<s>
In	O
computer	B-General_Concept
programming	I-General_Concept
,	O
genetic	B-Algorithm
representation	I-Algorithm
is	O
a	O
way	O
of	O
presenting	O
solutions/individuals	O
in	O
evolutionary	O
computation	O
methods	O
.	O
</s>
<s>
The	O
term	O
encompasses	O
both	O
the	O
concrete	O
data	B-General_Concept
structures	I-General_Concept
and	O
data	O
types	O
used	O
to	O
realize	O
the	O
genetic	O
material	O
of	O
the	O
candidate	O
solutions	O
in	O
the	O
form	O
of	O
a	O
genome	O
,	O
and	O
the	O
relationships	O
between	O
search	O
space	O
and	O
problem	O
space	O
.	O
</s>
<s>
Genetic	B-Algorithm
representation	I-Algorithm
can	O
encode	O
appearance	O
,	O
behavior	O
,	O
physical	O
qualities	O
of	O
individuals	O
.	O
</s>
<s>
Difference	O
in	O
genetic	B-Algorithm
representations	I-Algorithm
is	O
one	O
of	O
the	O
major	O
criteria	O
drawing	O
a	O
line	O
between	O
known	O
classes	O
of	O
evolutionary	O
computation	O
.	O
</s>
<s>
Genetic	B-Algorithm
algorithms	I-Algorithm
(	O
GAs	O
)	O
typically	O
linear	O
representations	O
;	O
these	O
are	O
often	O
,	O
but	O
not	O
always	O
,	O
binary	O
.	O
</s>
<s>
The	O
main	O
property	O
that	O
makes	O
these	O
genetic	B-Algorithm
representations	I-Algorithm
convenient	O
is	O
that	O
their	O
parts	O
are	O
easily	O
aligned	O
due	O
to	O
their	O
fixed	O
size	O
.	O
</s>
<s>
Variable	O
length	O
representations	O
were	O
also	O
explored	O
in	O
genetic	B-Algorithm
algorithms	I-Algorithm
,	O
but	O
crossover	O
implementation	O
is	O
more	O
complex	O
in	O
this	O
case	O
.	O
</s>
<s>
Evolution	B-Algorithm
strategy	I-Algorithm
uses	O
linear	O
real-valued	O
representations	O
,	O
e.g.	O
</s>
<s>
It	O
uses	O
mostly	O
gaussian	B-Application
mutation	O
and	O
blending/averaging	O
crossover	O
.	O
</s>
<s>
Genetic	B-Algorithm
programming	I-Algorithm
(	O
GP	O
)	O
pioneered	O
tree-like	O
representations	O
and	O
developed	O
genetic	O
operators	O
suitable	O
for	O
such	O
representations	O
.	O
</s>
<s>
Human-based	O
genetic	B-Algorithm
algorithm	I-Algorithm
(	O
HBGA	O
)	O
offers	O
a	O
way	O
to	O
avoid	O
solving	O
hard	O
representation	O
problems	O
by	O
outsourcing	O
all	O
genetic	O
operators	O
to	O
outside	O
agents	O
,	O
in	O
this	O
case	O
,	O
humans	O
.	O
</s>
<s>
The	O
algorithm	O
has	O
no	O
need	O
for	O
knowledge	O
of	O
a	O
particular	O
fixed	O
genetic	B-Algorithm
representation	I-Algorithm
as	O
long	O
as	O
there	O
are	O
enough	O
external	O
agents	O
capable	O
of	O
handling	O
those	O
representations	O
,	O
allowing	O
for	O
free-form	O
and	O
evolving	O
genetic	B-Algorithm
representations	I-Algorithm
.	O
</s>
<s>
Analogous	O
to	O
biology	O
,	O
evolutionary	B-Algorithm
algorithms	I-Algorithm
(	O
EA	O
)	O
distinguish	O
between	O
problem	O
space	O
(	O
corresponds	O
to	O
phenotype	O
)	O
and	O
search	O
space	O
(	O
corresponds	O
to	O
genotype	O
)	O
.	O
</s>
<s>
When	O
more	O
possible	O
genotypes	O
exist	O
than	O
phenotypes	O
,	O
the	O
genetic	B-Algorithm
representation	I-Algorithm
of	O
the	O
EA	O
is	O
called	O
