R Based Non-dominated Sorting Genetic Algorithm II

Share:

Description

A fast and elitist multiobjective genetic algorithm based on R.

Usage

1
2
3
nsga2R(fn, varNo, objDim, lowerBounds = rep(-Inf, varNo), upperBounds = rep(Inf, varNo),
  popSize = 100, tourSize = 2, generations = 20, cprob = 0.7, XoverDistIdx = 5,
  mprob = 0.2, MuDistIdx = 10)

Arguments

fn

Objective functions to be minimized

varNo

Number of decision variables

objDim

Number of objective functions

lowerBounds

Lower bounds of each decision variable

upperBounds

Upper bounds of each decision variable

popSize

Size of population

tourSize

Size of tournament

generations

Number of generations

cprob

Crossover probability

XoverDistIdx

Crossover distribution index, it can be any nonnegative real number

mprob

Mutation probability

MuDistIdx

Mutation distribution index, it can be any nonnegative real number

Value

The returned value is a 'nsga2R' object with the following fields in additional to above NSGA-II settings:

parameters

Solutions of decision variables found

objectives

Non-dominated objective function values

paretoFrontRank

Nondomination ranks (or levels) that each non-dominated solution belongs to

crowdingDistance

Crowding distance of each non-dominated solution

Author(s)

Ching-Shih (Vince) Tsou cstsou@mail.ntcb.edu.tw

References

Deb, K., Pratap, A., Agarwal, S., and Meyarivan, T. (2002), " A fast and elitist multiobjective genetic algorithm: NSGA-II", IEEE Transactions on Evolutionary Computation, 6(2), 182-197.

Examples

1
2
3
4
5
# find the non-dominated solutions of zdt3 test problem
results <- nsga2R(fn=zdt3, varNo=30, objDim=2, lowerBounds=rep(0,30), upperBounds=rep(1,30),
  popSize=50, tourSize=2, generations=50, cprob=0.9, XoverDistIdx=20, mprob=0.1,MuDistIdx=3)

plot(results$objectives)