Reactive Scheduling Systems using Tabu Search

Tan Tiong Guan

Abstract

The aim of this project is to develop a reactive scheduling system for a highly dynamic and unpredictable environment. We have earlier on used genetic algorithm. In this version, Tabu Search was used instead. Although GA has been shown to be successful, the effectiveness of other heuristic techniques have yet to be explored. A Tabu Search scheduler was implemented using C++ to search for the best solutions. The solutions from the scheduler is then compared with the solution obtained among the runs. It was found that the table search scheduler was also capable of producing good solutions within a short period of time.