weighted_sum_ga: Weighted sum genetic algorithm

In jiripetrlik/r-multiobjective-evolutionary-algorithms: Multiobjective Genetic Algorithms

Description

Use weighted sum approach to solve multiobjective optimization problem. Weighted sum approach transforms multiobjective optimization problem to single objective by multiplying objective functions values by weights. Then genetic algorithm is used to find optimal solution.

Usage

weighted_sum_ga(objective_functions_list, weights, chromosome_size,
  chromosome_type = "binary", population_size = 100,
  number_of_iterations = 100, elitism = TRUE, nc = 2,
  mutation_probability = 0.05, uniform_mutation_sd = 0.01)

Arguments

objective_functions_list	List of objective functions
weights	Objective functions weights
chromosome_size	Size of chromosome which represents candidate solutions
chromosome_type	Chromosome type ("binary" or "numeric")
population_size	Number of solutions evaluated in one iteration of genetic algorithm
number_of_iterations	Number of iterations (generations) of genetic algorithm
elitism	Use elitism
nc	NC for SBX crossover (valid if "numeric" chromosome is used)
mutation_probability	Probability of mutation (valid if "binary" chromosome is used)
uniform_mutation_sd	Standard deviation of mutation (valid if "numeric" chromosome is used)

Value

List which contains results of weighted sum genetic algorithm:

value - Sum of weighted objective funtions values for the best solution

best_solution - Chromosome which represents the best solution

best_solution_index - Index of the best solution in population

statistics - Statistics about run of genetic algorithm

parameters - Parameters of genetic algorithm

values - Values of objective functions for the best solution

weighted_values - Values of objective functions multiplied by weights for the best solution

jiripetrlik/r-multiobjective-evolutionary-algorithms documentation built on April 27, 2020, 12:12 p.m.