File Download
  Links for fulltext
     (May Require Subscription)
Supplementary

Article: Numerical algorithms for dynamic traffic demand estimation between zones in a network

TitleNumerical algorithms for dynamic traffic demand estimation between zones in a network
Authors
KeywordsConjugate Gradient Method
Entropy Maximization
Newton's Method
Traffic Demand Estimation
Traffic Network
Issue Date2004
PublisherTaylor & Francis Ltd. The Journal's web site is located at http://www.tandf.co.uk/journals/titles/0305215x.asp
Citation
Engineering Optimization, 2004, v. 36 n. 3, p. 379-400 How to Cite?
AbstractThis paper presents numerical methods for dynamic traffic demand estimation between N zones in a network, where the zones are disjoint subsets of nodes of the network. Traffic is assumed to be generated or absorbed only in the zones and nowhere else in the network. Traffic volumes between zones over a fixed period of time are modeled as independent random variables with unknown means which it is desired to estimate. For each zone, the volume of all incoming and outgoing traffic is counted on a regular basis but no information about the origin or destination of the observed traffic is used. Procedures are suggested for a regular update of estimates of the N(N - 1) mean traffic demands between the zones on the basis of an incoming stream of the 2N traffic counts. The procedures are based on an exponential smoothing scheme and are reminiscent of the expectation maximization (EM) algorithm if smoothing is removed. Fast and reliable numerical algorithms, based on the conjugate gradient method, are presented for normal as well as for Poisson traffic demands. The Poisson case is linked with entropy maximization. Computational tests based on simulated data demonstrate both the numerical and statistical efficiency of the procedures.
Persistent Identifierhttp://hdl.handle.net/10722/156144
ISSN
2023 Impact Factor: 2.2
2023 SCImago Journal Rankings: 0.621
ISI Accession Number ID
References

 

DC FieldValueLanguage
dc.contributor.authorChing, WKen_US
dc.contributor.authorScholtes, Sen_US
dc.contributor.authorZhang, SQen_US
dc.date.accessioned2012-08-08T08:40:34Z-
dc.date.available2012-08-08T08:40:34Z-
dc.date.issued2004en_US
dc.identifier.citationEngineering Optimization, 2004, v. 36 n. 3, p. 379-400en_US
dc.identifier.issn0305-215Xen_US
dc.identifier.urihttp://hdl.handle.net/10722/156144-
dc.description.abstractThis paper presents numerical methods for dynamic traffic demand estimation between N zones in a network, where the zones are disjoint subsets of nodes of the network. Traffic is assumed to be generated or absorbed only in the zones and nowhere else in the network. Traffic volumes between zones over a fixed period of time are modeled as independent random variables with unknown means which it is desired to estimate. For each zone, the volume of all incoming and outgoing traffic is counted on a regular basis but no information about the origin or destination of the observed traffic is used. Procedures are suggested for a regular update of estimates of the N(N - 1) mean traffic demands between the zones on the basis of an incoming stream of the 2N traffic counts. The procedures are based on an exponential smoothing scheme and are reminiscent of the expectation maximization (EM) algorithm if smoothing is removed. Fast and reliable numerical algorithms, based on the conjugate gradient method, are presented for normal as well as for Poisson traffic demands. The Poisson case is linked with entropy maximization. Computational tests based on simulated data demonstrate both the numerical and statistical efficiency of the procedures.en_US
dc.languageengen_US
dc.publisherTaylor & Francis Ltd. The Journal's web site is located at http://www.tandf.co.uk/journals/titles/0305215x.aspen_US
dc.relation.ispartofEngineering Optimizationen_US
dc.rightsThis work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.-
dc.subjectConjugate Gradient Methoden_US
dc.subjectEntropy Maximizationen_US
dc.subjectNewton's Methoden_US
dc.subjectTraffic Demand Estimationen_US
dc.subjectTraffic Networken_US
dc.titleNumerical algorithms for dynamic traffic demand estimation between zones in a networken_US
dc.typeArticleen_US
dc.identifier.emailChing, WK:wching@hku.hken_US
dc.identifier.authorityChing, WK=rp00679en_US
dc.description.naturepreprinten_US
dc.identifier.doi10.1080/0305215042000267045en_US
dc.identifier.scopuseid_2-s2.0-2542486440en_US
dc.identifier.hkuros88736-
dc.relation.referenceshttp://www.scopus.com/mlt/select.url?eid=2-s2.0-2542486440&selection=ref&src=s&origin=recordpageen_US
dc.identifier.volume36en_US
dc.identifier.issue3en_US
dc.identifier.spage379en_US
dc.identifier.epage400en_US
dc.identifier.isiWOS:000221762600006-
dc.publisher.placeUnited Kingdomen_US
dc.identifier.scopusauthoridChing, WK=13310265500en_US
dc.identifier.scopusauthoridScholtes, S=6602466118en_US
dc.identifier.scopusauthoridZhang, SQ=10143093600en_US
dc.identifier.issnl0305-215X-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats