Abstract
A probabilistic safety analysis methodology based on Bayesian networks models for the probabilistic safety assessment (PSA) of highways and roads is presented. The main idea consists of (a) identifying all the elements encountered when travelling the road, (b) reproducing these elements by sets of variables, (c) identifying the direct dependencies among variables, (d) building a directed acyclic graph to reproduce the qualitative structure of the Bayesian network, and (e) building the conditional probability tables for each variable conditioned on its parent nodes. Since human error is the most important cause of accidents, driver’s tiredness and attention are used to model how the driver’s behaviour evolves with driving and how it is affected by the environment, signs and other factors. A computer program developed in Matlab implements the Bayesian network model from the list of road items and a set of parameter values given by a group of experts. In this way, the most critical elements can be identified and sorted by importance, thus, an improvement of the global safety of the road can be done savings time and money. The proposed methodology is illustrated in real examples of a Spanish highway and a conventional road.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
ENSI refers to the expected number of equivalent severe incidents, where 6.4 medium incidents and 230 light incidents are considered equivalent to one severe incident.
References
Castillo, E., Grande, Z., Calviño, A.: Bayesian networks based probabilistic risk analysis for railway lines. Comput. Aided Civil Infrastruct. Eng. 31, 681–700 (2016)
Castillo, E., Grande, Z., Calviño, A.: A Markovian-Bayesian network for risk analysis of high speed and conventional railway lines integrating human errors. Comput. Aided Civil Infrastruct. Eng. 31, 193–218 (2016)
Pawlovich, M., Li, W., Carriquiry, A., Welch, T.: Experience with road diet measures: use of Bayesian approach to assess impacts on crash frequencies and crash rates. TRR: J. Transp. Res. Board 1953, 163–171 (2006)
Persaud, B., Lyon, C.: Empirical Bayes before-after safety studies: lessons learned from two decades of experience and future directions. Accid. Anal. Prev. 39, 546–555 (2007)
Deublein, M.K.: Roadway accident risk prediction based on Bayesian probabilistic networks. ETH Zürich. Ph.D (2013)
Deublein, M., Schubert, M., Adey, B.T., de Soto García, B.: A Bayesian network model to predict accidents on Swiss highways. Infrastruct. Asset Manag. 2, 145–158 (2015)
Grande, Z., Castillo, E., Mora, E.: Highway and road probabilistic safety assessment based on Bayesian network model. Computer Aided Civil and Infrastructure Engineering (2017, to appear)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2018 Springer International Publishing AG
About this paper
Cite this paper
Mora, E., Grande, Z., Castillo, E. (2018). Bayesian Networks Probabilistic Safety Analysis of Highways and Roads. In: Moreno-Díaz, R., Pichler, F., Quesada-Arencibia, A. (eds) Computer Aided Systems Theory – EUROCAST 2017. EUROCAST 2017. Lecture Notes in Computer Science(), vol 10672. Springer, Cham. https://doi.org/10.1007/978-3-319-74727-9_57
Download citation
DOI: https://doi.org/10.1007/978-3-319-74727-9_57
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-74726-2
Online ISBN: 978-3-319-74727-9
eBook Packages: Computer ScienceComputer Science (R0)