Review of Various Crossover Operators in Genetic Algorithms

Priyanka Yadlapalli Ambika Annavarapu Padmavati Kora Nagaja Katragadda

Department of ECE, GRIET, Hyderabad
Abstract Genetic Algorithms (GA) are the population based search and optimization technique that mimic the process of natural
evolution. Genetic algorithms are very effective way of finding a very effective way of quickly finding a reasonable solution to a complex
problem. Performance of genetic algorithms mainly depends on type of genetic operators which involve crossover and mutation operators.
Different crossover and mutation operators exist to solve the problem that involves large population size. Example of such a problem is
Travelling Sales Man problem, which is having a large set of solution. In this paper we will discuss various crossover operators that help
in solving the problem.

Keywords Genetic Algorithm; Mutation; crossover; Selection; travelling salesman problem.

I. INTRODUCTION accustomed to the fact that it is the population size that

Evolutionary a l g o r i t h m s always follow the Darwin has an impact on the performance of genetic algorithm
concept of Survival of the fittest which is extensively which results in the problem of genetic drift which is
used for optimization problems for nearly four decades rampant mostly in the case of multimodal search space.
[1]. Evolutionary algorithms are classic examples of By initiating a local search method within the genetic
heuristic search algorithms yet they do not yield exact operators we can produce new genes that can resolve the
optimal solutions, but will certainly help to find better problem of genetic drift and accelerate the search
optimal solutions when compared to their parallel towards global optima [6].When a genetic algorithm with
counterparts within less amount of time. Genetic a local search method is combined a hybrid genetic
algorithms and E v o l u t i on a r y programming algorithm/mimetic algorithm is evolved. Some crossover
combinedly provide an excellent system for effectively operators are utilized for exploitation as well as for
sampling large search areas with basic technique is widely exploration. The main emphasis of this paper is to
applicable. Genetic algorithms are by nature have adaptive study various types of crossover operators
optimization algorithms which resemble the process of
natural selection and genetics [2].Both exploitation and II. TRAVELLING SALESMAN PROBLEM:
exploration techniques are by and large accountable for
the performance of genetic algorithms. Exploitation in fact The Travelling Salesman Problem (TSP) is one of the
is defined as usage of already existing information to find most widely used and well studied for combinational
out the better solution and Exploration means to find out a optimization problems. The statement of Travelling
novel and unknown solution in investigation space. The Salesman Problem (TSP) was made by Karl Menger in
crux of the genetic algorithms emerges from the capability 1930 whose statement is much uncomplicated, but it is
to combine both exploration and exploitation in an ideal one of the most challenging and popular problems in the
way [3]. Biological genetics model is regarded as one field of Operational Research. Large number of articles
of the best role model of genetic algorithm. A generic and research papers have been written and published on
genetic algorithm comprises of the following operations this problem. The most common practical interpretation
namely: Initialization, Selection, Reproduction and of TSP is that a salesman seeking the shortest tour
Replacement. Initialization represents creation of initial through n clients or cities. The Travelling Salesman
population by utilizing suitable encoding scheme. Problem is one of the famous NP-hard problems, which
Selection operator chooses the individuals arbitrarily or means that there is no perfect algorithm to solve it in
chosen according to their fitness. Crossover and mutation polynomial time. The minimal time required to obtain
genetic operators are exclusively involved to strike optimal solution is exponential [7]. The main motive is
balance between exploitation and exploration. During to find the shortest path or the path with least cost. It is
replacement, the old ones are succeeded by new very complicated as starting and finishing points are
offsprings [4]. By definition, crossover selects genes from same and we have to return to the original point after
parent chromosomes and creates a new offspring. The visiting each of the vertexes only once. The most
simplest way is to choose randomly some crossover point, prevalent practical applications of TSP is that it not only
point copy from a first parent and then everything after a involves regular distribution of goods or resources, but
crossover point copy from the second parent. In genetic finding out the shortest path for costumer servicing route,
algorithms, chromosomes denoted as linear strings of planning bus routes , vehicle routing, computer wiring,
symbols [5]. machine sequencing and scheduling, frequency
It is very important that in order to enhance the assignment in communication networks , but also in the
functionality of genetic algorithm problem there is a areas that have nothing to do with travel routes [7].
specific requisition of crossover. After reproduction or
after the so called selection process, the population III. VARIOUS METHODS USED FOR RESOLVING TSP
