A bayesian belief network based delay risk assessment tool for tunnel projects – BBN tunnel

---

Tezin Türü: Doktora

Tezin Yürütüldüğü Kurum: Orta Doğu Teknik Üniversitesi, Fen Bilimleri Enstitüsü, Türkiye

Tezin Onay Tarihi: 2019

Tezin Dili: İngilizce

Öğrenci: GÜLSÜM ÇAĞIL KÖSEOĞLU BALTA

Asıl Danışman (Eş Danışmanlı Tezler İçin): Mustafa Talat Birgönül

Eş Danışman: Irem Dikmen Toker

Özet:

Tunnel constructions are characterized by high degrees of uncertainty, due to two major factors; -geologic conditions, which can seldom be exactly known and -uncertainties in construction process itself as it highly depends on the performance of the equipment and workmanship. Therefore, due to the specific properties of tunnel construction projects, there is an increasing urgency to assess and manage the risks systematically. Initially, an extensive literature review was carried out to identify risks and proposed methods for risk identification in tunneling projects. Then, to gain insight into the practice of risk assessment of tunneling projects within the industry, current practices in a construction company are investigated and research needs are determined. In the light of these findings, major aims of the thesis are identified as; construction of a risk taxonomy that links risk with delay, development of a methodology for risk assessment and a tool that can be used to identify risk mitigation strategies to minimize delay. In collaboration with a construction company, first, major risk events, vulnerability and risk factors were determined, and a taxonomy was developed. Then, a dependency based probabilistic risk analysis method based on Bayesian Belief Networks (BBNs) was proposed. BBN model was developed and validated by utilizing several expert knowledge elicitation techniques. Finally, a decision support tool, BBN Tunnel, that can predict delay and estimate the cost-time impact of utilizing different strategies was developed. BBN Tunnel was tested, validated and its utilization in a real project was demonstrated by a case study. Results demonstrate that the methodology and tool may be used to integrate several risk factors, draw a comprehensive risk map, predict delay and help decision-makers to formulate risk management strategies to mitigate delay. .