New efficiency-stability criterion in a rescheduling problem with dynamic jobs weights

Abstract

In the context of the fourth revolution, today’s industry is developing new technologies and solutions for the production management. Moreover, customers’ practices are also changing, they can modify or cancel an order at any time, thus impacting the already established planning. Therefore, rescheduling is necessary to revise the initial schedule in a cost-effective manner. In this work, a new performance measure is investigated, considering simultaneously the weighted waiting time as an efficiency criterion, and the total completion time deviation as a stability criterion. The objective function which considers several criteria, could be very significant and helpful for industrial actors. In addition, an innovative concept is introduced, consisting in increasing the jobs weight at each time they are rescheduled, to take fairly into account the low weight earliest jobs. This paper addresses a parallel machine rescheduling problem, disrupted by new jobs arrival. Based on predictive reactive strategy, a linear mathematical model is developed, as well as an iterative methodology to solve the described problem. Finally, numerical results are analyzed to show the effect produced by the combination of efficiency and stability criteria.