2013 | OriginalPaper | Buchkapitel
On the Integration of Theoretical Single-Objective Scheduling Results for Multi-objective Problems
verfasst von : Christian Grimme, Markus Kemmerling, Joachim Lepping
Erschienen in: EVOLVE- A Bridge between Probability, Set Oriented Numerics and Evolutionary Computation
Verlag: Springer Berlin Heidelberg
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
We present a modular and flexible algorithmic framework to enable a fusion of scheduling theory and evolutionary multi-objective combinatorial optimization. For single-objective scheduling problems, that is the optimization of task assignments to sparse resources over time, a variety of optimal algorithms or heuristic rules are available. However, in the multi-objective domain it is often impossible to provide specific and theoretically well founded algorithmic solutions. In that situation, multi-objective evolutionary algorithms are commonly used. Although several standard heuristics from this domain exist, most of them hardly allow the integration of available single-objective problem knowledge without complex redesign of the algorithms structure itself. The redesign and tuned application of common evolutionary multi-objective optimizers is far beyond the scope of scheduling research. We therefore describe a framework based on a cellular and agent-based approach which allows the straightforward construction of multi-objective optimizers by compositing single-objective scheduling heuristics. In a case study, we address strongly NP-hard parallel machine scheduling problems and compose optimizers combining the known single-objective results. We eventually show that this approach can bridge between scheduling theory and evolutionary multi-objective search.