Please use this identifier to cite or link to this item: http://www.repositorio.ufc.br/handle/riufc/31645
Title in Portuguese: A dynamic linear model for the estimation of time-varying origin–destination matrices from link counts
Author: Pitombeira Neto, Anselmo Ramalho
Loureiro, Carlos Felipe Grangeiro
Keywords: Transportes
Modelos lineares (Estatística)
Matrizes
Dynamic linear models
Matrices
Issue Date: 2016
Publisher: JOURNAL OF ADVANCED TRANSPORTATION
Citation: PITOMBEIRA NETO, A. R.; LOUREIRO, C. F. G. A dynamic linear model for the estimation of time-varying origin–destination matrices from link counts. J. Adv. Transp., v. 50, p. 2116-2129, 2016.
Abstract: We propose a dynamic linear model (DLM) for the estimation of day-to-day time-varying origin– destination (OD) matrices from link counts. Mean OD flows are assumed to vary over time as a locally constant model. We take into account variability in OD flows, route flows, and link volumes. Given a time series of observed link volumes, sequential Bayesian inference is applied in order to estimate mean OD flows. The conditions under which mean OD flows may be estimated are established, and computational studies on two benchmark transportation networks from the literature are carried out. In both cases, the DLM converged to the unobserved mean OD flows when given sufficient observations of traffic link volumes despite assuming uninformative prior OD matrices. We discuss limitations and extensions of the proposed DLM. Copyright © 2017 John Wiley & Sons, Ltd.
URI: http://www.repositorio.ufc.br/handle/riufc/31645
metadata.dc.type: Artigo de Periódico
ISSN: 0197-6729
Appears in Collections:DET - Artigos publicados em revista científica

Files in This Item:
File Description SizeFormat 
2016_art_cfgloureiro_dynamic.pdf719,3 kBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.