SD-GA Model for Intervention Cost Optimization of Construction Workers' Unsafe Behaviors

The construction industry is a high-risk industry, and more than 80% of construction accidents are caused by unsafe behavior of construction workers. Existing papers in this area focus on the influencing factors identification and the intervention effects validation, but no studies of intervention costs. First, 7 influence levels and 18 targeted intervention measures of construction workers' unsafe behaviors are identified and weighted using the AHP. Then, the SD (System Dynamics) model for intervention costs of construction workers' unsafe behaviors is constructed and codified in MATLAB to establish the planning problem with minimum unsafe behavior level as the objective function and solved using GA (Genetic Algorithm). Finally, the results of the empirical analysis show that: the level of construction workers' unsafe behavior that optimize by SD-GA model is lower than before, validating the feasibility of the optimization model. This study can provide an important reference for developing intervention strategies for construction workers' unsafe behavior.