This is the sixth of eight chapters on "Crime Travel Demand Modeling" from the user manual for CrimeStat IV, a spatial statistics package that can analyze crime incident location data.
This chapter, "Crime Network Assignment," explains the assignment of predicted trips to particular routes. The predicted trips are those that are either obtained from the trip distribution stage of the model or from the mode split stage. When predicted from the trip distribution stage, all trips from each origin zone to each destination zone are assigned to a particular travel route, usually on the assumption that they all travel with the same mode (usually walking, biking, or driving). When predicted trips are from the mode split stage, the predicted trips from each origin-destination zone pair by specific travel modes are assigned to a particular route for the predicted mode (routes appropriate for a bus, train, car, bike, or walking). Once the trips are assigned to routes, several statistics can be calculated. First, the predicted route from an origin zone to a destination zone can be displayed. Thus, police could increase their patrol on high crime routes. Second, the entire trip load on road segments can be calculated, since many crime trips pass over the same network segments (e.g., freeways, and major arterial roads), the total number of predicted trips on individual segments can be predicted. The result is a map of offenders' most heavily traveled segments of the network. This chapter discusses the theoretical background for this stage, the features of route networks, algorithms related to this stage, and the details of the CrimeStat network assignment module. The chapter concludes with discussions of the modeling of network assignment by crime type and uses for the network assignment of crime. Examples are from Baltimore County and Baltimore City (Maryland). 28 figures that include maps and computer screen displays, 12 references, and an attached paper on modeling bank robbery trips in Baltimore County