scispace - formally typeset
Search or ask a question

Showing papers on "Crossover published in 1995"


Journal Article
TL;DR: A real-coded crossover operator is developed whose search power is similar to that of the single-point crossover used in binary-coded GAs, and SBX is found to be particularly useful in problems having mult ip le optimal solutions with a narrow global basin where the lower and upper bo unds of the global optimum are not known a priori.
Abstract: Abst ract . T he success of binary-coded gene t ic algorithms (GA s) in problems having discrete sear ch space largely depends on the coding used to represent the prob lem var iables and on the crossover ope ra tor that propagates buildin g blocks from parent strings to children st rings . In solving optimization problems having continuous search space, binary-coded GAs discr et ize the search space by using a coding of the problem var iables in binary strings. However , t he coding of realvalued vari ables in finit e-length st rings causes a number of difficulties: inability to achieve arbit rary pr ecision in the obtained solution , fixed mapping of problem var iab les, inh eren t Hamming cliff problem associated wit h binary coding, and processing of Holland 's schemata in cont inuous search space. Although a number of real-coded GAs are developed to solve optimization problems having a cont inuous search space, the search powers of these crossover operators are not adequate . In t his paper , t he search power of a crossover operator is defined in terms of the probability of creating an arbitrary child solut ion from a given pair of parent solutions . Motivated by the success of binarycoded GAs in discrete search space problems , we develop a real-coded crossover (which we call the simulated binar y crossover , or SBX) operator whose search power is similar to that of the single-point crossover used in binary-coded GAs . Simulation results on a nu mber of realvalued test problems of varying difficulty and dimensionality suggest t hat the real-cod ed GAs with the SBX operator ar e ab le to perfor m as good or bet ter than binary-cod ed GAs wit h the single-po int crossover. SBX is found to be particularly useful in problems having mult ip le optimal solutions with a narrow global basin an d in prob lems where the lower and upper bo unds of the global optimum are not known a priori. Further , a simulation on a two-var iable blocked function shows that the real-coded GA with SBX work s as suggested by Goldberg

2,702 citations


Book ChapterDOI
01 May 1995
TL;DR: An abstraction of the genetic algorithm, termed population-based incremental learning (PBIL), that explicitly maintains the statistics contained in a GA''s population, but which abstracts away the crossover operator and redefines the role of the population results in PBIL being simpler, both computationally and theoretically, than the GA.
Abstract: We present an abstraction of the genetic algorithm (GA), termed population-based incremental learning (PBIL), that explicitly maintains the statistics contained in a GA''s population, but which abstracts away the crossover operator and redefines the role of the population. This results in PBIL being simpler, both computationally and theoretically, than the GA. Empirical results reported elsewhere show that PBIL is faster and more effective than the GA on a large set of commonly used benchmark problems. Here we present results on a problem custom designed to benefit both from the GA''s crossover operator and from its use of a population. The results show that PBIL performs as well as, or better than, GAs carefully tuned to do well on this problem. This suggests that even on problems custom designed for GAs, much of the power of the GA may derive from the statistics maintained implicitly in its population, and not from the population itself nor from the crossover operator.

