HSE Management Office, Energy Systems Digital Innovation Headquarters, Mitsubishi Heavy Industries, Ltd.
Nowadays, the construction industry faces problems such as aging of the workforce, retirement of skilled workers and a decline in the number of workers. As a result, the temporary employment of inexperienced workers is increasing at construction sites. Furthermore, in the construction of a thermal power plant, which is one of the businesses of the energy domain of Mitsubishi Heavy Industries, Ltd. (MHI), Energy Systems, there are dangers such as heavy machinery, fire, and working in high places, and the risk of accidents is increasing since safety awareness is declining due to lack of human resources and lack of experience. On the other hand, social demands and other factors require we further strengthen safety management. Therefore, in cooperation with Digital Innovation Headquarters, Energy Systems has created an unprecedented indicator for predicting whether an accident will occur (hereinafter, leading indicator) as an indicator for examining safety management measures. This indicator is calculated and output by Artificial Intelligence (AI), machine learning model, based on information such as the number of workers, working hours, and supervisors at construction sites, and its prediction accuracy improves along with data accumulation. Using this system, we are working to reduce the number of accidents at construction sites by implementing proactive safety activities to prevent accidents from happening. This report describes the process of creating this leading indicator and its current utilization.
