Social impacts on hazard perception of construction workers: A system dynamics model analysis

Hui Ma,Zhiguo Wu,Peng Chang

DOI: https://doi.org/10.1016/j.ssci.2021.105240

IF: 6.392

2021-06-01

Safety Science

Abstract:<p>The hazard perception of construction workers plays a predominant role in their behavioral decision-making process under risk environment. However, the evolution mechanisms of hazard perception are still unclear, not only because it is hard to observe and measure, but also because the hazard perception is influenced by multiple social factors and workers have different personalities as well. Interactions between individual and social environment is the root cause of hazard perception's adjustment. Therefore, understanding how these social factors interact and drive hazard perception is important for managers to understand how unsafe behavior is fostered. The study analyzed the interaction and causality between the social factors and individual's hazard perception and categorized multiple feedback loops. Hence a system dynamics model was established and a series of experiments were conducted to simulate how the hazard perception evolves under different group environment. The results indicated that (1) dangerous situation tend to result in high level of hazard perception, but the hazard perception would fluctuate inevitably which explains why fatal injury rate stays high for years. And there is a marginal decreasing effect on safety supervision investment, which means when the investment is over optimal value, even the cost of safety supervision increases, but the effect is not noticeable. (2) The unsafe behaviors which is unlikely to cause serious accident are most difficult to regular. Applying both strict safety supervision and regular safety education may achieve an acceptable result. The result also suggested that the plain and simple but regular education is more effective on contrast to in-depth and systematic but long interval learning of safety knowledge.</p>

engineering, industrial,operations research & management science