Forming hyper-heuristics with GAs when solving 2D-regular cutting stock problems

This paper presents a method for combining concepts of hyper-heuristics and genetic algorithms for solving 2D cutting stock problems. The idea behind hyper-heuristics is to find some combination of simple heuristics to solve a wide range of problems. To be worthwhile, such combination should outperform the single heuristics. When tackling optimization problems, a genetic algorithm (GA) has been often used to evolve individuals coding direct solutions. In this investigation, the hyper-heuristic is formed using a GA which evolves solution procedures when solving individual problems. The method finds very competitive results for most of the cases, when tested with a collection of different problems. The testbed is composed of problems used in other similar studies in the literature. Some additional instances of the testbed were randomly generated.