AN EVIDENCE-BASED RISK DECISION SUPPORT APPROACH FOR METRO TUNNEL CONSTRUCTION

The risk-informed decision-making of metro tunnel project is often faced with the problem of inadequate utilization of available information. In order to address the epistemic uncertainty problem caused by insufficient utilization of information in decision-making, this paper proposes a risk decisio...

Full description

Saved in:
Bibliographic Details
Published inJournal of civil engineering and management Vol. 28; no. 5; pp. 377 - 396
Main Authors Guo, Yifan, Zheng, Junjie, Zhang, Rongjun, Yang, Youbin
Format Journal Article
LanguageEnglish
Published Vilnius Vilnius Gediminas Technical University 01.05.2022
Subjects
Acceptability
Acrylonitrile butadiene resins
Anomalies
Bayesian analysis
Case studies
Construction
Continuous Time Bayesian Network
Decision making
Decision support systems
evidence
Fault trees
Feasibility studies
History
Information processing
knowledge
Methods
multi-source information
Probability theory
Railroad tunnels
Risk analysis
Risk factors
risk-informed decision-making
Safety engineering
Subway construction
Subway tunnels
Transportation authorities
Tunnel construction
Tunnels
China
Online AccessGet full text

Cover

More Information
Summary:The risk-informed decision-making of metro tunnel project is often faced with the problem of inadequate utilization of available information. In order to address the epistemic uncertainty problem caused by insufficient utilization of information in decision-making, this paper proposes a risk decision support approach for metro tunnel construction based on Continuous Time Bayesian Network (CTBN) technique. CTBN can factor the state space of variables in tunnel projects and perform evidence-based reasoning, which enables the diverse information of expert opinions, project-specific parameters, historical data and engineering anomalies to be the evidence to support decision-making. A concise CTBN model development method based on Dynamic Fault Trees is presented to replace the cumbersome model learning process. The proposed approach can utilize multi-source information as evidence to provide multi-form decision support both in the pre-construction stage and construction stage of the tunnel construction project, and the results can support the decisions on judging the acceptability of the risk, developing response strategies for risk factors and diagnosing the causes of the hazardous event. A case study on the water leakage risk of tunnel construction in China is presented to illustrate the feasibility of the approach. The case study shows that the approach can assist in making informed decisions, so as to improve the engineering safety.
ISSN:1392-3730
1822-3605
DOI:10.3846/jcem.2022.16807