Citation

Kent J, Leitner M, Curtis A. Comput. Environ. Urban Syst. 2006; 30(2): 181-200.

Copyright

(Copyright © 2006, Elsevier Publishing)

DOI

10.1016/j.compenvurbsys.2004.10.002

PMID

unavailable

Abstract

This research evaluates the usefulness of applying functional distance measures to criminal geographic profiles using mathematically calibrated distance decay models. Both the travel-path (i.e., shortest distance) and temporally optimized (i.e., quickest travel time) functional distance measures were calculated based on the impedance attributes stored within a linearly referenced transportation data layer of several parishes in Louisiana. Two different journey-to-crime distance decay functions (i.e., negative exponential, and truncated negative exponential) were mathematically calibrated for "best fit", based on the distribution of distances between homicide crime locations and offender's residences. Using the calibrated distance decay functions, geographic profiles were created for a localized serial killer from Baton Rouge, Louisiana. A probability score was calculated for every point within the study area to indicate the likelihood that it contained the serial offender's residence. A comparison between the predicted (highest probability score) and the actual residence of the serial offender determined the predictive value and procedural validity of functional distance metrics. 2004 Elsevier Ltd. All rights reserved.