Abstract
 Probability Grid Method for Crime Analysis in ArcGIS 9 Track: Law Enforcement and Criminal Justice Author(s): Richard Heimann, Bryan Hill The Office of the Chief Technology Officer, DC GIS group, has developed a comprehensive and streamlined predictive workflow within the ArcGIS geoprocessing framework called the Probability Grid Method (PGM). With multiple versions of the model, we have achieved a scalable tool that permits crime analysts to perform predictive modeling at all Esri license levels. With variables set as model parameters, this is a complex yet customizable process that yields optimal results. The tool has multiple output statistical tables in addition to the final "probability grid" that is derived from multiple combined and weighed datasets. PGM is a methodology that has been in the industry for nearly a decade and has yielded a remarkably high-accuracy rate for predicting next crimes in a series. This has been a collaborative effort to bridge the gap between proven PGM methods and proven spatial statistics. Richard Heimann DC GIS OCTO 441 4th Street NW Suite 930s Washington , DC 20001 US Phone: (254) 285-7200 E-mail: richard.heiman@dc.gov Bryan Hill Glendale Police Department Crime Analysis Unit 6835 N 57th Dr Glendale , AZ 85301 US Phone: (623) 930-3073 E-mail: bhill@glendaleaz.com |
|||||
