ABSTRACT
Chapter 13 furthers the endeavor of simulation to model individual spatiotemporal trajectories. The agent-based model (ABM) is used to represent discrete places with unique geographic features, heterogeneous population of autonomous agents, and dynamic interactions between an agent and places and among agents themselves. Simulated agents (various groups of individuals) guided by defined rules (routines) move in space (a study area) over time, and their convergence in both space and time predicts the likelihood for an event to take place. In a way, an agent-based model mimics human movements in real life. The case study of robberies in Baton Rouge examines how criminogenic features of local places and daily routines of vulnerable targets and motivated offenders affect the spatiotemporal crime pattern. The ABM crime simulation model first tests some hypotheses suggested by the routine activity theory and the environmental criminology theory. How well simulated robberies from this model predict hotspots in reported crimes sets a benchmark on the theory’s usefulness in applied crime research. The model then tests the impact or effectiveness of various police patrol strategies, such as random vs. hotspot policing. The simulated crime pattern by the model largely replicates prominent robbery hotspots in the study area.