627 citations


Posted Content
TL;DR: A new model of fitness landscapes suitable for the consideration of evolutionary and other search algorithms is developed, and an investigation into crossover landscapes and hillclimbing algorithms on them illustrates the dual role played by crossover in genetic algorithms.
Abstract: A new model of fitness landscapes suitable for the consideration of evolutionary and other search algorithms is developed and its consequences are investigated. Answers to the questions "What is a landscape?" "Are landscapes useful?" and "What makes a landscape difficult to search?" are provided. The model makes it possible to construct landscapes for algorithms that employ multiple operators, including operators that act on or produce multiple individuals. It also incorporates operator transition probabilities. The consequences of adopting the model include a "one operator, one landscape" view of algorithms that search with multiple operators. An investigation into crossover landscapes and hillclimbing algorithms on them illustrates the dual role played by crossover in genetic algorithms. This leads to the "headless chicken" test for the usefulness of crossover to a given genetic algorithm and to serious questions about the usefulness of maintaining a population. A "reverse hillclimbing" algorithms is presented that allows the determination of details of the basin of attraction points on a landscape. These details can be used to directly compare members of a class of hillclimbing algorithms and to accurately predict how long a particular hillclimber will take to discover a given point. A connection between evolutionary algorithms and the heuristic search algorithms of Artificial Intelligence and Operations Research is established. One aspect of this correspondence is investigated in detail: the relationship between fitness functions and heuristic functions. By considering how closely fitness functions approximate the ideal heuristic functions, a measure of search difficult is obtained. The measure, fitness distance correlation, is a remarkably reliableble indicator of problem difficulty for a genetic algorithm on many problems taken from the genetic algorithms literature, even though the measure incorporates no knowledge of the operation of a genetic algorithm. This leads to one answer to the question "What makes a problem hard (or easy) for a genetic algorithm?" The answer is perfectly in keeping with what has been well known in Artificial Intelligence for over thirty years.

445 citations


ReportDOI
01 Jan 1995
TL;DR: This paper attempts to reconcile opposing views of uniform crossover and present a framework for understanding its virtues, as a growing body of experimental evidence suggests otherwise.
Abstract: : Traditionally, genetic algorithms have relied upon 1 and 2-point crossover operators. Many recent empirical studies, however, have shown the benefits of higher numbers of crossover points. Some of the most intriguing recent work has focused on uniform crossover, which involves on the average L/2 crossover points for strings of length L. Theoretical results suggest that, from the view of hyperplane sampling disruption, uniform crossover has few redeeming features. However, a growing body of experimental evidence suggests otherwise. In this paper, we attempt to reconcile these opposing views of uniform crossover and present a framework for understanding its virtues.

413 citations


01 Jan 1995
TL;DR: In this article, the authors proposed a greedy genetic algorithm for the quadratic assignment problem (QAP), which incorporates many greedy principles in its design and hence, is called the greedy GA.
Abstract: The Quadratic Assignment Problem (QAP) is one of the classical combinatorial optimization problems and is known for its diverse applications. Applications of QAP include assignments of tools to robots in flexible manufacturing systems, allocation of blades to hydraulic turbine runners, sequencing problems in production systems, and placement problem in VLSI design. In this paper, we suggest a genetic algorithm for the QAP and report its computational behaviour. The genetic algorithm incorporates many greedy principles in its design and hence, is called the greedy genetic algorithm. The ideas we incorporate in the greedy genetic algorithm include (i) generating the initial population using a randomized construction heuristic; (ii) new crossover schemes; (iii) a special purpose immigration scheme that promotes diversity; (iv) periodic local optimization of a subset of the population; (v) tournamenting among different populations; and (vi) an overall design that attempts to strike a balance between diversity and a bias towards fitter individuals. We test our algorithm on all the benchmark instances of QAPLIB, a well-known library of QAP instances. Out of the 132 total instances in QAPLIB of varied sizes, the greedy genetic algorithm obtained the best known solutions for 103 instances, and for the remaining instances (except one) found solutions within 1% of the best known solutions. Based on our computational testing, we believe that the greedy genetic algorithm is the best heuristic algorithm for dense QAP developed to date in terms of the quality of the solution.

