INTERNATIONAL JOURNAL OF ADVANCED MANUFACTURING TECHNOLOGY, sa.5-6, ss.551-565, 2007 (SCI-Expanded)
This paper considers the m-machine no-wait flowshop problem with the objective of minimizing the maximum lateness where setup times are considered as separate from processing times and treated as sequence-independent. A dominance relation is developed for the case of three machines and several heuristics and four new effective and efficient genetic algorithms are proposed. The genetic algorithms make use of advanced concepts like steady-state and elitist generational scheme as well as new fast-selection operators. Extensive experimentation is conducted to evaluate the performance of the dominance rule, the proposed heuristics, and the genetic algorithms. The computational and statistical analyses by means of experimental designs show that the genetic algorithms provide better results than the existing literature under the same conditions. Additionally, the proposed dominance rule shows great potential for instances where the processing and setup times are tightly distributed.