Abstract--This paper presents experimental results associated with human factors aspects of using three-dimensional (3D) visualizations to display electric power system generation information on one-line diagrams. The paper's results are based on an experiment performed at the University of Illinois at Urbana-Champaign using electric power system students. The results indicate that compared to standard 2D one-line displays, 3D visualizations of generator output and reserves can be used successfully on one-line displays to improve both the speed and accuracy of certain tasks. Index Terms-Power System Operations and Planning, Power System Visualization, 3D, Human Factors I. INTRODUCTIONPower system analysis and operations requires the consideration of a large amount of multivariate data. For example, even in a simple power flow application data of interest includes a potentially large number of independent and dependent variables, such as transmission lines flows, bus voltages, generator real/reactive outputs and reserves, transformer tap positions, flowgate values, and scheduled versus actual power transactions. With systems containing tens of thousands of buses, a key challenge is to present this data in a form so one can assess the state of the system in a quick and intuitive manner.The information associated with power systems has usually been presented using a two-dimensional (2D) display space, often consisting of either a one-line diagram or tabular list displays. However, over the last several years this pattern has begun to change as new visualization techniques are developed and integrated into both power system analysis software and utility control centers. One such technique, made possible by recent increases in computing power, is the interactive three-dimensional (3D) visualization of power system information. An early application of 3D for power system information visualization is [1] in which simple 3D graphics are used to show power system voltage security. A few years later the use of 3D is presented for plant andThe authors would like to acknowledge the support of NSF through its grants EEC 96-15792 and DMI 00-60329, PSERC, and the U.S. DOE through its Consortium for Electric Reliability Technology Solutions (CERTS) program.