327 citations


Journal ArticleDOI
TL;DR: In this article, a new representation called permutation with repetition (P-R) is presented, which is similar to the permutation scheme of the traveling salesman problem (TSP) in the sense that it cannot produce illegal operation sequences.
Abstract: In order to sequence the tasks of a job shop problem (JSP) on a number of machines related to the technological machine order of jobs, a new representation technique — mathematically known as “permutation with repetition” is presented. The main advantage of this single chromosome representation is — in analogy to the permutation scheme of the traveling salesman problem (TSP) — that it cannot produce illegal operation sequences. As a consequence of the representation scheme a new crossover operator preserving the initial scheme structure of permutations with repetition will be sketched. Its behavior is similar to the well known Order-Crossover for simple permutation schemes. Actually theGOX operator for permutations with repetition arises from aGeneralisation ofOX. Computational experiments show, that GOX passes the information from a couple of parent solutions efficiently to offspring solutions. Together, the new representation and GOX support the cooperative aspect of genetic search for scheduling problems strongly.

294 citations


Journal ArticleDOI
TL;DR: In this paper, the genetic algorithm is examined as a method for solving optimization problems in econometric estimation and compared to Nelder-Mead simplex, simulated annealing, adaptive random search, and MSCORE.
Abstract: The genetic algorithm is examined as a method for solving optimization problems in econometric estimation. It does not restrict either the form or regularity of the objective function, allows a reasonably large parameter space, and does not rely on a point-to-point search. The performance is evaluated through two sets of experiments on standard test problems as well as econometric problems from the literature. First, alternative genetic algorithms that vary over mutation and crossover rates, population sizes, and other features are contrasted. Second, the genetic algorithm is compared to Nelder–Mead simplex, simulated annealing, adaptive random search, and MSCORE.

246 citations


Journal Article
TL;DR: An adaptive mechanism for controlling the use of crossover in an EA is described and an improvement to the adaptive mechanism is presented, which can also be used to enhance performance in a non-adaptive EA.
Abstract: One of the issues in evolutionary algorithms (EAs) is the relative importance of two search operators: mutation and crossover. Genetic algorithms (GAs) and genetic programming (GP) stress the role of crossover, while evolutionary programming (EP) and evolution strategies (ESs) stress the role of mutation. The existence of many different forms of crossover further complicates the issue. Despite theoretical analysis, it appears difficult to decide a priori which form of crossover to use, or even if crossover should be used at all. One possible solution to this difficulty is to have the EA be self-adaptive, i.e., to have the EA dynamically modify which forms of crossover to use and how often to use them, as it solves a problem. This paper describes an adaptive mechanism for controlling the use of crossover in an EA and explores the behavior of this mechanism in a number of different situations. An improvement to the adaptive mechanism is then presented. Surprisingly this improvement can also be used to enhance performance in a non-adaptive EA.

242 citations


Journal Article
TL;DR: Simulation results suggest the applica t ion of realcod ed GAs with SBX operator to rea l-world optimization problems at large and observe that the real-coded GAs perform equally well or bet ter than binary coded GAs in solving a number of test problems.
Abstract: Real-coded genet ic algorit hms (GAs) do not use any coding of the problem variab les, instead they work dir ectly with the variab les . The main difference in the implementation of real-coded GAs and binary-coded GAs is in their recombination op erators. Alt ho ugh a number of real-cod ed crossover implementations were suggested, most of them were developed wit h intuition and wit hout much analysis. Recen tly, a real-cod ed crossover operator has been developed based on the search characteristics of t he single-point crossover operator used in binary-coded GAs. T his simulated binary crossover (SBX) operator has been found to work well in many test problems having continuous search space when compared to exis t ing real-coded crossover implementations. In this paper the performan ce of the real-cod ed GA with SBX in solving mult imodal and multiob jective problems is further investigated . Sharing function approach and nond ominated sort ing implementati ons are includ ed in the real-coded GA with SBX to solve mult imodal and mult iobjective problems, resp ecti vely. It is observed that the real-coded GAs perform equally well or bet ter than binarycoded GAs in solving a nu mber of test problems . One advant age of the SBX operator is that it can restri ct childr en solut ions to any arb it rary closeness to the parent solutions , t hereby not requi rin g any separate mating restrict ion scheme for bet ter performance. F inally, rea l-coded GAs with SBX have been successfully used to find mu lt iple P aretooptimal solut ions in solving a welded beam design pr oblem . These simulation results ar e encour aging and suggest the applica t ion of realcod ed GAs with SBX operator to rea l-world optimization problems at large. *Electronic mail address: debClliitk. ernet. in. 432 K alyanmoy Deb and Amarendra Kumar

217 citations


Journal ArticleDOI
TL;DR: In this article, a two-stage mapping process is constructed by the mapping relationships between unsigned decimal integers and discrete values, which can significantly reduce the computational effort and promote the computational efficiency.

