U.S. flag

An official website of the United States government, Department of Justice.

NCJRS Virtual Library

The Virtual Library houses over 235,000 criminal justice resources, including all known OJP works.
Click here to search the NCJRS Virtual Library

Attempt To Predict the Residence of Serial Arsonists: Geographic Profiling Using Center of Gravity Model

NCJ Number
190275
Journal
Reports of the National Research Institute of Police Science Volume: 40 Issue: 1 Dated: September 1999 Pages: 23-36
Author(s)
Terumi Mimoto; Naoki Fukada
Date Published
September 1999
Length
14 pages
Annotation
In this report on two related studies, the authors proposed a geographic profiling support system as a tool for geographic analysis of crime sites and described a practical strategy for predicting the location of serial arsonists' bases.
Abstract
In the construction of a geographic profiling support system, the authors developed software that unifies and manages large amounts of data on serial crimes, termed "Power Plot Professional." Currently, the main functions of the software include file management of the plot data, display of the results of analyses by using the circle hypothesis and the center-of- gravity hypothesis, as well as the computation of basic statistics of measured data. In addition, it supports BMP format graphic files as background maps. In the geographic profiling of serial arsonists, the center-of-gravity hypothesis posits that the base location of a serial arsonist tends to be near the geographic "center of gravity" of the offense sites. Using data from 14 Japanese serial arsonists (defined as those who commit five or more arsons), the authors examined the validity of the center-of-gravity model. The arsonists were classified into three categories based on offense duration (long-term arsonists, short-term arsonists, and mass arsonists) and into two categories based on spatial behavior types (marauders and commuters). Power Plot Professional was used not only to measure the distances between each offense site and the offender's base location, but also to calculate the geographic center of gravity of each offender's offense sites. The analyses showed that the center of gravity of offense sites did not coincide with the residence of the arsonist in any of the 14 cases. In the case of marauder-type offenders, however, the results for both the long-term arson type and the short-term arson type indicated that the distance between the offenders' residences and the center of gravity of their offense sites was relatively small. This suggests that the center-of-gravity model has a potential capability of predicting the base locations of these types of offenders. On the other hand, the center-of-gravity model apparently does not apply to the commuter-type arsonists. The authors propose a practical method of prediction that uses the concept of a "suspicion area," which is defined as a round area whose radius is equal to the average distance between the center of gravity and the offense sites. The use of the center-of-gravity model with the suspicion area can be a useful method for setting priorities for the target areas of an investigation. 4 tables and 8 figures