A reliability-based network design problem

Piya Chootinan, S. C. Wong, Anthony Chen

Research output: Journal article publicationJournal articleAcademic researchpeer-review

67 Citations (Scopus)

Abstract

This paper presents a reliability-based network design problem. A network reliability concept is embedded into the continuous network design problem in which travelers' route choice behavior follows the stochastic user equilibrium assumption. A new capacity-reliability index is introduced to measure the probability that all of the network links are operated below their capacities when serving different traffic patterns deviating from the average condition. The reliability-based network design problem is formulated as a bi-level program in which the lower level sub-program is the probit-based stochastic user equilibrium problem and the upper level sub-program is the maximization of the new capacity reliability index. The lower level sub-program is solved by a variant of the method of successive averages using the exponential average to represent the learning process of network users on a daily basis that results in the daily variation of traffic-flow pattern, and Monte Carlo stochastic loading. The upper level sub-program is tackled by means of genetic algorithms. A numerical example is used to demonstrate the concept of the proposed framework.
Original languageEnglish
Pages (from-to)247-270
Number of pages24
JournalJournal of Advanced Transportation
Volume39
Issue number3
DOIs
Publication statusPublished - 1 Jan 2005
Externally publishedYes

Keywords

  • Bi-level program
  • Capacity reliability
  • Continuous network design problem
  • Reliability analysis
  • Stochastic user equilibrium

ASJC Scopus subject areas

  • Automotive Engineering
  • Economics and Econometrics
  • Mechanical Engineering
  • Computer Science Applications
  • Strategy and Management

Cite this