Workforce health is one of the most important fundamental prerequisites for productivity and plays a crucial role in achieving sustainable socio-economic development. Therefore, preventing accidents in work environments, as a public necessity, has always attracted the attention of specialists. Given the key role of unsafe behaviors in the occurrence of accidents, some researchers have focused on monitoring unsafe behaviors using statistical process control (SPC) charts. However, human behavior is influenced by various factors such as environmental conditions and individual characteristics. These factors increase process variance and are often referred to as common causes of variation. In such processes, identifying the most influential factors contributing to variance can significantly improve process performance. Nevertheless, such sources of variability have not been adequately considered in previous related studies. The main objective of this paper is to model and monitor different levels contributing to variance in individuals’ unsafe behaviors. The proposed model was implemented in a manufacturing company. After determining the variance associated with each factor, the variability of each component was analyzed separately. The proposed approach, which demonstrates a high capability in reducing unsafe behaviors, can be used as an effective tool for safety management in organizations.