This paper presents an approach for representing and analyzing random motions and hazardous events in a simulated three-dimensional workplace, providing designers and analysts with a new technique for evaluating operator-machine interaction hazards in virtual environments. Technical data in this paper is based upon a project striving to reduce workers’ risks from being hit by underground mining machinery in a confined space. The project’s methodology includes human factors design considerations, ergonomic modeling and simulation tools, laboratory validation, and collaboration with a mining equipment manufacturer. Hazardous conditions can be analyzed in virtual environments using collision detection. By simulating an operator’s random behavior and machine’s appendage velocity, researchers can accurately identify hazards, and use that information to form safe design parameters for mining equipment. Analysts must be discerning with the model and not read more from the databases than what the simulation model was designed to deliver. Simulations provided an interesting approach to data gathering in that there was no need for live subjects and logistics - test sites and costs associated with experiments-became insignificant. Collisions versus speed, operators’ size, and risk behaviors proved the versatility found in the data obtained from the model. Preliminary results show that response time significantly affects the number of collisions experienced by the virtual subject. Also simulation data suggests that more mishaps occur with hand-on-boom-arm risk behavior.