215 citations


Journal ArticleDOI
TL;DR: An improved simple genetic algorithm developed for reactive power system planning and a new population selection and generation method which makes the use of Benders' cut is presented.
Abstract: This paper presents an improved simple genetic algorithm developed for reactive power system planning. Successive linear programming is used to solve operational optimization sub-problems. A new population selection and generation method which makes the use of Benders' cut is presented in this paper. It is desirable to find the optimal solution in few iterations, especially in some test cases where the optimal results are expected to be obtained easily. However, the simple genetic algorithm has failed in finding the solution except through an extensive number of iterations. Different population generation and crossover methods are also tested and discussed. The method has been tested for 6 bus and 30 bus power systems to show its effectiveness. Further improvement for the method is also discussed.

Journal ArticleDOI
TL;DR: This paper compares the performance of several crossover operators, including two new operators and a new faster formulation of a previously published operator and describes a method for designing problem specific crossover incorporating a novel tie-breaking algorithm.

Journal ArticleDOI
TL;DR: This paper describes the development and testing of a Genetic Algorithm for the generation of multiple solutions to the assembly line balancing (ALB) problem, and results are achieved by combining the genetic approach with a simple local optimization procedure.

Journal ArticleDOI
01 Feb 1995-Genetics
TL;DR: It is found that some biologically inspired point process models incorporating one or two additional parameters provide a dramatically better fit to the data than the usual "no-interference" Poisson model.
Abstract: In analyzing genetic linkage data it is common to assume that the locations of crossovers along a chromosome follow a Poisson process, whereas it has long been known that this assumption does not fit the data. In many organisms it appears that the presence of a crossover inhibits the formation of another nearby, a phenomenon known as "interference." We discuss several point process models for recombination that incorporate position interference but assume no chromatid interference. Using stochastic simulation, we are able to fit the models to a multilocus Drosophila dataset by the method of maximum likelihood. We find that some biologically inspired point process models incorporating one or two additional parameters provide a dramatically better fit to the data than the usual "no-interference" Poisson model.

Proceedings Article
15 Jul 1995
TL;DR: A simple method for testing the usefulness of crossover for a particular problem is presented, which makes it possible to identify situations in which crossover is apparently useful but is in fact producing gains that are only equal to those that can be obtained with macromutation and no population.
Abstract: A Genetic Algorithm (GA) maintains a population of individuals for the express purpose of improving performance via communication of information between contemporary individuals. This is achieved in a GA through the use of a crossover operator. If crossover is not a useful method for this exchange, the GA should not, on average, perform any better than a variety of simpler algorithms that are not population-based. A simple method for testing the usefulness of crossover for a particular problem is presented. This makes it possible to identify situations in which crossover is apparently useful but is in fact producing gains that are only equal to (or less than) those that can be obtained with macromutation and no population.

Journal ArticleDOI
TL;DR: This paper describes the application of genetic algorithms to nonlinear constrained mixed discrete-integer optimization problems with optimal sets of parameters furnished by a meta-genetic algorithm.
Abstract: This paper describes the application of genetic algorithms to nonlinear constrained mixed discrete-integer optimization problems with optimal sets of parameters furnished by a meta-genetic algorithm. Genetic algorithms are combinatorial in nature, and therefore are computationally suitable for treating discrete and integer design variables. Careful attention has been paid to modify the genetic algorithms to promote computational efficiency. Some numerical experiments were performed so as to determine the appropriate range of genetic parameter values. Then the meta-genetic algorithm was employed to optimize these parameters to locate the best solution. Three examples are given to demonstrate the effectiveness of the methodology developed in this paper. Four crossover operators have been compared and the results show that a four-point crossover operator performs best.

