Extending the shifting bottleneck procedure to real-life applications

Abstract

Much research has been devoted to the job shop scheduling problem since its introduction in the late 50's. Despite these efforts, even moderate sized benchmarking problems remained unsolved for many years. Given the complexity of the job shop scheduling problem, there is little hope for solving large real-life problems optimally within reasonable time. We therefore rely on heuristics, of which the Shifting Bottleneck Procedure, developed by Adams et al. (1988), is performing excellently. We examine several extensions of the Shifting Bottleneck Procedure towards real-life applications. We introduce due dates, release dates, assembly structures, split structures, overlapping operations, setup times, transportation times, parallel machines and beginning inventory. This generalized shifting bottleneck procedure is compared with priority dispatching rules on a set of large test problems.

Local search and genetic algorithm for the job shop scheduling problem with sequence dependent setup times

“…This heuristic has also been extended to face practical characteristics of actual job shops other than setup times; for example, in Mönch et al (2007), the authors face a complex job shop problem with setup times and other characteristics such as parallel batching machines and re-entrant process flows, by means of a genetic algorithm hybridized with a local search method based on this heuristic. Ives and Lambrecht (1996) describe a similar approach and consider a larger number of real situations. Also, in Zoghby et al (2005) a neighborhood search with heuristic repairing is proposed that is an extension of the local search methods for the JSP.…”

Job Shop Scheduling

“…As examples, Balas et al [29] include sequence-dependent setup times (setup time before processing begins depends on the prior task), release dates, and deadlines, and Ivens and Lambrecht [30] consider release and due dates, assembly and splitting of jobs, overlapping processing times, setup times, transportation times, and multiple machines at a work station. As a final example, Mason et al [23] minimize weighted tardiness where constraints include multiple and batching machines (several tasks are processed simultaneously) at a work center, sequence-dependent setup times, and reentrant flows (a job visits a machine more than once).…”

Job scheduling with technical constraints

“…Ramudhin and Marier 4 extended the disjunctive graph model for a combination of job-shop, open-shop and assembly-shop structures. Another study by Ivens and Lembrecht 5 proposed a disjunctive graph model to incorporate many real-life scenarios such as set-up times, assembly and job-splitting structures, and job transfer between machines in small sub-batches. These models deal with aspects common to many industrial applications.…”

“…The shifting bottleneck procedure was shown to be quite versatile when applied to more realistic job-shop models. 4,5 This procedure relies on exact (or near-exact) algorithms for single-machine problems. These algorithms are mostly branch-and-bound techniques, which are difficult to implement and lack versatility when the production environment changes.…”