Modeling Unsafe Behaviors within Construction Crews: A Safety-Based Cognitive Perspective

Abstract Unsafe behaviors among workers are one of the root causes of accidents. Construction crews’ behavioral decision-making is closely correlated with factors such as safety cognition, interpersonal interactions, and environmental conditions. Building upon a comprehensive consideration of the multidimensional factors influencing individual safety cognition and behavioral decisions, this study aims to explore the safety cognitive process for behavioral decision-making and subject interaction within construction crews. An agent-based modeling (ABM) model was developed for the experimental simulation. The model quantitatively assessed the impacts of individual psychological characteristics, supervisors’ management measures, and work situational factors on the emergence and evolution of unsafe behaviors. The results reveal that (1) enhancing workers’ self-efficacy and supervisor identification can benefit safety management within the crew; (2) supervisors’ intensified safety coaching and safety control can reduce the occurrence of unsafe behaviors; and (3) implementing strategies to reduce safety behavior cost for workers can effectively improve safety performance. Unsafe behaviors are more likely to proliferate in low-risk environments, while managers in high-risk environments need to take strong measures to reduce workers’ safety behavior cost and prevent the occurrence of large-scale unsafe behaviors. The research provides a mechanistic explanation for safety decision-making, quantifies the nonlinear effects of psychological characteristics, supervisory interventions and situational factors, and proposes a differentiated safety management strategy based on computational experiments. This study expands the application of safety cognition models and ABM methods in enhancing safety performance.