Mapping of attributes used in construction accidents prediction models using Machine Learning techniques
Mapping of attributes used in construction accidents prediction models using ML techniques
DOI:
https://doi.org/10.46421/entac.v20i1.6180Keywords:
Safety management, Accident, Machine learning, Predictive modelsAbstract
The low occupational health and safety performance faced by the construction industry impacts countries' gross domestic product and brings financial losses to companies and workers. In this context, several predictive models using machine learning (ML) have been developed using historical data on construction accidents. However, there is still no consensus on the types of attributes that should be input into these models to achieve more assertive predictions. Therefore, this paper aims to survey and categorize the most commonly used attributes in predictive models using ML. To this end, a systematic review aims to identify the main research of accident predictive models. The findings show six main categories of attributes; the most frequent attributes were those related to workers, organization, and safety management. Thus, this paper aims to define what types of data related to accidents should be collected to assist in predicting accidents and consequently improving safety performance.
References
POH, C. Q. X.; UBEYNARAYANA, C. U.; GOH, Y. M. Safety leading indicators for construction sites: A machine learning approach.
Automation in Construction, v. 93, p. 375–386, 1 set. 2018.
CHOI, J. et al. Machine learning predictive model based on national data for fatal accidents of construction workers. Automation in
Construction, v. 110, 2020.
ANAMT - ASSOCIAÇÃO NACIONAL DE MEDICINA DO TRABALHO. Divulgadas as estatísticas de acidentes de trabalho para o ano de
Disponível em: <https://www.anamt.org.br/portal/2023/02/08/divulgadas-as-estatisticas-de-acidentes-de-trabalho-para-o-
ano-de-2021/>. Acesso em: 29 mar. 2023.
OIT - ORGANIZAÇÃO INTERNACIONAL DO TRABALHO. Série SmartLab de Trabalho Decente 2022: acidentes de trabalho e mortes
acidentárias voltam a crescer em 2021. Disponível em: <https://www.anamt.org.br/portal/2023/02/08/divulgadas-as-estatisticas-
de-acidentes-de-trabalho-para-o-ano-de-2021/>. Acesso em: 29 mar. 2023.
SAURIN, T. A. Safety inspections in construction sites: A systems thinking perspective. Accident Analysis & Prevention, v. 93, p.
-250, 2016.
MELO, R. R. S.; COSTA, D. B. Integrating resilience engineering and UAS technology into construction safety planning and control.
Engineering, Construction and Architectural Management, v. 26, n. 11, p. 2705-2722, 2019.
LIN, K. et al. A user-centered information and communication technology (ICT) tool to improve safety inspections. Automation in
construction, v. 48, p. 53-63, 2014.
LIU, Y. et al. Computer Vision Technologies and Machine Learning Algorithms for Construction Safety Management: A Critical
Review. [s.l: s.n.].
KANG, K.; KOO, C.; RYU, H. An interpretable machine learning approach for evaluating the feature importance affecting lost
workdays at construction sites. Journal of Building Engineering, v. 53, p. 104534, 2022.
SHUANG, Q.; ZHANG, Z. Determining critical cause combination of fatality accidents on construction sites with machine learning
techniques. Buildings, v. 13, n. 2, p. 345, 2023.
ZHU, J. et al. Developing predictive models of construction fatality characteristics using machine learning. Safety science, v. 164, p.
, 2023.
JAFARI, P. et al. Leading safety indicators: Application of machine learning for safety performance measurement. In: ISARC.
Proceedings of the International Symposium on Automation and Robotics in Construction. IAARC Publications, 2019. p. 501-506.
KOC, K.; EKMEKCIOĞLU, Ö.; GURGUN, A. P. Integrating feature engineering, genetic algorithm and tree-based machine learning
methods to predict the post-accident disability status of construction workers. Automation in Construction, v. 131, p. 103896, 2021.
ZERMANE, A. et al. Predicting fatal fall from heights accidents using random forest classification machine learning model. Safety
science, v. 159, p. 106023, 2023.
LEE, G. et al. Assessment of construction workers’ perceived risk using physiological data from wearable sensors: A machine
learning approach. Journal of Building Engineering, v. 42, p. 102824, 2021.
KOC, K.; EKMEKCIOĞLU, Ö.; GURGUN, A. P. Determining susceptible body parts of construction workers due to occupational
injuries using inclusive modelling. Safety science, v. 164, p. 106157, 2023.
KOC, K.; EKMEKCIOĞLU, Ö.; GURGUN, A. P. Developing a national data-driven construction safety management framework with
interpretable fatal accident prediction. Journal of Construction Engineering and Management, v. 149, n. 4, p. 04023010, 2023b.
ZHU, R. et al. Application of machine learning techniques for predicting the consequences of construction accidents in China.
Process Safety and Environmental Protection, v. 145, p. 293–302, 1 jan. 2021.
LEE, S. H.; SON, J. Development of a safety management system tracking the weight of heavy objects carried by construction
workers using fsr sensors. Applied Sciences, v. 11, n. 4, p. 1–15, 2021.
SEONG, H.; SON, H.; KIM, C. A Comparative Study of Machine Learning Classification for Color-based Safety Vest Detection on
Construction-Site Images. KSCE Journal of Civil Engineering, v. 22, n. 11, p. 4254–4262, 2018.
RADZI, A. R. et al. Relationship between digital twin and building information modeling: a systematic review and future directions.
Construction Innovation, v. 24, n. 3, p. 811-829, 2024.
KOC, K.; EKMEKCIOĞLU, Ö.; GURGUN, A. P. Prediction of construction accident outcomes based on an imbalanced dataset through
integrated resampling techniques and machine learning methods. Engineering, Construction and Architectural Management, v.
, n. 9, p. 4486-4517, 2022.
