Research on psychophysiological characteristics of construction workers during consciously unsafe behaviors

• Published in: Heliyon Vol. 9; no. 10; p. e20484
• Main Authors: Li, Xiangchun, Long, Yuzhen, Yang, Chunli, Li, Qin, Lu, Weidong, Gao, Jiaxing
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.10.2023; Elsevier
• Subjects: decision support systems; Decision tree regressor; Multiple linear regression; Physiological characteristic; prediction; psychology; questionnaires; regression analysis; risk perception; Risky psychology; skin temperature; Unsafe behavior prediction; Risky psychology; Decision tree regressor; Physiological characteristic; Multiple linear regression; Unsafe behavior prediction
• ISSN: 2405-8440; 2405-8440
• DOI: 10.1016/j.heliyon.2023.e20484

Workers' unsafe behavior is a primary cause leading to falling accidents on construction sites. This study aimed to explore how to utilize psychophysiological characteristics to predict consciously unsafe behaviors of construction workers. In this paper, a psychological questionnaire was compiled to measure risky psychology, and wireless wearable physiological recorders were employed to real-timely measure the physiological signals of subjects. The psychological and physiological characteristics were identified by correlation analysis and significance test, which were then utilized to develop unsafe behavior prediction models based on multiple linear regression and decision tree regressor. It was revealed that unsafe behavior performance was negatively correlated with task-related risk perception, while positively correlated with hazardous attitude. Subjects experienced remarkable increases in skin conductivity, while notable decreases in the inter-beat interval and skin temperature during consciously unsafe behavior. Both models developed for predicting unsafe behavior were reliably and well-fitted with coefficients of determination higher than 0.8. Whereas, each model exhibited its unique advantages in terms of prediction accuracy and interpretability. Not only could study results contribute to the body of knowledge on intrinsic mechanisms of unsafe behavior, but also provide a theoretical basis for the automatic identification of workers’ unsafe behavior. •Unsafe behaviors were simulated and physiological signals were measured.•Unsafe behavior was influenced by risky psychological factors.•Subjects showed remarkable changes in SC, IBI, and SKT during unsafe behaviors.•Unsafe behavior prediction models were developed based on these characteristics.