The Joint Forces Command (JFCOM) demonstrated the Joint National Training Capability (JNTC) as part of a special event, called the Joint Virtual Training Special Event (JVTSE), during the 2004 Interservice/Industry Training, Simulation, and Education Conference (I/ITSEC). JVTSE implemented a prototypical live-virtual-constructive (LVC) training event that was supported by participants from the government, industry, and academia. The assessment of large-scale, joint exercises is a laborious and challenging task. The current method employs human observers/trainers (O/Ts) monitoring the training audience's performance. Given the pace that these exercises are conducted and the large number of simultaneously occurring events, this assessment process is fraught with errors, including the omission of key events, and O/T bias. The need for automated performance assessment technologies to provide detailed and accurate reports of performance has been advocated (Cardinal et al., 2004; Meliza et al., 1992; Watz et al., 2003). This paper will target the design, development, and implementation of an automated performance assessment capability that; logs, analyzes, and presents data obtained from an LVC environment. It will provide information regarding the performance of the trainees as well as the training system. Additionally, this paper will describe how this system can be used to diagnose specific performance issues and make appropriate adjustments to improve human and/or system performance.