Scheduling Jobs on Parallel Machines with Sequence-Dependent Family Set-up Times

Abstract

This paper proposes an efficient heuristic to minimise the total weighted tardiness of a set of tasks with known processing times, due dates, weights and family types for parallel machines. A three-phase heuristic is presented to minimise total weighted tardiness. In the first phase, jobs are listed by the earliest due date and then divided into small job-sets according to a decision parameter. In the second phase, jobs are grouped by the due date within applicable families using apparent tardiness cost with set-up (ATCS), and the sequence of jobs within families is improved through the use of the tabu search method. In the third phase, jobs are allocated to machines using a threshold value and a look-ahead parameter. The comprehensive simulation results show that the proposed heuristic performs better than the ATCS and rolling horizon procedure at a significantly reduced total weighted tardiness.