Predicting Safety Risks in Deep Foundation Pits in Subway Infrastructure Projects: Support Vector Machine Approach

Abstract

Accurately predicting risks during the construction of deep foundation pits is pivotal to ensuring the safety of the workforce of public and adjacent structures. Existing methods for assessing such risks are cumbersome and are unable to accurately provide the certainty required to ensure safety levels. This paper presents a novel prediction method that utilizes the support vector machine (SVM) to determine the safety risks that can materialize during the construction of deep pit foundations in subway infrastructure projects. The development of the SVM risk prediction model involves the following steps: (1) identification of risk factors from industry experts; (2) processing the sampled data; and (3) training and testing. A case study is used to demonstrate the predictive capability of the developed SVM approach. By inputting data on a daily basis, the safety risks associated with deep foundation pits can be monitored; this enables decision-makers to formulate appropriate control measures.