comprises of better individuals. Reproduction or
selection produces same individual but the progeny are In 1997, Rong Yang initiated several knowledge-
improved when compared to the older ones. The main augmented genetic operators that guide the genetic
feature of genetic algorithms is to merge both exploration algorithm more directly towards better quality of the
and exploitation in an ideal way. In reality, we are population but not trapped in local optima prematurely.
 The algorithm uses an avid crossover operator along with break up and if both head and tail have good genetic
two advanced mutation operations based on the 2-opt and material then none of the offspring will get the both
3-opt heuristics [8].In 2001, Chiung Moon introduced the good features directly.
concept of topological sort (TS), which is as an or
interpreted as ordering of vertices in a directed graph. 2) N Point Crossover
Moreover, a new crossover operation is introduced for
the proposed GA [9].In 2004, new knowledge based It was first introduced by De Jong in 1975 [18]. It has
multiple inversion operators and neighborhood swapping many cross over sites but rule is similar as we used in
operator is proposed by Shubhra Sankar Ray [10].In single point crossover. In 2-point crossover significance
2005, Lawrence V. Snyder gave a heuristic to resolve the of crossover sites is 2. Adding more and more crossover
generalized travelling salesman problem. The approach sites effects the disruptions of building blocks that
incorporates a local tour development heuristic into a sometimes reduce the performance of GA. But it allows
random-key genetic algorithm. The algorithm was very the tail and head part of a chromosome to be accepted
much successful when tested on a set of 41 standard together in the offspring
problems with familiar optimal objective values [11].
In 2005, Milena Karova found that genetic 3) Uniform Crossover
algorithm when implemented gives a maximal It does not fragment the chromosomes for recombination.
approximation of the problem, thus reforming the Each gene in progeny is created by replicating it
acquired solution with genetic operators [12].In the year from the parent selected according to the
2006, Plamenka Borovska analysed the efficiency of corresponding bit in the binary crossover mask of same
parallel computation of travelling salesman problem length as the length of parent chromosomes [17]. If the
applying the genetic approach on a casual multicomputer bit in crossover mask is 1, then the resultant gene is
cluster [13].In 2007, a two-level genetic algorithm derivative from the first parent and if the bit in crossover
(TLGA) was developed for a case, with special treatment mask is 0, then the resultant gene is derivative from the
consisting of neither intra-cluster paths nor inter- cluster second parent. A new crossover mask is produced
paths, thus realizing an integrated evolutionary arbitrarily for each pair of parent chromosomes. The
optimization for both levels of the CTSP [14].In 2007, A quantity of crossover point is not fixed initially. So, the
new technique called Particle Swarm Optimization progeny have a mix of genes from both the parents
(PSO)- algorithm has been initiated for travelling
salesman problem (TSP) is presented, and is compared 4) Three Parent Crossover
with the prevailing algorithms for resolving TSP based In this three parents are chosen randomly. Each gene of
on swarm intelligence [15].In 2008, a software system the 1st parent is compared with the equivalent gene of the
is developed to determine the ideal route for 2nd parent. If both genes are alike, the gene is occupied
Travelling Salesman Problem by employing Genetic for offspring or else the equivalent gene from the third
Algorithm technique [16].In 2009, S.N. Sivanandam parent is taken for the progeny. It is mostly used in
presented two approaches i.e. Genetic Algorithms and case of binary encoded chromosomes.
Particle swarm optimization in order to find a solution to
the given objective function by applying different 5) Arithmetic Crossover
procedures and computational techniques; as a result of It is used in case of real-value encoding. Arithmetic
which their performance can be evaluated and compared crossover operator linearly combines two parent
[17]. chromosomes [17]. Two chromosomes are particular
randomly for crossover and create two offsprings
IV. GA M ETHODOLOGY which are linear mixture of their parents
A genetic algorithm defines how an algorithm can
generate a modified copy of itself in order to achieve an 6) Partially mapped Crossover
objective. The progeny would match against the criteria to (PMC)
ensure that the objective would be eventually attained. Partially Matched or Mapped Crossover (PMX) is the
Genetic algorithm which is inspired by Darwin's theory is most normally used crossover operator. It was proposed
all about evolution. It is not that difficult to program or by Goldberg and Lingle [19] for Travelling
realize, since they are biological based. Salesman Problem in PMC, two chromosomes are linked
and two crossover sites are chosen arbitrarily. The
V. TYPES OF CROSSOVER OPERATORS fraction of chromosomes between the two crossover
points gives a related selection that undergoes the
1) Single Point Crossover crossover process through position-by- position exchange
Single point crossover is the most approved crossover operations(PEO) [2, 17]. PMX be likely to respect the
which is widely in use. A crossover site is randomly absolute positions
