The broader impact/commercial potential of this Small Business Innovation Research (SBIR) Phase I project is in the field of predictive policing, with broader applications to crime and terrorism-fighting efforts. The product will be the first software to assimilate real-time crime data and provide dynamic model-based forecasting of policing resources necessary to fight criminal activity in specific locations within a city. Over a third of local police agencies in the US with 100 or more officers have implemented a program similar to Compstat (crime information and record management software), which shows their interest in computer-aided policing. The proposed software will be an independent product, but compatible with such current police software tools. Expansions for anti-terrorist applications are envisioned. This Small Business Innovation Research Phase I project will address the development of a gang activity agent-based model, from which a software tool for law enforcement decision making - "Computer Aided Policing" (CAP) - will be developed. An agent-based model of gang activity, which will include individual behavior of an agent or a group in interaction with other agents or groups, as well as an individual's reaction to regulatory, economic and other societal pressures, will be developed based on criminology theory and practical experience. Data management and data mining techniques will be used for development and validation of the agent-based model. Using data assimilation, real-time observations will be incorporated into the model. Tools that enable rapid and automated visualization and analysis of simulation results and the optimization of force deployment for crime mitigation will be developed. This model-based development will be first-of-a-kind in computational social science and will go well beyond the current state of the art in predictive policing.
