NSGA-II

NSGA-II (Non-dominated Sorting Genetic Algorithm II) is the reference multi-objective evolutionary algorithm: it ranks the population using non-dominated sorting and uses the crowding distance as a density estimator to preserve diversity.

from jmetal.algorithm.multiobjective.nsgaii import NSGAII
from jmetal.operator.crossover import SBXCrossover
from jmetal.operator.mutation import PolynomialMutation
from jmetal.problem import ZDT1
from jmetal.util.plotting import save_plt_to_file
from jmetal.util.solution import (
    get_non_dominated_solutions,
    print_function_values_to_file,
    print_variables_to_file,
    read_solutions,
)
from jmetal.util.termination_criterion import StoppingByEvaluations

"""
 Program to  configure and run the NSGA-II algorithm configured with standard settings.
"""
if __name__ == "__main__":
    problem = ZDT1()

    problem.reference_front = read_solutions(filename="resources/reference_fronts/ZDT1.pf")

    max_evaluations = 25000
    algorithm = NSGAII(
        problem=problem,
        population_size=100,
        offspring_population_size=100,
        mutation=PolynomialMutation(
            probability=1.0 / problem.number_of_variables(), distribution_index=20
        ),
        crossover=SBXCrossover(probability=1.0, distribution_index=20),
        termination_criterion=StoppingByEvaluations(max_evaluations=max_evaluations),
    )

    algorithm.run()

    front = get_non_dominated_solutions(algorithm.result())

    # Save results to file
    print_function_values_to_file(front, "FUN." + algorithm.label)
    print_variables_to_file(front, "VAR." + algorithm.label)

    # Save a PNG visualization of the front (and optional HTML if Plotly available)
    try:
        png = save_plt_to_file(front, "FUN." + algorithm.label, out_dir=".", html_plotly=True)
        print(f"Saved front plot to: {png}")
    except Exception as e:
        print(f"Warning: could not generate front plot: {e}")

    print(f"Algorithm: {algorithm.get_name()}")
    print(f"Problem: {problem.name()}")
    print(f"Computing time: {algorithm.total_computing_time}")

examples/multiobjective/nsgaii/ has many more variants: steady-state, dynamic, distributed (Dask, Spark, multiprocessing), preference-based, real-time plotting, binary and mixed encodings, constrained problems, and archive-based variants.