Book ChapterDOI
01 Jan 1995
TL;DR: The function f mdG seems to be a powerful new tool for generalizing deception and relating hillclimbers (and Hamming space) to GAs and crossover and allows us to create functions, such as the minimum distance function fmdG, with k isolated global optima and multiple local optima attractive to both crossover and hillClimbers.
Abstract: We assume that the modality (i.e., number of local optima) of a fitness landscape is related to the difficulty of finding the best point on that landscape by evolutionary computation (e.g., hillclimbers and genetic algorithms (GAs)). We first examine the limits of modality by constructing a unimodal function and a maximally multimodal function. At such extremes our intuition breaks down. A fitness landscape consisting entirely of a single hill leading to the global optimum proves to be harder for hillclimbers than GAs. A provably maximally multimodal function, in which half the points in the search space are local optima, can be easier than the unimodal, single hill problem for both hillclimbers and GAs. Exploring the more realistic intermediate range between the extremes of modality, we construct local optima with varying degrees of “attraction” to our evolutionary algorithms. Most work on optima and their basins of attraction has focused on hills and hillclimbers, while some research has explored attraction for the GA's crossover operator. We extend the latter results by defining and implementing maximal partial deception in problems with k arbitrarily placed global optima. This allows us to create functions, such as the minimum distance function f mdG , with k isolated global optima and multiple local optima attractive to both crossover and hillclimbers. The function f mdG seems to be a powerful new tool for generalizing deception and relating hillclimbers (and Hamming space) to GAs and crossover.

Journal ArticleDOI
TL;DR: A genetic algorithm takes an initial set of possible starting solutions, and iteratively improves them by means of crossover and mutation operators that are related to those involved in Darwinian evolution.

Journal ArticleDOI
TL;DR: A genetic algorithm has been designed which generates molecular structures within constraints which are modified to ‘grow’ into families of structures which, using the evolutionary operators of selection, crossover and mutation evolve to better fit the constraints.
Abstract: A genetic algorithm has been designed which generates molecular structures within constraints. The constraints may be any useful function, for example an enzyme active site, a pharmacophore or molecular properties from pattern recognition or rule-induction analyses. The starting point may be random or may utilise known molecules. These are modified to 'grow' into families of structures which, using the evolutionary operators of selection, crossover and mutation evolve to better fit the constraints. The basis of the algorithm is described together with some applications in lead generation, 3D database construction and drug design. Genetic algorithms of this type may have wider applications in chemistry, for example in the design and optimisation of new polymers, materials (e.g. superconducting materials) or synthetic enzymes.

Journal ArticleDOI
Chae Y. Lee1, J. Y. Choi1
TL;DR: An optimal timing algorithm is presented which decides the optimal starting time of each job in a given job sequence which is proved to outperform existing heuristic methods.

Posted Content
TL;DR: In this article, a simple method for testing the usefulness of crossover for a particular problem is presented, which makes it possible to identify situations in which crossover is apparently useful but is in fact producing gains that are only equal to (or less than) those that can be obtained with macromutation and no population.
Abstract: A Genetic Algorithm (GA) maintains a population of individuals for the express purpose of improving performance via communication of information between contemporary individuals. This is achieved in a GA through the use of a crossover operator. If crossover is not a useful method for this exchange, the GA should not, on average, perform any better than a variety of simpler algorithms that are not population-based. A simple method for testing the usefulness of crossover for a particular problem is presented. This makes it possible to identify situations in which crossover is apparently useful but is in fact producing gains that are only equal to (or less than) those that can be obtained with macromutation and no population.

