A Decision Support System for predictive police patrolling

• A DSS for designing patrol districts based on forecasted crime risk is presented.
• The DSS has been defined in collaboration with the Spanish National Police Corps.
• The system is comprised of a predictive policing and a police districting model.
• Crime risk forecast based on exponential smoothing is reliable.
• According to practitioners, model configurations are better than the current ones.

In the current economic climate, many police agencies have reduced resources, especially personnel, with a consequential increase in workload and deterioration in public safety. A Decision Support System (DSS) can help to optimize effective use of the scarce human resources available. In this paper we present a DSS that merges predictive policing capabilities with a patrolling districting model, for the design of predictive patrolling areas. The proposed DSS, developed in close collaboration with the Spanish National Police Corps (SNPC), defines partitions of the territory under the jurisdiction of a district that are efficient and balanced at the same time, according to the preferences of a decision maker. To analyze the crime records provided by the SNPC, a methodology for the description of spatially and temporally indeterminate crime events has been developed. The DSS has been tested with a case study in the Central District of Madrid. The results of the experiments show that the proposed DSS clearly outperforms the patrolling area definitions currently in use by the SNPC. To compare the solutions in terms of efficiency loss, we discuss how to build an operational envelope for the problem considered, which can be used to identify the range of performances associated with different patrolling strategies.