Dynamic selection of dispatching rules for manufacturing system scheduling

The scheduling strategy proposed is based on a dynamic selection of certain pre-determined dispatching rules. A new selection is carried out each time a machine becomes available, depending on: (1) primary and secondary performance objectives, (2) operating conditions, (3) an analysis of the system state, which aims at detecting particular symptoms from the numerical values taken by state variables. This selection method is improved through the optimization of the numerical thresholds used in the detection of symptoms. A simple heuristic dispatching strategy, called ‘shift from standard rules’ (SFSR), has been designed based on this dynamic selection approach. It is compared with SPT, SI^x, COVERT, CEXSPT, and CR/SPT on a job shop problem, already used in other research works. The results indicate significant improvements on the mean tardiness and the conditional mean tardiness. Moreover, this new approach allows several performance criteria, such as the mean flowtime and the mean tardiness, to be taken into account simultaneously.