selected along the length of the mated strings and bits
wh i c h a r e ve r y next to the cross-sites are 7) Crossover ORDER (OX)
interchanged. When suitable site is chosen then better It was proposed by Davis and used for chromosomes
progeny can be obtained by combining the good qualities with permutation encoding [20]. The process starts in a
of the parents. If appropriate site is chosen, better way similar to that of PMX by choosing two crossover
offspring can be obtained by combining good quality points. But in its place of using point-by-point
parents else it harshly hampers string quality. In single exchanges as in case of PMX, order crossover apply
point crossover the head and tail of one chromosome sliding motion to fill the left out holes by sending the
 mapped positions. It copies the portion of permutation undertake most of the real life applications of travelling
elements among the crossover points from the cut string salesman problem.
directly to the offspring, insertion them in the similar
absolute position [2, 17]. OX tends to respect the relative VIII. FUTURE SCOPE
positions.
It appears that major dilemma with the genetic
8) Cycle Crossover (CX) algorithm for TSP is that to preserve the formation from
This is used for chromosomes with permutation encoding. the parents chromosomes and still finish up with a
Throughout recombination in cyclic crossover there is a permissible tour in the child chromosomes. Possibly a
limitation that each gene comes from the one parent or the improved crossover or mutation routine that remain
other [21]. The fundamental model at the back cycle structure from parent chromosome would give a far
crossover(BCC) is that each allele comes from one better clarification than we have previously establish for
parent together with its position. To construct a cycle of some TSP. As a potential work we want to prolong the
alleles from parent1 we have to begin with the first same process with different other crossover and mutation
allele of parent1. After that come across at the allele at operator to give an improved solution to a TSP.
the equal position in parent2 and go to the position with
the same allele in Parent1.insert this allele to the cycle and REFERENCES
do again above step until you reach your destination at the
[1] J. Holland, Adaptation in natural and artificial systems, University
first allele of parent1. Put the alleles of the cycle in the of Michigan Press, Ann Arbor, 1975.
first child on the positions they have in the first parent and [2] D.E. Goldberg, Genetic algorithms in search, optimisation,
the left behind alleles of first child come from the second and machine learning, Addison Wesley Longman, Inc., ISBN
parent along with their position. Produce next cycle from 0-201-15767-5, 1989.
[3] J. Holland, Adaptation in natural and artificial systems, University
parent2. of Michigam Press, Ann Arbor, 1975.
[4] Manju Sharma , Novel Knowledge based Selective Tabu
VI. SOLVING TSP USING GENETIC ALGORITHM Initialization in Genetic algorithm, IJARCSSE, Volume 3, Issue
5, May 2013
A genetic algorithm is popular to find a solution of [5] C. Ryan, Automatic re-engineering of software using genetic
programming, Genetic Programming Series. Kluwer Academic
a problem in much less time. Even though it might not Publishers, ISBN 0-7923-8653-1, 2000.
find the most excellent solution of the problem but it can [6] W. E. Hart, Adaptive global optimization with local search,
help in finding a near ideal solution for a 100 city tour Doctoral diss., San Diego, University of California, 1994.
in less than a minute. The following are some [7]Ivan Brezina Jr.,ZuzanaCickova, Solving the Travelling Salesman
Problem using the Ant colony Optimization,
fundamental steps of our proposed work
Management Information Systems, 2011, Vol. (6), No. (4).
Encoding: Permutation encoding is most popular way to [8] Rong Yang, Solving Large Travelling Salesman Problems
solve the TSP. We characterize cities with an integer value with Small Populations. IEEE 1997.
prior to the population initializion. [9] Chiung Moon, Jongsoo Kim, GyunghyunChoi ,YoonhoSeo,
An efficient genetic algorithm for the traveling salesman
Distance matrix: distance matrix is problem with precedence constraints, European Journal of
a representation of matrix of point to Operational Research 140 (2002) 606617, accepted 28 February
point distances. 2001
[10] ShubhraSankar Ray, Sanghamitra Bandyopadhyay and Sankar
Selection based on fitness function: the
K.Pal, New Operators of Genetic Algorithms for Traveling
fitness function will be total expenditure of the Salesman Problem, 2000IEEE.
visit represented by each chromosome. The
smaller the sum, the fitter the solution
represented by that chromosome
Generating arbitrary numbers equal to
population size. Best of two routes will be
chosen using tournament selection to relate
Mutation.
Next generation of population size will
be produced. Process will go through
predefined iterations.
After the final iteration the smallest number
of distance size will be displayed as result.

VII. CONCLUSION

In this paper, we have discussed eight types of

mutations strategy in the genetic algorithm procedure
to create new and better generations. As we learnt that
genetic algorithms appear to be a good solution for TSP,
however it very much depends on the technique the
difficulty is encoded and which crossover and mutation
methods are used. So our labors will decide which
mutation operator will always gives end result to