Solving the single machine total weighted tardiness scheduling problem using a hybrid simulated annealing algorithm

An application of a new hybrid optimization approach for the single-machine total weighted tardiness problem (SMTWTP) is presented in this article. SMTWTP is known Io be NP-hard, and thus the right way to proceed is through the use of heuristic techniques. The proposed approach combines the stochastic sampling of a simulated annealing algorithm (SAA) together with features borrowed from the field of evolutionary algorithms, such as the population of individual solutions and a special unary recombination operator. The performance of the hybrid SSA is tested over multiple benchmark problems with up to 50 jobs on a single machine and compared to that of other previously studied heuristics. The results obtained are in average less than 1% above the known global optimum solutions.