Technological evaluation of the phase I AVM system used tests which showed that 95 percent of the vehicle location estimates to be within 625 feet of the true location, and 80 percent of the estimates to be within 90 feet of the true location. However, because of the frequency of lost vehicles, an analysis was made to determine factors responsible for diminished accuracy. These are random error; quantization in time, distance, and angle; systematic error; open-loop tracking; missed signals; and susceptibility to subversion. Many of these errors are to be corrected in the phase 2 operations of the system. Evaluation indicated a reduction in dispatch time (12 percent in district 3) and an increase in dispatcher workload. Travel time dropped by 11 to 15 percent. Overall response time dropped about 12 percent, suggesting that the police department should also concentrate on other aspects of the police response system not directly related to the AVM in order to effect greater improvements in response time. Attitudinal evaluation showed a significant drop in positive attitudes of the police toward the AVM (from 64.4 percent before using the system to 39.8 percent after phase 1 implementation). The drop can be largely attributed to problems in system technical performance and reliability. Figures, footnotes, and 10 references are included.
Evaluating a Police-implemented AVM (Automatic Vehicle Monitoring) System - The St Louis Experience (Phase 1)
NCJ Number
77771
Journal
IEEE Transactions on Vehicular Technology Volume: VT-26 Issue: 1 Dated: (February 1977) Pages: 60-70
Date Published
1977
Length
11 pages
Annotation
Focusing on technology, operations, and attitudes, this paper presents a summary of an 18-month evaluation of the phase 1 Automated Vehicle Monitoring (AVM) System in district 3 in the St. Louis Metropolitan Police Department (Missouri).
Abstract
Date Published: January 1, 1977