Comparing the Performance of Different Meta-Heuristics for Unweighted Parallel Machine Scheduling

Abstract

This article considers the due window scheduling problem to minimise the number of early and tardy jobs on identical parallel machines. This problem is known to be NP complete and thus finding an optimal solution is unlikely. Three meta-heuristics and their hybrids are proposed and extensive computational experiments are conducted. The purpose of this paper is to compare the performance of these meta-heuristics and their hybrids and to determine the best among them. Detailed comparative tests have also been conducted to analyse the different heuristics with the simulated annealing hybrid giving the best result. OPSOMMINGDie minimering van vroeë en trae take op identiese parallelle masjien met behulp van die gepaste gleufskeduleringsprobleem word oorweeg. Die probleem is nie-deterministies polinomiese tyd volledig en die vind van 'n optimale oplossing is dus onwaarskynlik. Drie meta-heuristieke en hul hibriede word voorgestel en uitgebreide berekeningseksperimente word uitgevoer. Die doel van hierdie artikel is om die vertoning van hierdie metaheuristieke en hul hibriede met mekaar te vergelyk om sodoende die beste te identifiseer. Gedetaileerde vergelykende toetse is ook uitgevoer om die verskillende heuristieke te ondersoek; daar is gevind dat die gesimuleerde uitgloei hibried die beste resultaat lewer.

Advanced discrete firefly algorithm with adaptive mutation‐based neighborhood search for scheduling unrelated parallel machines with sequence‐dependent setup times

“…19 It is equally important to mention here that because in the proposed scheduling formulation, each firefly was used to represent a candidate schedule of parallel machines, the maximum makespan C max which is obtained by the firefly is used to measure its performance. The fitness value of the individual firefly is computed using the expression given in Equation (20).…”

“…Similarly, Ezugwu 5 introduced an improved symbiotic organism search (SOS) algorithm to solve the UPMSP, with schemes to handle candidates' and individual organisms' one‐to‐one mappings. Adamu and Adewumi 20 compared six metaheuristics' performances to find the best algorithm to solve the UPMSP. Arnaout 21 implemented a worm optimization algorithm to solve the UPMSP with a sequence‐dependent and machine‐dependent setup.…”

A performance comparison and evaluation of metaheuristics for a batch scheduling problem in a multi-hybrid cell manufacturing system with skilled workforce assignment

“…The process is fundamentally based on evolution in nature, and uses selection, crossover, and mutation operators. Because the effectiveness of GA for solving discrete optimization problems has been demonstrated in many studies [2]- [1], it is included in this study for comparison purposes.…”

“…The SA algorithm has been implemented for optimization problems over the past few decades, and has demonstrated efficiency for discrete optimization problems [ [1], [51], and [12]]. Therefore, the SA is included in this study for comparison purposes.…”

“…The number of batches on each cell is denoted by NB, and listed in Table 5. For the comparison analysis, the batch sizes were randomly generated from a discrete uniform distribution in the interval U [1,20]. The total number of employees in the system was equal to nine.…”