In order to increase productivity and reduce labour costs, many manufacturing companies have adopted automated systems such as automated guided vehicles (AGVs) for performing material handling and transportation tasks in the production process. In the study, virtual reality is utilised to simulate the pre-defined path of AGVs, and examine the effects of operating conditions on perception of hazards in virtual and real operating conditions. A series of experiments were conducted to develop an empirical relationship with the psychophysical Stevens' power law. Based on the analytical results from the experiments, empirical models of Stevens' psychophysical power functions for perception of acceptability and hazards within a dynamic virtual work environment are developed. Using the sequential experiments technique, a comparison of perception in virtual and real operating conditions was made for female subjects (n = 20) after data bridging. Results show that speed of the AGV and distance to the AGV significantly influence the perception of hazards in virtual operating conditions.