Publication detail

Hybrid Algorithm for Wait-and-see Network Design Problem

HRABEC, D. POPELA, P. ROUPEC, J. JINDRA, P. HAUGEN, K. NOVOTNÝ, J. OLSTAD, A.

Original Title

Hybrid Algorithm for Wait-and-see Network Design Problem

English Title

Hybrid Algorithm for Wait-and-see Network Design Problem

Type

journal article in Scopus

Language

en

Original Abstract

The aim of the paper is to introduce a modified hybrid algorithm to solve a wait-and-see reformulation of transportation optimization model with random demand parameters and 0-1 network design variables. Firstly, the deterministic linear transportation model with network design variables is reviewed. Then, uncertain demand parameters are introduced and modeled by random variables. The following deterministic reformulation is based on the wait-and-see (WS) approach. Finite discrete probability distributions are assumed for all random variables, and hence, the obtained separable scenario-based model can be repeatedly solved as a finite set of mixed integer linear programs (MILPs) by means of integer programming techniques or some heuristics. However, the authors combine a traditional optimization algorithm and a suitable genetic algorithm to obtain a hybrid algorithm that is modified for the WS case. Its implementation and test results illustrated by figures are also discussed in the paper.

English abstract

The aim of the paper is to introduce a modified hybrid algorithm to solve a wait-and-see reformulation of transportation optimization model with random demand parameters and 0-1 network design variables. Firstly, the deterministic linear transportation model with network design variables is reviewed. Then, uncertain demand parameters are introduced and modeled by random variables. The following deterministic reformulation is based on the wait-and-see (WS) approach. Finite discrete probability distributions are assumed for all random variables, and hence, the obtained separable scenario-based model can be repeatedly solved as a finite set of mixed integer linear programs (MILPs) by means of integer programming techniques or some heuristics. However, the authors combine a traditional optimization algorithm and a suitable genetic algorithm to obtain a hybrid algorithm that is modified for the WS case. Its implementation and test results illustrated by figures are also discussed in the paper.

Keywords

transportation model, network-design problem, stochastic programming, wait-and-see approach, genetic algorithm, hybrid algorithm

RIV year

2014

Released

01.07.2014

Location

Brno

ISBN

978-80-214-4984-8

Book

20th International Conference of Soft Computing, MENDEL 2014

Pages from

97

Pages to

104

Pages count

8

Documents

BibTex


@article{BUT108247,
  author="Dušan {Hrabec} and Pavel {Popela} and Jan {Roupec} and Petr {Jindra} and Kjetil Kare {Haugen} and Jan {Novotný} and Asmund {Olstad}",
  title="Hybrid Algorithm for Wait-and-see Network Design Problem",
  annote="The aim of the paper is to introduce a modified hybrid algorithm to solve a wait-and-see reformulation of transportation optimization model with random demand parameters and 0-1 network design variables. Firstly, the deterministic linear transportation model with network design variables is  reviewed. Then, uncertain demand parameters are introduced and modeled by random variables. The following deterministic reformulation is based on the wait-and-see (WS) approach. Finite discrete probability distributions are assumed for all random variables, and hence, the obtained separable scenario-based model can be repeatedly solved as a finite set of mixed integer linear programs (MILPs) by means of integer programming techniques or some heuristics. However, the authors combine a traditional optimization algorithm and a suitable genetic algorithm to obtain a hybrid algorithm that is modified for the WS case. Its implementation and test results illustrated by figures are also discussed in the paper.",
  booktitle="20th International Conference of Soft Computing, MENDEL 2014",
  chapter="108247",
  howpublished="print",
  number="1",
  volume="2014",
  year="2014",
  month="july",
  pages="97--104",
  type="journal article in Scopus"
}