 18 N° 1 (Janvier 2010) . - pp. 104-117 Mots-clés : Decision  tree-based  heuristic  Heavy  plate  Hotrolling  scheduling  Reheating  furnace  scheduling  Scatter  search Index. décimale : 629.1 Résumé : In this paper, the production scheduling of heavy plates is considered. Different from general hot rolling mill, the reheating furnace scheduling shares the same importance with the hot rolling scheduling due to the special characters of heavy plates such as the large differences in weight and dimension. One major characteristic of the reheating furnace is that it can simultaneously process multiple slabs. When the furnace is full, a slab can be drawn in only if the head slab in the furnace has been drawn out. Therefore, the practical heating time of a slab is variable and besides the weight and dimension, is also dependent on not only the slab heating sequence but also the furnace capacity. Since there is no buffer between the reheating furnace and the hot rolling line, a finished slab will reside in the furnace until the hot rolling line is available. To solve this problem, a two-phase heuristic is proposed. In the first phase, only the hot rolling scheduling problem is considered and a scatter search is developed for it. In the second phase, a decision tree-based heuristic is constructed to obtain the corresponding reheating furnace schedule. The experimental results on real production data show that the two-phase heuristic outperforms the current manual scheduling system. DEWEY : 629.1 ISSN : 1063-6536 En ligne : http://ieeexplore.ieee.org/xpl/freeabs_all.jsp?arnumber=5337884