Heuristics and Metaheuristics for heavily constrained hybrid Flowshop Problems

Autor/en:
T. Urlings
Umfang:
348
EAN/ISBN:
978-3-89838-331-8
Erscheinungsdatum:
Freitag, 31. Dezember 2010
Band:
331
Ausgabe:
softcover
Buchreihe:
Dissertationen zur Künstlichen Intelligenz
Kategorien:
Buch
Informatik
Künstliche Intelligenz
Allgemeine Computer- und Kommunikationswissenschaft
Dissertationen zur Künstlichen Intelligenz
Englisch
Gesamtverzeichnis AKA Verlag
Preis:
50,00 €
inkl. 7% MwSt.
Many researchers agree about the existence of a gap between machine scheduling literature on the one hand and scheduling in real-world production environments on the other hand. This book aims to contribute in diminuishing this gap. A hybrid flowshop problem with a combination of many constraints that occur in reality, is adressed. An exact solution method is developped, and some heuristics. Also, a number of metaheuristics is adapted to the problem and several new ones are presented. Comparison is done, enhancing statistical analyses. The solution representation appears to have an important influence on the quality of the results. Multiobjective scheduling is a common need in practice. For Pareto optimisation in a hybrid flowshop, two algorithms and a complete methodology for the comparison of approximation algorithms are presented. From the obtained results, one can conclude that methods that work well for relatively simple, theoretical problems do often not behave so well for more complex real-life environments. Therefore, research to this latter kind of problems is highly relevant.