Experimentation on Iterated Local Search Hyper-heuristics for Combinatorial Optimization Problems

Designing effective algorithms to solve cross-domain combinatorial optimization problems is an important goal for which manifold search methods have been extensively investigated. However, finding optimal combination of perturbation operations solving hard because the different characteristics each problem and discrepancies in strengths operations. The algorithm that works effectively one domain may completely falter instances other problems. objectives this study are describe three categories a hyper-heuristic combine low-level heuristics with acceptance mechanism problems, compare against existing benchmark experimentally determine effects heuristic categorization on standard from flexible framework. based Thompson sampling iterated local control behavior search. performances configurations were tested Study findings suggested most improved performance when compared hyper-heuristics investigated be good balance between “single shaking” “double strategies. not only provide foundation establishing comparisons but also demonstrate alternative investigate complex