Journal ArticleDOI
TL;DR: This paper considers two-dimensional Fermi liquids in the vicinity of a quantum transition to a phase with commensurate, antiferromagnetic long-range order and proposes a universal scaling function which determines the entire, temperature-, wave-vector-, and frequency-dependent, dynamic, staggered spin susceptibility in terms of four experimentally measurable, T=0 parameters.
Abstract: We consider two-dimensional Fermi liquids in the vicinity of a quantum transition to a phase with commensurate, antiferromagnetic long-range order. Depending upon the Fermi-surface topology, mean-field spin-density-wave theory predicts two different types of such transitions, with mean-field dynamic critical exponents z=1 (when the Fermi surface does not cross the magnetic zone boundary, type A) and z=2 (when the Fermi surface crosses the magnetic zone boundary, type B). The type-A system only displays z=1 behavior at all energies and its scaling properties are similar (though not identical) to those of an insulating Heisenberg antiferromagnet. Under suitable conditions precisely stated in this paper, the type-B system displays a crossover from relaxational behavior at low energies to type-A behavior at high energies. A scaling hypothesis is proposed to describe this crossover: we postulate a universal scaling function which determines the entire, temperature-, wave-vector-, and frequency-dependent, dynamic, staggered spin susceptibility in terms of four experimentally measurable, T=0 parameters. The scaling function contains the full scaling behavior in all regimes for both type-A and -B systems. The crossover behavior of the uniform susceptibility and the specific heat is somewhat more complicated and is also discussed. Explicit computation of the crossover functions is carried out in a large N expansion on a mean-field model. Some new results for the critical properties on the ordered side of the transition are also obtained in a spin-density-wave formalism. The possible relevance of our results to the doped cuprate compounds is briefly discussed.

Journal ArticleDOI
TL;DR: Edge-recombination crossover used in conjunction with several specialized operators is found to perform best in these experiments; these operators solved a 10KB sequence, consisting of 177 fragments, with no manual intervention.
Abstract: We study different genetic algorithm operators for one permutation problem associated with the Human Genome Project—the assembly of DNA sequence fragments from a parent clone whose sequence is unknown into a consensus sequence corresponding to the parent sequence. The sorted-order representation, which does not require specialized operators, is compared with a more traditional permutation representation, which does require specialized operators. The two representations and their associated operators are compared on problems ranging from 2K to 34K base pairs (KB). Edge-recombination crossover used in conjunction with several specialized operators is found to perform best in these experimentss these operators solved a 10KB sequence, consisting of 177 fragments, with no manual intervention. Natural building blocks in the problem are exploited at progressively higher levels through “macro-operators.” This significantly improves performance.

Journal ArticleDOI
01 Feb 1995-Genetics
TL;DR: Under the assumption of no chromatid interference, the probability for any single spore or tetrad joint recombination pattern is derived under the chi-square model and comparisons are made between this model and some other tractable models in the literature.
Abstract: The chi-square model (also known as the gamma model with integer shape parameter) for the occurrence of crossovers along a chromosome was first proposed in the 1940's as a description of interference that was mathematically tractable but without biological basis. Recently, the chi-square model has been reintroduced into the literature from a biological perspective. It arises as a result of certain hypothesized constraints on the resolution of randomly distributed crossover intermediates. In this paper under the assumption of no chromatid interference, the probability for any single spore or tetrad joint recombination pattern is derived under the chi-square model. The method of maximum likelihood is then used to estimate the chi-square parameter m and genetic distances among marker loci. We discuss how to interpret the goodness-of-fit statistics appropriately when there are some recombination classes that have only a small number of observations. Finally, comparisons are made between the chi-square model and some other tractable models in the literature.

