Early-warning of unsafe hoisting operations: An integration of digital twin and knowledge graph

Abstract

Unsafe hoisting operations have been consistently associated with numerous safety incidents involving tower cranes. Currently, the predominant measures to mitigate these operations center around comprehensive training and education, emphasizing standardized protocols prior to hoisting activities. Despite concerted efforts in this direction, a conspicuous research gap persists in early-warning mechanisms during the construction phase. This paper aims to address this gap by proposing an innovative early-warning methodology, inspired by the principles of digital twin and knowledge graph. We firstly introduce a digital twin framework designed to mirror the real-time operational status of the tower crane. This framework enables the immediate detection of deviations or infractions as they occur. Subsequently, we develop a knowledge graph capable of promptly identifying unsafe hoisting operations by leveraging real-time data obtained from the digital twin. To validate the efficacy of our proposed methodology, we construct a scaled-down replica of a tower crane and establish a tailored digital twin system. The findings of a series of experimental trials prominently underscore the system's capability to generate timely alerts in response to unsafe hoisting operations while maintaining an impressively low rate of false alarms.