Critical Link Identification for Traffic Incident Management Programme under Degradable Stochastic Network

Agachai Sumalee, Paramet Luathep, Hing Keung William Lam

Research output: Journal article publicationJournal articleAcademic researchpeer-review

Abstract

This paper proposed a new index called traffic incident management ratio (TIMR) for identifying appropriate roads to be installed with traffic incident management system. The TIMR is defined as the ratio between the expected total travel costs of the degradable transport network without and with traffic incident management programme (TIMP). The traffic flow pattern under the degradation network without TIMP is assumed to follow User Equilibrium (UE) condition, whereas the system optimum (SO) assignment is used to characterise the degradable network condition controlled by TIMP. The main function of TIMP considered in this paper is re-routing of traffic after the incident. An optimisation programme is then proposed for identifying critical roads in the network for an installation of TIMP. The critical roads for TIMP is identified by maximising the TIMR which implies that incidents on these critical roads can significantly increase the total network travel time without TIMP in which the installation of TIMP on these links can decrease the impacts significantly. The proposed model can also quantify the critical level of link capacity degradation. The model and algorithm are tested with a network to illustrate the application of the proposed model.
Original languageEnglish
Pages (from-to)48-55
Number of pages8
JournalHKIE Transactions Hong Kong Institution of Engineers
Volume16
Issue number4
DOIs
Publication statusPublished - 1 Jan 2009

Keywords

  • Capacity Degradation Evaluation
  • Degradable Transport Network
  • Stochastic Network Design
  • Traffic Incident Management

ASJC Scopus subject areas

  • General Engineering

Fingerprint

Dive into the research topics of 'Critical Link Identification for Traffic Incident Management Programme under Degradable Stochastic Network'. Together they form a unique fingerprint.

Cite this