Search

The significance of accident prevention in the construction industry has been consistently emphasized. Despite the research into various preventive methodologies, their application in actual construction sites still needs to be improved. Therefore, it is necessary to develop accident prevention methodologies and make improvements so that practitioners can utilize them in actual construction sites. This study sets out to propose methods for accident prevention using construction accident data and to develop a system that construction practitioners can easily utilize. Firstly, a methodology was developed to assess the risk of construction sites by considering site-specific characteristics through actual construction accident data. This led to the identification of five principal risk factors (i.e., construction type, facility type, ordering organization, construction cost, and safety management plan), and the results of risk assessment for different site classifications were derived by combining these factors. Then, machine-learning models were developed to predict accident-causing objects, accident types, and injury-death using four algorithms: random forest, light gradient boosting model, eXtreme gradient boosting (XGBoost), and categorical boosting. As a result, XGBoost demonstrated the most outstanding predictive performance, with averaged F1 scores of 0.839, 0.749, and 0.977, respectively. A web-based prototype was also developed to deploy the proposed methods, confirming the practical utility of the accident prevention strategies outlined in this research. These findings have the potential to improve the efficacy of site management by enabling proactive identification of high-risk areas, thereby promoting more effective accident prevention initiatives. [ABSTRACT FROM AUTHOR]

The growth of the global construction market has attracted international companies to participate in overseas projects. Overseas projects are extremely dynamic with numerous uncertainties, raising the need to collect information about construction in host countries. Due to the vast amounts of text data in the construction industry, an automated method, specifically information retrieval, is required to find the necessary information. Previous studies have suggested automated methods to review various construction documents. However, these studies required substantial manual effort and mainly focused on only one language, resulting in loss of vital information because it is buried in documents written in the host country's language. To address these limitations, this study proposes a cross-lingual information retrieval (CLIR) framework using pretrained Bidirectional Encoder Representations from Transformers (BERT) models to retrieve information from multilingual construction documents. The proposed framework employs language models (i.e., monolingual, multilingual, and cross-lingual) and trains these models on a construction data set to enhance their ability in construction-specific text. The framework achieved reliable performance of retrieval, even with minimal additional training using domain-specific data. The results indicate that training on the domain data set raises the level of retrieval, increasing the mean reciprocal rank of a specific task by up to 0.2128. With the employment of a monolingual model with machine translation, CLIR in a specific domain could be performed effectively without the need for a labeled data set. The suggested CLIR framework offers a practical alternative for dealing with construction documents in overseas projects, reducing time and cost while improving risk identification and mitigation. [ABSTRACT FROM AUTHOR]