Implications of the computational complexity of transit route network redesign for metaheuristic optimisation systems

Jeremy Joseph Blum, T. V. Mathew

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

The transit route network design problem is a computationally intractable optimisation problem that seeks a set of bus routes and frequencies that minimises the operator cost while maximising passenger utility. Previous attempts to optimise this problem have used metaheuristic and heuristic techniques to find solutions that allow for a complete redesign of the transit network. In reality, however, a complete redesign of the network may encounter political resistance from existing transit users if the routes that they use are eliminated. Here, an intelligent agent optimisation system is used to optimise the transit route network redesign (TRNR) problem, which is subject to the additional constraint that existing routes in the network remain, although perhaps serviced with lower frequency. When applied to the transit network in Mumbai, India, the optimisation system found significant improvement in the route network, even when subject to these constraints. In a scenario in which the current routes were maintained with frequency of stops reduced by no more than 50, operator cost could be improved by 18.1 while maintaining the current level of passenger utility. On the other hand, passenger utility could be improved by 5.5 at current levels of operator cost.

Original languageEnglish (US)
Pages (from-to)124-131
Number of pages8
JournalIET Intelligent Transport Systems
Volume6
Issue number2
DOIs
StatePublished - Jun 1 2012

Fingerprint

Computational complexity
cost
Costs
network design
Intelligent agents
heuristics
costs
India
scenario
bus

All Science Journal Classification (ASJC) codes

  • Transportation
  • Environmental Science(all)
  • Mechanical Engineering
  • Law

Cite this

@article{4033854667b142a1897a06cc5b6bcbaf,
title = "Implications of the computational complexity of transit route network redesign for metaheuristic optimisation systems",
abstract = "The transit route network design problem is a computationally intractable optimisation problem that seeks a set of bus routes and frequencies that minimises the operator cost while maximising passenger utility. Previous attempts to optimise this problem have used metaheuristic and heuristic techniques to find solutions that allow for a complete redesign of the transit network. In reality, however, a complete redesign of the network may encounter political resistance from existing transit users if the routes that they use are eliminated. Here, an intelligent agent optimisation system is used to optimise the transit route network redesign (TRNR) problem, which is subject to the additional constraint that existing routes in the network remain, although perhaps serviced with lower frequency. When applied to the transit network in Mumbai, India, the optimisation system found significant improvement in the route network, even when subject to these constraints. In a scenario in which the current routes were maintained with frequency of stops reduced by no more than 50, operator cost could be improved by 18.1 while maintaining the current level of passenger utility. On the other hand, passenger utility could be improved by 5.5 at current levels of operator cost.",
author = "Blum, {Jeremy Joseph} and Mathew, {T. V.}",
year = "2012",
month = "6",
day = "1",
doi = "10.1049/iet-its.2011.0021",
language = "English (US)",
volume = "6",
pages = "124--131",
journal = "IET Intelligent Transport Systems",
issn = "1751-956X",
publisher = "Institution of Engineering and Technology",
number = "2",

}

Implications of the computational complexity of transit route network redesign for metaheuristic optimisation systems. / Blum, Jeremy Joseph; Mathew, T. V.

In: IET Intelligent Transport Systems, Vol. 6, No. 2, 01.06.2012, p. 124-131.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Implications of the computational complexity of transit route network redesign for metaheuristic optimisation systems

AU - Blum, Jeremy Joseph

AU - Mathew, T. V.

PY - 2012/6/1

Y1 - 2012/6/1

N2 - The transit route network design problem is a computationally intractable optimisation problem that seeks a set of bus routes and frequencies that minimises the operator cost while maximising passenger utility. Previous attempts to optimise this problem have used metaheuristic and heuristic techniques to find solutions that allow for a complete redesign of the transit network. In reality, however, a complete redesign of the network may encounter political resistance from existing transit users if the routes that they use are eliminated. Here, an intelligent agent optimisation system is used to optimise the transit route network redesign (TRNR) problem, which is subject to the additional constraint that existing routes in the network remain, although perhaps serviced with lower frequency. When applied to the transit network in Mumbai, India, the optimisation system found significant improvement in the route network, even when subject to these constraints. In a scenario in which the current routes were maintained with frequency of stops reduced by no more than 50, operator cost could be improved by 18.1 while maintaining the current level of passenger utility. On the other hand, passenger utility could be improved by 5.5 at current levels of operator cost.

AB - The transit route network design problem is a computationally intractable optimisation problem that seeks a set of bus routes and frequencies that minimises the operator cost while maximising passenger utility. Previous attempts to optimise this problem have used metaheuristic and heuristic techniques to find solutions that allow for a complete redesign of the transit network. In reality, however, a complete redesign of the network may encounter political resistance from existing transit users if the routes that they use are eliminated. Here, an intelligent agent optimisation system is used to optimise the transit route network redesign (TRNR) problem, which is subject to the additional constraint that existing routes in the network remain, although perhaps serviced with lower frequency. When applied to the transit network in Mumbai, India, the optimisation system found significant improvement in the route network, even when subject to these constraints. In a scenario in which the current routes were maintained with frequency of stops reduced by no more than 50, operator cost could be improved by 18.1 while maintaining the current level of passenger utility. On the other hand, passenger utility could be improved by 5.5 at current levels of operator cost.

UR - http://www.scopus.com/inward/record.url?scp=84866863844&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84866863844&partnerID=8YFLogxK

U2 - 10.1049/iet-its.2011.0021

DO - 10.1049/iet-its.2011.0021

M3 - Article

AN - SCOPUS:84866863844

VL - 6

SP - 124

EP - 131

JO - IET Intelligent Transport Systems

JF - IET Intelligent Transport Systems

SN - 1751-956X

IS - 2

ER -