Intensification and diversification strategies with tabu search: One-machine problem with weighted tardiness objective

Abstract

This article investigates several intensification and diversification strategies for the one machine problem with weighted tardiness objective. The aim of this study was to achieve a balance between intensification and diversification strategies. The use of intensification by decomposition, path relinking, frequency-based memory and large step optimization in intensification and diversification process, are combined in several procedures. We perform several computational experiments to study the effect of several combination of these strategies. Our result indicates that combined "large step optimization" with "intensification-diversification approach" and adicional intensification with "path relinking" achieve the betters performance. © Springer-Verlag Berlin Heidelberg 2000.