01 Jan 1995
TL;DR: This paper investigates the phenomenon of multi-parent reproduction, i.e. recombination mechanisms where an arbitrary n>1 number of parents participate in creating children, and discusses scanning crossover that generalizes the standard uniform crossover and diagonal crossover thatgeneralizes 1-point crossover.
Abstract: In this paper we investigate the phenomenon of multi-parent reproduction, i.e. we study recombination mechanisms where an arbitrary $n>1$ number of parents participate in creating children. In particular, we discuss scanning crossover that generalizes the standard uniform crossover and diagonal crossover that generalizes 1-point crossover, and study the effects of different number of parents on the GA behavior. We conduct experiments on tough function optimization problems and observe that by multi-parent operators the performance of GAs can be enhanced significantly. We also give a theoretical foundation by showing how these operators work on distributions.

Journal ArticleDOI
TL;DR: In this article, a relatively straightforward method for efficiently reducing the ERS-1 orbit error using Topex/Postidon data is presented. The method is based on a global minimization of TOPEX/Poscidon-ERS-1 (TP-E) dual crossover differences.
Abstract: This paper presents a relatively straightforward method for efficiently reducing the ERS-1 orbit error using Topex/Postidon data. The method is based on a global minimization of Topex/Poscidon-ERS-1 (TP-E) dual crossover differences. The TP-E crossover differences give an estimate of the ERS-1 radial orbit error almost directly, leading to a geometric estimation of orbit error. Smoothing cubic-spline functions are then used to obtain a continuous estimation of the orbit error over time. The splines can also be adjusted to minimize the ERS-1-ERS-1 (E-E) crossover differences. This allows a better estimation of the orbit error, especially poleward of 66° where no TP-E crossovers are available. The method was successfully applied to the final TP and ERS-1 datasets [i.e., the TP GDRs (geophysical data records) and the ERS-1 OPRs (ocean products)]. The authors used one full 35-day ERS-1 cycle and five TP cycles concurrent with ERS-1 data. Only crossovers with time differences lm than 5 days are used i...

Journal ArticleDOI
TL;DR: In this article, a feasibility study is described in which a simple genetic algorithm has been developed in order to examine the suitability of such an approach, and the results of the present experiments are presented, and further complexities are discussed in the context of the genetic approach.

Book ChapterDOI
04 Jun 1995
TL;DR: In this paper, the authors investigate the phenomenon of multi-parent reproduction, i.e., recombination mechanisms where an arbitrary n>1 number of parents participate in creating children.
Abstract: In this paper we investigate the phenomenon of multi-parent reproduction, i.e. we study recombination mechanisms where an arbitrary n>1 number of parents participate in creating children. In particular, we discuss scanning crossover that generalizes the standard uniform crossover and diagonal crossover that generalizes 1-point crossover, and study the effects of different number of parents on the GA behavior. We conduct experiments on tough function optimization problems and observe that by multi-parent operators the performance of GAs can be enhanced significantly. We also give a theoretical foundation by showing how these operators work on distributions.

Proceedings ArticleDOI
12 Sep 1995
TL;DR: A new crossover called multi-step crossover (MSX) which utilizes a neighborhood structure and a distance in the problem space and successively generates their descendents along the path connecting both of them.
Abstract: Genetic algorithms (GAs) have been designed as general purpose optimization methods. GAs can be uniquely characterized by their population-based search strategies and their operators: mutation, selection and crossover. In this paper, we propose a new crossover called multi-step crossover (MSX) which utilizes a neighborhood structure and a distance in the problem space. Given parents, MSX successively generates their descendents along the path connecting both of them. MSX was applied to the job-shop scheduling problem (JSSP) as a high-level crossover to work on the critical path. Preliminary experiments using JSSP benchmarks showed the promising performance of a GA with the proposed MSX.

Journal ArticleDOI
Chae Y. Lee1, Seok Kim1
TL;DR: This paper develops parallel genetic algorithms for a job scheduling problem on a single machine to minimize the total generally weighted earliness and tardiness penalties from a common due date.