Performance of a genetic algorithm for solving the multi-objective, multimodel transportation network design problem

Ties Brands, Eric C. van Berkum

Research output: Contribution to journalArticleAcademicpeer-review

68 Downloads (Pure)

Abstract

The optimization of infrastructure planning in a multimodal network is defined as a multi-objective network design problem, with accessibility, use of urban space by parking, operating deficit and climate impact as objectives. Decision variables are the location of park and ride facilities, train stations and the frequency of public transport lines. For a case study the Pareto set is estimated by the Non-dominated Sorting Genetic Algorithm (NSGA-II). Such a Pareto set is one specific outcome of the optimization process, for a specific value of the parameters generation size, number of generations and mutation rate and for a specific outcome of the Monte Carlo simulation within NSGA-II. Similar issues exist for many other metaheuristics. However, when applied in practice, a policy maker desires a result that is robust for these unknown aspects of the method. In this paper Pareto sets from various runs of the NSGA-II algorithm are analyzed and compared. To compare the values of the decision variables in the Pareto sets, new methods are necessary, so these are defined and applied. The results show that the differences concerning decision variables are considerably larger than the differences concerning objectives. This indicates that the randomness of the algorithm may be a problem when determining the decisions to be made. Furthermore, it is concluded that variations caused by different parameters are comparable with the variations caused by randomness within the algorithm.
Original languageEnglish
Pages (from-to)1-20
JournalInternational journal of transportation
Volume2
Issue number1
DOIs
Publication statusPublished - 2014

Fingerprint

network design
genetic algorithm
park and ride
infrastructure planning
parking
public transport
climate effect
sorting
accessibility
train
mutation
decision
simulation
method
parameter

Keywords

  • IR-101342
  • METIS-294089

Cite this

@article{5ea4b713f2214136a6ed32285ae9d1ab,
title = "Performance of a genetic algorithm for solving the multi-objective, multimodel transportation network design problem",
abstract = "The optimization of infrastructure planning in a multimodal network is defined as a multi-objective network design problem, with accessibility, use of urban space by parking, operating deficit and climate impact as objectives. Decision variables are the location of park and ride facilities, train stations and the frequency of public transport lines. For a case study the Pareto set is estimated by the Non-dominated Sorting Genetic Algorithm (NSGA-II). Such a Pareto set is one specific outcome of the optimization process, for a specific value of the parameters generation size, number of generations and mutation rate and for a specific outcome of the Monte Carlo simulation within NSGA-II. Similar issues exist for many other metaheuristics. However, when applied in practice, a policy maker desires a result that is robust for these unknown aspects of the method. In this paper Pareto sets from various runs of the NSGA-II algorithm are analyzed and compared. To compare the values of the decision variables in the Pareto sets, new methods are necessary, so these are defined and applied. The results show that the differences concerning decision variables are considerably larger than the differences concerning objectives. This indicates that the randomness of the algorithm may be a problem when determining the decisions to be made. Furthermore, it is concluded that variations caused by different parameters are comparable with the variations caused by randomness within the algorithm.",
keywords = "IR-101342, METIS-294089",
author = "Ties Brands and {van Berkum}, {Eric C.}",
year = "2014",
doi = "10.14257/ijt.2014.2.1.01",
language = "English",
volume = "2",
pages = "1--20",
journal = "International journal of transportation",
issn = "2287-7940",
publisher = "Science and Engineering Research Support Society",
number = "1",

}

Performance of a genetic algorithm for solving the multi-objective, multimodel transportation network design problem. / Brands, Ties; van Berkum, Eric C.

In: International journal of transportation, Vol. 2, No. 1, 2014, p. 1-20.

Research output: Contribution to journalArticleAcademicpeer-review

TY - JOUR

T1 - Performance of a genetic algorithm for solving the multi-objective, multimodel transportation network design problem

AU - Brands, Ties

AU - van Berkum, Eric C.

PY - 2014

Y1 - 2014

N2 - The optimization of infrastructure planning in a multimodal network is defined as a multi-objective network design problem, with accessibility, use of urban space by parking, operating deficit and climate impact as objectives. Decision variables are the location of park and ride facilities, train stations and the frequency of public transport lines. For a case study the Pareto set is estimated by the Non-dominated Sorting Genetic Algorithm (NSGA-II). Such a Pareto set is one specific outcome of the optimization process, for a specific value of the parameters generation size, number of generations and mutation rate and for a specific outcome of the Monte Carlo simulation within NSGA-II. Similar issues exist for many other metaheuristics. However, when applied in practice, a policy maker desires a result that is robust for these unknown aspects of the method. In this paper Pareto sets from various runs of the NSGA-II algorithm are analyzed and compared. To compare the values of the decision variables in the Pareto sets, new methods are necessary, so these are defined and applied. The results show that the differences concerning decision variables are considerably larger than the differences concerning objectives. This indicates that the randomness of the algorithm may be a problem when determining the decisions to be made. Furthermore, it is concluded that variations caused by different parameters are comparable with the variations caused by randomness within the algorithm.

AB - The optimization of infrastructure planning in a multimodal network is defined as a multi-objective network design problem, with accessibility, use of urban space by parking, operating deficit and climate impact as objectives. Decision variables are the location of park and ride facilities, train stations and the frequency of public transport lines. For a case study the Pareto set is estimated by the Non-dominated Sorting Genetic Algorithm (NSGA-II). Such a Pareto set is one specific outcome of the optimization process, for a specific value of the parameters generation size, number of generations and mutation rate and for a specific outcome of the Monte Carlo simulation within NSGA-II. Similar issues exist for many other metaheuristics. However, when applied in practice, a policy maker desires a result that is robust for these unknown aspects of the method. In this paper Pareto sets from various runs of the NSGA-II algorithm are analyzed and compared. To compare the values of the decision variables in the Pareto sets, new methods are necessary, so these are defined and applied. The results show that the differences concerning decision variables are considerably larger than the differences concerning objectives. This indicates that the randomness of the algorithm may be a problem when determining the decisions to be made. Furthermore, it is concluded that variations caused by different parameters are comparable with the variations caused by randomness within the algorithm.

KW - IR-101342

KW - METIS-294089

U2 - 10.14257/ijt.2014.2.1.01

DO - 10.14257/ijt.2014.2.1.01

M3 - Article

VL - 2

SP - 1

EP - 20

JO - International journal of transportation

JF - International journal of transportation

SN - 2287-7940

IS - 1

ER -