A Hybrid Genetic Algorithm for the Early/Tardy Scheduling Problem

Abstract

In this paper, we present a hybrid genetic algorithm for a version of the early/tardy scheduling problem in which no unforced idle time may be inserted in a sequence. The chromosome representation of the problem is based on random keys. The genetic algorithm is used to establish the order in which the jobs are initially scheduled, and a local search procedure is subsequently applied to detect possible improvements. The approach is tested on a set of randomly generated problems and compared with existing efficient heuristic procedures based on dispatch rules and local search. The computational results show that this new approach, although requiring slightly longer computational times, is better than the previous algorithms in terms of solution quality.

Biased random-key genetic algorithms for combinatorial optimization

“…For all instances, in absolute terms, algorithm GA-SlackMod obtained earliness, tardiness, and flow time deviation close to the optimum value. Valente et al (2006) describe a BRKGA for a single machine scheduling problem with earliness and tardiness costs and no unforced machine idle time. Such problems arise in just-in-time production, where goods are produced only when they are needed, since jobs are scheduled to conclude as close as possible to their due dates.…”

“…If a swap decreases the cost of the schedule, the swap is done, the cost recomputed, and the scan continues from that job until the last two jobs are tested. Valente et al (2006) tested six BRKGA variants for the early tardy scheduling problem. The genetic algorithms were compared with the NSearch heuristic of Li (1997).…”

A Hybrid Permutation-Coded Evolutionary Algorithm for the Early/Tardy Scheduling Problem

“…Therefore, these models support the philosophy of just-in-time production, i.e., producing a commodity only when there is a need to produce it (Valente et al, 2006). Therefore, these models support the philosophy of just-in-time production, i.e., producing a commodity only when there is a need to produce it (Valente et al, 2006).…”

“…Adopting the notational conventions of Valente et al (2006), the problem can be defined as follows: A list of n independent jobs J 1 , J 2 , . Adopting the notational conventions of Valente et al (2006), the problem can be defined as follows: A list of n independent jobs J 1 , J 2 , .…”

“…Li (1997) developed a heuristic neighborhood search procedure that outperformed the filtered beam search approach of Ow and Morton (1989) both in terms of solution quality as well as running time. Valente et al (2006) presented 12 generational genetic algorithms that differ slightly from one another. Valente and Alves (2005b) proposed an improvement in the lower bounds developed in Li (1997) and Liaw (1999).…”