An automated safety risk recognition mechanism for underground construction at the pre-construction stage based on BIM

A highly challenging problem that has been plaguing regulators is the inaccurate collection and sharing of safety risks for underground construction. In order to establish a mechanism that could achieve timely and accurate recognition at the pre-construction stage, Building information model (BIM) is adopted as the risk recognition platform. This paper provides the automated safety risk recognition process based on BIM which is generally composed of three parts. The first part is to build the risk database. By means of knowledge structuralizing, questionnaires, depth interviews and group decision-making, explicit and tacit knowledge source are acquired. The safety risk knowledge source is divided into three categories. SQL database is used to express the safety risk knowledge, and all safety risks are stored in the BIM-cloud. The second part is to analyse the relation between engineering information and safety risks. Risk-related engineering information is extracted from BIM models. Backus-Naur form is used to describe the syntax of languages used in computing. A mapping table of engineering information and safety risks is established. The third part concerns the automated safety risk recognition mechanism in the BIM platform. The safety risk recognition mechanism is expressed as “If e, then h (CF (h, e), λ)”. The confidence level as the link is adopted to reveal the mechanism. Finally, a case about flowing sand risk at the foundation pit bottom is conducted. The reasoning process and recognition results are demonstrated. The paper concludes by summarizing the main scientific contribution and giving direction to future research in this field.