EM Approach for Weibull Analysis in a Strongly Censored Data Context - Application to Road Markings

doi:10.23940/ijpe.21.04.p1.333342

Abstract

Abstract:

Road surface markings organize road traffic and therefore play a major role in road safety. This is one reason why a part of the current guidance system is to model traffic lanes based on pavement marking detection. Consequently, road managers must guarantee detectable marking lines. An adapted maintenance strategy is a potential solution. Today, marking lines are inspected periodically. A direct consequence of this is that the failure moment of a given marking is generally strongly censored (left, interval or right). Sathyanarayanan et al.present the first Weibull approach for road markings [1]. Weibull distributions are estimated by the maximum likelihood estimation (MLE) method adapted to this censored problem. The MLE is computed by a numerical approach such as the Newton-Raphson algorithm. In the context of interval-censored data, Pradhan and Kundu proposed an alternative method based on an Expectation-Maximisation EM algorithm [2]. This paper proposes replacing this EM approach in the road markings context. A case study regarding the broken center line of a section of French National Road 4 illustrates this methodology. Lifetime markings are distributed by the Weibull distribution.

Key words: reliability analysis, pavement marking, retroreflection, EM algorithm, maximum likelihood estimator, censor, Weibull distribution

Redondin Maxime, Bouillaut Laurent, and Daucher Dimitri. EM Approach for Weibull Analysis in a Strongly Censored Data Context - Application to Road Markings [J]. Int J Performability Eng, 2021, 17(4): 333-342.

Add to citation manager EndNote|Reference Manager|ProCite|BibTeX|RefWorks

References 23.

[1].	SathyanarayananS., ShankarV. and DonnellE.T. Pavement marking retroreflectivity inspection data: a Weibull analysis. Transportation Research Record, 2055(1), pp.63-70, 2008.
[2].	PradhanB. and KunduD. Analysis of interval-censored data with Weibull lifetime distribution. Sankhya B, 76(1), pp.120-139, 2014.
[3].	AFNOR, NF EN1436+A 1 - Maintenance - Maintenance terminology, 2009.
[4].	VECTRA, Ecodyn MLPC - Signing visibility, https://www.nextroad. com/product/ecodyn-3/, 2018.
[5].	LuJ.J. Performance of Traffic Markings in cold regions. Transportation Research Center, Institute of Northen Engineering, University of Alaska Fairbanks, 1995.
[6].	AbboudN. and BowmanB.L. Cost-and longevity-based scheduling of paint and thermoplastic striping. Transportation Research Record, 1794(1), pp.55-62, 2002.
[7].	SitzabeeW.E., HummerJ.E.and RasdorfW. Pavement marking degradation modeling and analysis. Journal of infrastructure systems, 15(3), pp.190-199, 2009.
[8].	SarasuaW.A., ClarkeD.B.and DavisW.J. Evaluation of interstate pavement marking retroreflectivity, 2003.
[9].	KarwaV. and DonnellE.T. Predicting pavement marking retroreflectivity using artificial neural networks: Exploratory analysis. Journal of Transportation Engineering, 137(2), pp.91-103, 2011.
[10].	CastetJ.F.and SalehJ.H. Single versus mixture Weibull distributions for nonparametric satellite reliability. Reliability Engineering & System Safety, 95(3), pp.295-300, 2010.
[11].	McLachlanG.J.and KrishnanT., 2007. The EM algorithm and extensions (Vol. 382). John Wiley & Sons.
[12].	ChauveauD. A stochastic EM algorithm for mixtures with censored data. Journal of statistical planning and inference, 46(1), pp.1-25, 1995.
[13].	BordesL. and ChauveauD. Stochastic EM algorithms for parametric and semiparametric mixture models for right-censored lifetime data. Computational Statistics, 31(4), pp. 1513-1538, 2016.
[14].	ElmahdyE.E. A new approach for Weibull modeling for reliability life data analysis. Applied Mathematics and computation, 250, pp.708-720, 2015.
[15].	BalakrishnanN. and MitraD. Left truncated and right censored Weibull data and likelihood inference with an illustration. Computational Statistics & Data Analysis, 56(12), pp.4011-4025, 2012.
[16].	BagheriS.F., SamaniE.B.and GanjaliM. The generalized modified Weibull power series distribution: Theory and applications. Computational Statistics & Data Analysis, 94, pp.136-160, 2016.
[17].	KunduD. and DeyA.K. Estimating the parameters of the Marshall-Olkin bivariate Weibull distribution by EM algorithm. Computational Statistics & Data Analysis, 53(4), pp.956-965, 2009.
[18].	BainL. Statistical analysis of reliability and life-testing models: theory and methods. Routledge, 2017.
[19].	RedondinM., FaulN., BouillautL., SameA. and DaucherD. Alternative Weibull analysis for road markings: an EM approach. In ESREL2018-European Safety and Reliability Conference ( p.7p), June 2018.
[20].	EvansJ.W. Two-and three-parameter Weibull goodness-of-fit tests ( Vol.493). US Department of Agriculture, Forest Service, Forest Products Laboratory, 1989.
[21].	RedondinM., BouillautL. and DaucherD. A clustering-based approach to segment a pavement markings line. International Journal of Performability Engineering, 16(10), pp-1497, 2020.
[22].	DucrosF. and PamphileP. Bayesian estimation of Weibull mixture in heavily censored data setting. Reliability Engineering & System Safety, 180, pp.453-462, 2018.
[23].	MarsiliF., BödefeldJ., CroceP. and LandiF. Bayesian approaches to lifetime prediction. Safety and Reliability-Safe Societies in a Changing World, pp.707-715, 2018.