The University of Southampton
University of Southampton Institutional Repository

Speed limit reduction in urban areas: A before-after study using Bayesian generalized mixed linear models

Speed limit reduction in urban areas: A before-after study using Bayesian generalized mixed linear models
Speed limit reduction in urban areas: A before-after study using Bayesian generalized mixed linear models

In fall 2009, a new speed limit of 40 km/h was introduced on local streets in Montreal (previous speed limit: 50 km/h). This paper proposes a methodology to efficiently estimate the effect of such reduction on speeding behaviors. We employ a full Bayes before-after approach, which overcomes the limitations of the empirical Bayes method. The proposed methodology allows for the analysis of speed data using hourly observations. Therefore, the entire daily profile of speed is considered. Furthermore, it accounts for the entire distribution of speed in contrast to the traditional approach of considering only a point estimate such as 85th percentile speed. Different reference speeds were used to examine variations in the treatment effectiveness in terms of speeding rate and frequency. In addition to comparing rates of vehicles exceeding reference speeds of 40 km/h and 50 km/h (speeding), we verified how the implemented treatment affected "excessive speeding" behaviors (exceeding 80 km/h). To model operating speeds, two Bayesian generalized mixed linear models were utilized. These models have the advantage of addressing the heterogeneity problem in observations and efficiently capturing potential intra-site correlations. A variety of site characteristics, temporal variables, and environmental factors were considered. The analyses indicated that variables such as lane width and night hour had an increasing effect on speeding. Conversely, roadside parking had a decreasing effect on speeding. One-way and lane width had an increasing effect on excessive speeding, whereas evening hour had a decreasing effect. This study concluded that although the treatment was effective with respect to speed references of 40 km/h and 50 km/h, its effectiveness was not significant with respect to excessive speeding-which carries a great risk to pedestrians and cyclists in urban areas. Therefore, caution must be taken in drawing conclusions about the effectiveness of speed limit reduction. This study also points out the importance of using a comparison group to capture underlying trends caused by unknown factors.

Bayesian generalized mixed linear models, Before-after studies, Excessive speeding, Speed limit reduction, Speeding
0001-4575
252-261
Heydari, Shahram
0d12a583-a4e8-4888-9e51-a50d312be1e9
Miranda-Moreno, Luis F.
b61c4a8f-b48e-4c04-b051-3184945da9e4
Liping, Fu
5a8cfcc4-d76e-4456-b4e0-7877de2a0eb1
Heydari, Shahram
0d12a583-a4e8-4888-9e51-a50d312be1e9
Miranda-Moreno, Luis F.
b61c4a8f-b48e-4c04-b051-3184945da9e4
Liping, Fu
5a8cfcc4-d76e-4456-b4e0-7877de2a0eb1

Heydari, Shahram, Miranda-Moreno, Luis F. and Liping, Fu (2014) Speed limit reduction in urban areas: A before-after study using Bayesian generalized mixed linear models. Accident Analysis and Prevention, 73, 252-261. (doi:10.1016/j.aap.2014.09.013).

Record type: Article

Abstract

In fall 2009, a new speed limit of 40 km/h was introduced on local streets in Montreal (previous speed limit: 50 km/h). This paper proposes a methodology to efficiently estimate the effect of such reduction on speeding behaviors. We employ a full Bayes before-after approach, which overcomes the limitations of the empirical Bayes method. The proposed methodology allows for the analysis of speed data using hourly observations. Therefore, the entire daily profile of speed is considered. Furthermore, it accounts for the entire distribution of speed in contrast to the traditional approach of considering only a point estimate such as 85th percentile speed. Different reference speeds were used to examine variations in the treatment effectiveness in terms of speeding rate and frequency. In addition to comparing rates of vehicles exceeding reference speeds of 40 km/h and 50 km/h (speeding), we verified how the implemented treatment affected "excessive speeding" behaviors (exceeding 80 km/h). To model operating speeds, two Bayesian generalized mixed linear models were utilized. These models have the advantage of addressing the heterogeneity problem in observations and efficiently capturing potential intra-site correlations. A variety of site characteristics, temporal variables, and environmental factors were considered. The analyses indicated that variables such as lane width and night hour had an increasing effect on speeding. Conversely, roadside parking had a decreasing effect on speeding. One-way and lane width had an increasing effect on excessive speeding, whereas evening hour had a decreasing effect. This study concluded that although the treatment was effective with respect to speed references of 40 km/h and 50 km/h, its effectiveness was not significant with respect to excessive speeding-which carries a great risk to pedestrians and cyclists in urban areas. Therefore, caution must be taken in drawing conclusions about the effectiveness of speed limit reduction. This study also points out the importance of using a comparison group to capture underlying trends caused by unknown factors.

Full text not available from this repository.

More information

Accepted/In Press date: 11 September 2014
e-pub ahead of print date: 27 September 2014
Published date: December 2014
Keywords: Bayesian generalized mixed linear models, Before-after studies, Excessive speeding, Speed limit reduction, Speeding

Identifiers

Local EPrints ID: 424167
URI: http://eprints.soton.ac.uk/id/eprint/424167
ISSN: 0001-4575
PURE UUID: 7122f018-ab3d-430f-9367-d6575ae431ec

Catalogue record

Date deposited: 05 Oct 2018 11:31
Last modified: 07 Oct 2020 00:21

Export record

Altmetrics

Download statistics

Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.

View more statistics

Atom RSS 1.0 RSS 2.0

Contact ePrints Soton: eprints@soton.ac.uk

ePrints Soton supports OAI 2.0 with a base URL of http://eprints.soton.ac.uk/cgi/oai2

This repository has been built using EPrints software, developed at the University of Southampton, but available to everyone to use.

We use cookies to ensure that we give you the best experience on our website. If you continue without changing your settings, we will assume that you are happy to receive cookies on the University of Southampton website.

×