Building hyper-heuristics through ant colony optimization for the 2D bin packing problem Chapter in Scopus uri icon

Abstract

  • Convergence proofs for ant colony optimization are limited [1], only in some cases it is possible to assure that the algorithm will find an optimal solution. It is even more difficult to state how long it will take, but it has been found experimentally that the computing time increases at least exponentially with the size of the problem [2]. To overcome this, the concept of hyper-heuristics could be applied. The idea behind hyper-heuristics is to find some combination of simple heuristics to solve a problem instead than solving it directly. In this paper we introduce the first attempt to combine hyper-heuristics with an AGO algorithm. The resulting algorithm was applied to the two-dimensional bin packing problem, and encouraging results were obtained when solving classic instances taken from the literature. The performance of our approach is always equal or better than that of any of the simple heuristics studied, and comparable to the best metaheuristics known. © Springer-Verlag Berlin Heidelberg 2005.

Publication date

  • December 1, 2005