15.9.4 Mutation
Afier crossover, the strings
subjected to mutation, Mut
. Mutation plays the role of recoverin
geneticinformation. Ieis
ation prevents the algorithm to be trapped in a local
na the lost genetic materials as well as for randomly distributing
aninsurance policy against the irreversible loss of genetic material. Mutation hasbecr
«raitionally considered as a simple search operator, If crossover i supposed to exploit the current solution
to find better ones, mutation is supposed to help for the exploration of the whole search space. Mutation is
viewed asa background operator to maintain genetic diversity in the population. It introduces new genetic
structures in the population by randomly modifying some of its building blocks, Mutation helps escape from
local minima’ trap and maintains diversity in the population, It also keeps the gene pool well stocked, thus
ensuring ergodicity. A search space is said to be ergodic if there is a non-zero probability of generating any
solution from any population state.
Thereare many different forms of mutation for the different kinds of representation. For binary representa-
ton asimple mutation can consist in inverting the value of ach gene with asta probability. The probability
Susy aken about 1/2, where L is the length of the chromosome. It is also possible to implement kind
of ill climbing mutation operators that do mutation only if it improves the quality of the solution. Such an
operator can accelerate the search; however,
care should be taken, because it might also reduce the diversity
in the population and make the algori
hm converge toward some local optima. Mutation of a bit involves
Sipping a bit, changing 0 to 1 and vice-versa.
15.9.4,1 Flipping
'ipping ofa bic involves changing 0 to 1 and 1 to 0 based ona mutation chromosome generated. Figure 15-30
pkins mutation-flipping concept. A parent is considered and a mutation chromosome is randomly gener-
“tel. Fora 1 in mutation chromosome, the corresponding bie in parent chromosome is flipped (0 to 1 and
' 00) and child chtomosome is produced. In the cas illustrated in Figure 15-30, 1 occurs at 3 places of
Mutation chromosome, the corresponding bits in parent chromosome are flipped and the child is generated.
159.49 Interchanging :
|'otandom positions ofthe string are chosen and th bts corresponding to those positions are interchanged
"gure 15,31)Parent 1011 01014
Mutation Sogo 1004
chromosome
Chita oo11 1100
Figure 15-30 Mutation flipping.
10110101
7111100071
Parent
Child
Figure 15-31 Incerchanging.
Parent | 10110:101
Child 101101010
Figure 15-32 Reversing.
15.9.4.3. Reversing
A random position is chosen and the bits next to that position are reversed and child chromosome is produce
15.9.4.4 Mutation Probability
An important parameter in the mutation technique is the mutation probabili
of chromosome will be mutated. If there is no mutation, offspring are generated immediately after cros
(or directly copied) without any change. If mutation is performed, one or more parts of a chromosome 2
changed, If mutation probability is 100%, whole chromosome is changed; if it is 0%, nothing is
Mutation generally prevents the GA from falling into local extremes. Mutation should not occur very
because then GA will in fact change to random search.
y (Pm). Ie decides how often’
0 15.10 Stopping Condition for Genetic Algorithm Flow
In short, the various stopping condition are listed as follows:
1. Maximum generation: The GA stops when the specified number of generations has evolved.
dime. The genetic process will end when a specified time has elapsed.
L : °
f the maximum number of generation has been reached before the specified time has elapsed:
ss will end.
The genetic process will end if there is no change to the population's best fines!
a specified number of generations. :
ie : - ic
Note: If the maximum number of generation has been reached before the specified number of gene"
with no changes has been reached, the process will end,4,11, Constraints in Genetic Algorithm
1
4, Sullgenertions: The algorithm sops theres yo}
of consscutive generations of length “Sill genera
5 Stall time limit. The algorithm Stops if there
sarerval of time in seconds equal to “Se
ement in the objective function fora se
is no
alltime lime,”
The ermination oF convergence ctiteion finally brings the
methods of termination techniques, a
improvement in the
nective function d.
arch toa halt, The following are the few
15.10.1 Best Individual
A best individual convergence criterion sto
below the convergence value. This brings
solution.
Ps the search once th
the search to a faster
he minimum fitness in the populatio
1 drops
conclusion, guaranteeing at least one
good
45.10.2 Worst Individual
‘Worst individual terminates the search when the least fi
the convergence criteria. This guarantees the entire population to be of minimum standard, sthough she
best individual may not be significantly better than the worst. In this case, astringent enrve :
never be met, in which case the search will terminate after the maximum has been exceeded
15.10.3. Sum of Fitness
In this termination scheme, the search is considered to have satisfaction converged when the sum of the
fitness in the entire population is less than or equal to the convergence value in the population record. This
guarantees that virtually all individuals in the population will be within a particular fitness range, although
ic is better to pair this convergence criteria with weakest gene replacement, otherwise a few unfit individuals
in the population will blow out the fitness sum. The population size has to be considered while setting the
convergence value,
0 15.10.4 Median Fitness
individuals in the population have fitness less than
gence value may
Here at least half of the individuals will be better than or equal to the convergence value, which should give
@ good range of solutions to choose from.