Construction worker safety and safety training continue to be the main issues in the construction industry. As a means of improving construction worker safety, this study focuses on safety training at an actual construction worksite. In order to promote safety awareness among workers, it is imperative to develop more effective safety training. This study examined safety training as a method of improving construction worker safety, focusing on the effectiveness of the instructional delivery method. Effectiveness pertains to level of understanding of instruction and can be enhanced through improving instructional delivery method. This study aims to examine two different types of safety training methods: (1) the conventional lecture method and (2) innovative method using the 3D Building Information Modeling (BIM) simulation, reflecting the hazard condition of the actual site. An experiment is conducted, in which the two types of training are implemented and assessed through testing trainees’ understanding. The workers trained via BIM simulation showed a higher level of understanding than the group of workers who were trained conventionally. Also, a survey was conducted targeting safety managers, in which the workers evaluated lifelike quality of the training, active learning and enjoyment that each of the training methods can promote. This research will provide implications that an innovative method using the virtual reality is more effective than the conventional lecture method.

This study goals to develop a model for predicting financial loss at construction sites using a deep learning algorithm to reduce and prevent the risk of financial loss at construction sites. Lately, as the construction of high-rise buildings and complex buildings increases and the scale of construction sites surges, the severity and frequency of accidents occurring at construction sites are swelling, and financial losses are also snowballing. Singularly, as natural disasters rise and construction projects in urban areas increase, the risk of financial loss for construction sites is mounting. Thus, a financial loss prediction model is desired to mitigate and manage the risk of such financial loss for maintainable and effective construction project management. This study reflects the financial loss incurred at the actual construction sites by collecting claim payout data from a major South Korean insurance company. A deep learning algorithm was presented in order to develop an objective and scientific prediction model. The results and framework of this study provide critical guidance on financial loss management necessary for sustainable and successful construction project management and can be used as a reference for various other construction project management studies.

This study analyzed the relative risks of migrant workers, and identified risk factors based on quantitative data for the systematic safety management of migrant workers. Many studies have found that migrant workers are more vulnerable to safety accidents than non-migrant workers. Nevertheless, there are few quantitative studies of migrant workers’ accident-risk in the construction industry, where safety accidents are most frequent. In addition, safety management for the identified accident risk factors has not been implemented systematically. To fill the gap, this study uses safety accident data from construction sites, from the +, for the methodical safety management of migrant workers. The t-test and multiple regression analysis methods are used to define the variance in non-migrant and migrant workers, and the risk indicators, respectively. The two analyses show that the results for migrant construction workers were 2.2% higher in safety accident severity than non-migrant workers, and significant factors are also different. This study’s results will provide critical guidance for the safety management of migrant construction workers.

Extensive use has been made of lifecycle-cost assessment to enhance the cost-effectiveness and resilience of facilities management. However, if such assessments are to be truly effective, supplemental information will be needed on the major costs to be expected over buildings’ entire lives. Electricity generation and distribution systems, for example, are absolutely indispensable to industry and human society, not least in the operation of buildings and other infrastructure as networks. The widespread disruption that ensues when such power systems are damaged often carries considerable repair costs. Natural disasters likewise can cause extensive societal, economic, and environmental damage. Such damage is often associated with lengthy power outages that, as well as being directly harmful, can hinder emergency response and recovery. Accordingly, the present study investigated the correlations of natural hazard indicators such as wind speed and rainfall, along with environmental data regarding the power failure in Florida caused by Hurricane Irma in 2017 utilizing multiple regression analysis. The environmental data in question, selected on the basis of a thorough literature review, was tree density. Our analysis indicated that the independent variables, maximum wind speed, total rainfall, and tree density, were all significantly correlated with the dependent variable, power failure. Among these, rainfall was the least significant. Despite there being only three independent variables in the model, its adjusted coefficient of determination (0.512) indicated its effectiveness as a predictor of the power outages caused by Hurricane Irma. As such, our results can serve the construction industry’s establishment of advanced safety guidelines and structural designs power transmission systems in regions at risk of hurricanes and typhoons. Additionally, insurance companies’ loss-assessment modeling for power-system facilities would benefit from incorporating the three identified risk indicators. Finally, our findings can serve as a useful reference to policymakers tasked with mitigating power outages’ effects on infrastructure in hurricane-prone areas. It is hoped that this work will be extended, facilitating infrastructure restoration planning and making societies and economies more sustainable.

So far, studies for predicting construction safety accidents have mostly been conducted by statistical analysis methods that assume linear models, such as regression and time series analysis. However, it is difficult for this statistical analysis method to reflect the nonlinear characteristics of construction safety accidents determined by complex influencing factors. In general, deep learning techniques are used to analyze the nonlinear characteristics of complex influencing factors. Therefore, the purpose of this study is to propose a framework for developing a deep learning model for predicting safety accidents for sustainable construction. For this study, 1766 cases of actual accidents were collected by the Korea Occupational Safety Authority (KOSHA) over the 10-year period from 2010 to 2019. Eight factors influencing accident prediction such as medical day, progress rate, and construction scale were selected. Subsequently, the predictive power between deep learning models and conventional multi-regression models was compared using actual accident data at construction sites. As a result, a deep neural network (DNN) improved predictive power by 9.3% in mean absolute error (MAE) and 10.6% in root mean square error (RMSE) compared to a conventional multi-regression model. The results of this study provide guidelines for the introduction of deep learning technology to construction safety management.