To main content

An Estimator for Traffic Breakdown Probability Based on Classification of Transitional Breakdown Events

Abstract

In this paper we propose a new estimator for calculating the probability of traffic breakdown as a function of traffic demand. Traffic breakdown is a well-studied phenomena within previous literature and is of great importance to traffic planners and controllers. The proposed estimator has an appealing intuition and is able to overcome several of the problems associated with previously proposed methodology. The input to the estimator is a set of aggregated (typically five minute) traffic observations classified to either a breakdown or nonbreakdown state, and a customized and fast algorithm for this purpose is proposed. Last, we apply the classification algorithm and breakdown probability estimator to a large data set consisting of several observation sites on the Norwegian road network, and we compare our estimator to a previously defined estimator.
Read publication

Category

Academic article

Language

English

Affiliation

  • SINTEF Community / Mobility and Economics

Year

2018

Published in

Transportation Science

ISSN

0041-1655

Publisher

Institute for Operations Research and the Management Sciences (INFORMS)

Volume

52

Issue

3

Page(s)

593 - 602

View this publication at Cristin