SAMPLE-based Model for Intelligent Task performance by Human operators (SMITH)

The Army has engaged in a modeling and simulation effort to develop advanced technologies and tactics, techniques, and procedures (TTPs) supporting the transformation to the network-centric warfare (NCW) concept. These activities have culminated in the Modeling Architecture for Technology, Research, and Experimentation (MATREX), a fully distributed and service-based architecture encouraging the "plug-and-play" of individual models. Within the MATREX, the Command, Control, and Communications Grid (C3Grid) models operator behaviors using a behavior engine based on the Java Expert System Shell (JESS), and has been deemed insufficient to effectively model the expanding range of behaviors, due to its brittleness and lack of maintainability. We propose to develop an enhanced behavior engine for the MATREX by applying an overarching software architecture (based on our existing SAMPLE architecture) allowing for the integration of a range of artificial intelligence (AI) technologies (fuzzy logic, Bayesian reasoning, genetic algorithms, etc.) to model decision-making more robustly and efficiently, with specific focus on enhanced usability. The SAMPLE-based Model of Intelligent Task performance by Human operators (SMITH) will be integrated into the MATREX with specific focus on its interactions with the C3Grid's C3 human performance model (C3HPM), providing an effective representation of both decision-making behavior and task performance.