The social contradiction is becoming complex, and the workload of police officers is also growing while there is no significant increase in the number of police officers. The efficient optimization on the allocation of police resources has become an urgent issue for public security organs. Big data technology provides a new method for the optimization of police resources. This paper designed a potential criminal prediction and police resource optimization system based on existing allocation of police resources in the jurisdiction. Based on the data analysis of police work, case information, and public security organs, the system will predict the key locations where potential criminals exist, so as to provide scientific and optimized allocation strategies of police resources, effectively improve the police's response capability.