redundant	O
.	O
</s>
<s>
This	O
has	O
motivated	O
researchers	O
in	O
the	O
EA	O
community	O
to	O
examine	O
whether	O
neutral	O
mutations	O
can	O
improve	O
EA	O
functioning	O
by	O
giving	O
populations	O
that	O
have	O
converged	O
to	O
a	O
local	O
optimum	O
a	O
way	O
to	O
escape	O
that	O
local	O
optimum	O
through	O
genetic	B-Application
drift	I-Application
.	O
</s>
<s>
The	O
locality	O
of	O
a	O
genetic	B-Algorithm
representation	I-Algorithm
corresponds	O
to	O
the	O
degree	O
to	O
which	O
distances	O
in	O
the	O
search	O
space	O
are	O
preserved	O
in	O
the	O
problem	O
space	O
after	O
genotype-phenotype	O
mapping	O
.	O
</s>
<s>
In	O
order	O
for	O
successful	O
schemata	B-Algorithm
not	O
to	O
be	O
destroyed	O
by	O
genotype-phenotype	O
mapping	O
after	O
a	O
minor	O
mutation	O
,	O
the	O
locality	O
of	O
a	O
representation	O
must	O
be	O
high	O
.	O
</s>
<s>
An	O
obvious	O
and	O
commonly	O
used	O
encoding	O
for	O
the	O
traveling	B-Algorithm
salesman	I-Algorithm
problem	I-Algorithm
and	O
related	O
tasks	O
is	O
to	O
number	O
the	O
cities	O
to	O
be	O
visited	O
consecutively	O
and	O
store	O
them	O
as	O
integers	O
in	O
the	O
chromosome	O
.	O
</s>
<s>
In	O
order	O
to	O
be	O
able	O
to	O
create	O
this	O
,	O
as	O
many	O
allocation	O
matrices	B-Architecture
must	O
be	O
created	O
as	O
resources	O
can	O
be	O
allocated	O
to	O
one	O
subtask	O
at	O
most	O
.	O
</s>
<s>
The	O
creation	O
of	O
allocation	O
matrices	B-Architecture
ensures	O
firstly	O
that	O
there	O
are	O
no	O
inadmissible	O
multiple	O
allocations	O
.	O
</s>
<s>
To	O
achieve	O
this	O
in	O
a	O
timely	O
manner	O
,	O
which	O
is	O
a	O
common	O
optimization	O
goal	O
and	O
not	O
a	O
constraint	O
,	O
a	O
simple	O
heuristic	B-Algorithm
can	O
be	O
used	O
:	O
Allocate	O
the	O
required	O
resource	O
for	O
the	O
desired	O
time	O
period	O
as	O
early	O
as	O
possible	O
,	O
avoiding	O
duplicate	O
reservations	O
.	O
</s>
<s>
If	O
the	O
scheduling	O
problem	O
is	O
modified	O
to	O
the	O
scheduling	O
of	O
workflows	B-Operating_System
instead	O
of	O
independent	O
subtasks	O
,	O
at	O
least	O
some	O
of	O
the	O
work	O
steps	O
of	O
a	O
workflow	B-Operating_System
have	O
to	O
be	O
executed	O
in	O
a	O
given	O
order	O
.	O
</s>
<s>
If	O
the	O
previously	O
described	O
scheduling	O
heuristic	B-Algorithm
now	O
determines	O
that	O
the	O
predecessor	O
of	O
a	O
work	O
step	O
is	O
not	O
completed	O
when	O
it	O
should	O
be	O
started	O
itself	O
,	O
the	O
following	O
repair	O
mechanism	O
can	O
help	O
:	O
Postpone	O
the	O
scheduling	O
of	O
this	O
work	O
step	O
until	O
all	O
its	O
predecessors	O
are	O
finished	O
.	O
</s>
<s>
The	O
following	O
layout	O
planning	O
task	O
is	O
intended	O
to	O
illustrate	O
a	O
different	O
use	O
of	O
a	O
heuristic	B-Algorithm
in	O
genotype-phenotype	O
mapping	O
:	O
On	O
a	O
rectangular	O
surface	O
different	O
geometric	O
types	O
of	O
objects	O
are	O
to	O
be	O
arranged	O
in	O
such	O
a	O
way	O
that	O
as	O
little	O
area	O
as	O
possible	O
remains	O
unused	O
.	O
</s>
