The AEC industry is nowadays one of the most hazardous industries in the world. The construction sector employees about 7% of the world’s work force but is responsible for 30-40% of fatalities. As statistics demonstrate, interferences between workers-on-foot and moving vehicles have caused several injuries and fatalities over the years. Despite safety organizational measures, passive safety devices imposed by regulations and efforts from training procedures, scarce improvements have been recorded. Recent research studies propose technology driven approaches as the key solutions to integrate standard health and safety management practices. This is motivated by the evidence that the dynamics of complex systems can hardly be predicted; rather a proactive approach to health and safety is more effective. Current technologies installed on construction equipment can usually react according to a strict logic, such as sending proximity alerts when workers and equipment are too close. Nevertheless, these approaches barely do make informed decisions in real-time, e.g. including the level of reactiveness of the endangered worker. In similar circumstances a digital twin of the construction site, updated by real-time data from sensors and enriched by artificial intelligence, can pro-actively support activities, forecasting dangerous scenarios on the base of several factors. In this paper a laboratory mock-up has been assumed as the test case, supported by a game engine, which is able to replicates the job site for the execution of bored piles. In such a scenario populated by an avatar of a sensor-equipped worker and a virtual driller, a Bayesian network, implemented within the game engine and fed in runtime by sensor data, works out collision probability in real-time in order to send warnings and avoid fatal accidents.

Development of a Digital Twin Model for Real-Time Assessment of Collision Hazards / Messi, Leonardo; Naticchia, Berardo; Carbonari, Alessandro; Ridolfi, Luigi; DI GIUDA, GIUSEPPE MARTINO. - ELETTRONICO. - (2020), pp. 14-19. (Intervento presentato al convegno Creative Construction e-Conference tenutosi a on-line nel 28 June – 1 July 2020) [10.3311/CCC2020-003].

Development of a Digital Twin Model for Real-Time Assessment of Collision Hazards

Leonardo Messi
Primo
;
Berardo Naticchia;Alessandro Carbonari;Luigi Ridolfi;Giuseppe Martino Di Giuda
2020-01-01

Abstract

The AEC industry is nowadays one of the most hazardous industries in the world. The construction sector employees about 7% of the world’s work force but is responsible for 30-40% of fatalities. As statistics demonstrate, interferences between workers-on-foot and moving vehicles have caused several injuries and fatalities over the years. Despite safety organizational measures, passive safety devices imposed by regulations and efforts from training procedures, scarce improvements have been recorded. Recent research studies propose technology driven approaches as the key solutions to integrate standard health and safety management practices. This is motivated by the evidence that the dynamics of complex systems can hardly be predicted; rather a proactive approach to health and safety is more effective. Current technologies installed on construction equipment can usually react according to a strict logic, such as sending proximity alerts when workers and equipment are too close. Nevertheless, these approaches barely do make informed decisions in real-time, e.g. including the level of reactiveness of the endangered worker. In similar circumstances a digital twin of the construction site, updated by real-time data from sensors and enriched by artificial intelligence, can pro-actively support activities, forecasting dangerous scenarios on the base of several factors. In this paper a laboratory mock-up has been assumed as the test case, supported by a game engine, which is able to replicates the job site for the execution of bored piles. In such a scenario populated by an avatar of a sensor-equipped worker and a virtual driller, a Bayesian network, implemented within the game engine and fed in runtime by sensor data, works out collision probability in real-time in order to send warnings and avoid fatal accidents.
2020
978-615-5270-62-8
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11566/290354
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact