Accident risk assessment in marine transportation via Markov modelling and Markov Chain Monte Carlo simulation

Research output: Journal Publications and ReviewsRGC 21 - Publication in refereed journalpeer-review

102 Scopus Citations
View graph of relations

Author(s)

  • Shahrzad Faghih-Roohi
  • Min Xie
  • Kien Ming Ng

Detail(s)

Original languageEnglish
Pages (from-to)363-370
Journal / PublicationOcean Engineering
Volume91
Online published15 Oct 2014
Publication statusPublished - 15 Nov 2014

Abstract

There are many technological and environmental safety factors involved in marine accidents. This paper deals with an analytic approach to accident risk modelling when data for analyzing safety factors is limited or unavailable. The purpose of this paper is to propose a simulated accident model for assessing accident risk in marine transportation. The proposed approach is based on Markov modelling and Markov Chain Monte Carlo (MCMC) simulation and it is illustrated using an example from marine transportation. A three-state continuous time Markov model is used to record and estimate marine occurrence rates and probabilities. The MCMC simulation requires the occurrence data of the Markov model to estimate the accident risk. However, it can be used when only a limited amount of information is available. Compared with other models, the approach in this paper is applicable to any type of marine accident or marine transportation system. A numerical example is also given to illustrate the procedure.

Research Area(s)

  • Accident risk modeling, Marine transportation, Markov Chain Monte Carlo simulation, Markov model, Slice sampling algorithm