Methods and Applications of Meta-Heuristics

Abstract

This special issue presents 5 articles that cover the methods and applications of meta-heuristics and other related models, which highlight innovation and recent development for challenging problems in industry and science. The issue begins with an article by Bouhmala and Granmo, “Stochastic Learning for SATEncoded Graph Coloring Problems”. The article presents a new algorithm based on the random walk technique for graph coloring problem using finite learning automata. The results of the new algorithm are illustrated by a benchmark set containing SAT-encoding graph coloring. The article “A Reinforced Tabu Search Approach for 2D Strip Packing” by GómezVillouta et al. presents a local search algorithm based on Tabu Search to a particular “packing” problem known as 2D Strip Packing (2D-SPP). The authors introduce a fitness function targeting the specific features in the packing problems and present a diversification scheme in Tabu Search to improve the search strategy. Due to the preliminary successful experience on 2D-SPP, the authors plan to extend this approach to other optimization problems. He and Wang’s paper on “The Analysis of Zero Inventory Drift Variants Based on Simple and General Order-Up-To Policies” presents an order-up-to model for demand forecasting to eliminate the inherent offset in the supply chain. Two zero inventory drift variants are evaluated via spreadsheet simulation and proved to be effective. In Li’s paper entitled “ Pagenumber and Graph Treewidth”, the author considers the problem of book-embedding of graph and assign the edges in graph to the pages of the book. This article investigates the relationship between pagenumber and treewidth of G and presents the results with improved upper bounds. The article by Nguyen and Kachitvichyanukul, “Movement Strategies for Multi-Objective Particle Swarm Optimization” reports the application on multi-objective optimization using an improved particle swarm optimization algorithm. The authors present the framework on the improve algorithm and present the results on a set of test problems in Engineering Design and portfolio optimization. The results indicated that the proposed algorithm are very effective in high dimensional problems. We are pleased with the response to the special conference and call for papers, and would like to take this opportunity to thank all of the contributors and anonymous reviewers for their support. The support from the Editor-in-Chief and editorial staff at IJAMC are also greatly appreciated.

Topics

0 Figures and Tables

    Download Full PDF Version (Non-Commercial Use)