Each decision variable is treated as a gene and encoded by a floating-point number. Each
chromosome representing a feasible solution is encoded as a vector T n
n
x xx 21 x ] [
,
where xi denotes the value of the ith gene and n is the number of design variables in an
MOP. Because the lower bound T
n
l ll l ] [ 21
and the upper bound T
n
u uu u ] [
21
define
the feasible solution space, the domain of each xi is denoted as interval [li, ui].
The main components of the E-NSGA-II are chromosome encoding, fitness function,
selection, recombination and replacement. An initial population with P chromosomes is
randomly generated within the predefined feasible region. At each generation, E-NSGA-II
applies the fast non-dominated sorting of NSGA-II to identify non-dominated solutions and
construct the non-dominated front. And then, E-NSGA-II executes the rank comparison in
selection operation to decide successive population by elitism strategy as the diversity
preservation in NSGA-II (Deb et al., 2002). Therefore, the following sections only describe
the details of the evaluative crossover operator and the diverse replacement.