An Enhanced Grouping Genetic Algorithm for solving the cell formation problem

Abstract

International audienceCell formation is often the first step in solving facility layout design problems. The objective is to group part families and machines so that they can be assigned to manufacturing cells. The cell formation problem is a non-deterministic polynomial (NP) complete problem which means that the time taken to produce solutions increases exponentially with problem size. This paper presents the Enhanced Grouping Genetic Algorithm (EnGGA) that has been developed for solving the cell formation problem. The EnGGA replaces the replacement heuristic in a standard Grouping Genetic Algorithm with a Greedy Heuristic and employs a rank-based roulette-elitist strategy, which is a new mechanism for creating successive generations. The EnGGA was tested using well-known data sets from the literature. The quality of the solutions was compared with those produced by other methods using the grouping efficacy measure. The results show that the EnGGA is effective and outperforms or matches the other methods

An iterated local search heuristic for cell formation

“…As has been noted in previous comparative analyses (James et al, 2007;Tunnukij & Hicks, 2009), there is the propensity for inconsistency between the original published source for test problems and data that have been used and/or published in other sources. To the greatest extent possible, we relied on the original sources of data in our analyses, and commentary is provided where our results differ from those reported by other authors.…”

“…The part-machine clustering problems used in our computational analyses were collected for a study by Goncalves and Resende (2004), and been have used (either all or in part) for testing purposes by several other authors (James et al, 2007;Li et al, 2010;Noktehdan et al, 2010;Paydar & Saidi-Mehrabad, 2013;Tunnukij & Hicks, 2009). Although there were originally 35 test problems in this suite, we consider two versions of two of the problems, which yields a total of 37.…”

“…These methods include, but are not necessarily limited to, genetic algorithms (Tariq et al, 2009;Tunnukij & Hicks, 2009), simulated annealing (Wu et al, 2008), ant colony methods (Li et al, 2010), water flow-like algorithms (Wu et al, 2010), and evolutionary algorithms with embedded variable neighborhood search (Paydar & SaidiMehrabad, 2013). Reviews of implementations of different metaheuristics are provided by Papaioannou and Wilson (2010) and Ghosh, Sengupta, Chattopadhyay, and Dan (2011).…”

“…Throughout the remainder of this paper, we focus on a particular objective criterion known as the grouping efficacy index (GEI: Kumar & Chandrasekharan, 1990). The GEI is arguably one of the most popular indices used in the cell formation literature (Brusco, 2015;Brusco & Steinley, 2007;Bychkov, Batsyn, & Pardalos, 2014;James, Brown, & Keeling, 2007;Li, Baki, & Aneja, 2010;Noktehdan, Karimi, & Kashan, 2010;Paydar & Saidi-Mehrabad, 2013;Tariq, Hussain, & Ghafoor, 2009;Tunnukij & Hicks, 2009;Wu, Chang, & Chung, 2008, Wu, Chung, & Chang, 2010.…”

A Comparison of a Standard Genetic Algorithm with a Hybrid Genetic Algorithm Applied to Cell Formation Problem

“…Chee and Ponnambalam [17] GA-particle swarm optimization Defersha and Chen [18] GA-linear programming Tunnukij and Hicks [19] GGA-greedy heuristic Tariq et al [20] G A -l o c a l s e a r c h h e u r i s t i c…”