A fuzzy evidential reasoning-based approach for risk assessment of deep foundation pit

Wei, D; Xu, D; Zhang, Y

Abstract

Traditional risk assessment methods, such as the probabilistic methods, are not effectively used in the construction works of a deep foundation pit (DFP) when data set collected are incomplete or vague input takes place. A new method based on fuzzy evidential reasoning approach is proposed in this paper to assess the overall risk level of a DFP construction project. Firstly, the method defines risks as the products of occurrence likelihood multiplying consequence severity, which is further depicted by trapezoidal fuzzy numbers. Thereafter, the fuzzy analytical hierarchy process is adopted to calculate the weighs of different hazardous events that may occur in a DFP construction project. The overall risk level of a DFP project therefore could be achieved through aggregating the risk level of all hazardous events based on evidential reasoning algorithm. However, due to the existence of intersections among more than two continuous fuzzy evaluation grades rather than between two adjacent grades, the prevailing aggregation method is not suitable any more. So, a new aggregated probability mass along with the reassigning method in relation to the degree of belief belonging to the fuzzy intersection of two grades is thus put forward in this paper, as a result to make the evidential reasoning possible. A case study on risk assessment of the DFP of underground traffic project of Zhengzhou comprehensive transportation hub in China is introduced to illustrate the application of the proposed method. The result indicates that the overall risk level of a DFP project could be assessed effectively under the scenario that more than two continuous fuzzy evaluation grades intersect rather than only two adjacent grades. Moreover, comparing with the traditional methods, the result obtained in the case study by using the proposed method seems to be more reasonable.

Citation

Wei, D., Xu, D., & Zhang, Y. (2020). A fuzzy evidential reasoning-based approach for risk assessment of deep foundation pit. Tunnelling and Underground Space Technology, 97(103232), https://doi.org/10.1016/j.tust.2019.103232

• Journal Article Type: Article
• Acceptance Date: Dec 6, 2019
• Online Publication Date: Dec 20, 2019
• Publication Date: Mar 1, 2020
• Deposit Date: Mar 16, 2020
• Publicly Available Date: Dec 20, 2020
• Journal: Tunnelling and Underground Space Technology
• Print ISSN: 0886-7798
• Publisher: Elsevier
• Volume: 97
• Issue: 103232
• DOI: https://doi.org/10.1016/j.tust.2019.103232
• Publisher URL: https://doi.org/10.1016/j.tust.2019.103232
• Related Public URLs: https://www.sciencedirect.com/journal/tunnelling-and-underground-space-technology
• Additional Information: Funders : ShanXi Science and Technology Department, Xi'An City, Shanxi Province, China; Grant Number: 2018ZDXM-SF-096.