Information modeling in construction integrates with artificial intelligence technologies, digital twins and sensors, which makes it possible to predict risks and improve occupational safety. The article presents a mathematical model for assessing the safety of a construction worker based on a system of Kolmogorov–Chapman differential equations and taking into account the four states of the system associated with the identification and neutralization of labor factors. The model calculates the probabilities of finding a worker in each condition using the intensity of transitions, which optimizes risk management on the construction site. To identify hazards, a method is proposed for creating a register of hazards with a two-stage process: forming a model of the source data and decomposing it into blocks of the production environment, the production process and the labor process, followed by a matrix analysis. To neutralize hazards, a method is proposed to reduce the identification time and neutralize dangerous factors by increasing the efficiency of identification systems and combining organizational, medical and technical measures. As a result of the research, a model of occupational safety management in construction was developed, based on the calculation of the safety index through the assessment of the time of identification and neutralization of hazards and a combination of organizational, medical and technical measures.