Robert K. Bratt, Joseph R. Lake, Jr., and Theresa Whistler
In August 1994, the Criminal Division of the Department of Justice initiated an outreach program to test the viability of using its Geographic Information System (GIS) to assist law enforcement agencies surrounding the Washington, D.C. area. Two pilot-project sites were chosen to participate in the program: the Warrenton Police Department in Virginia, and the Montgomery County Police Department headquartered in Rockville, Maryland. The focus of the projects is to measure how effectively GIS boosts criminal analysis capabilities and law enforcement performance. Both projects entail a needs assessment, system design, application development, software and hardware installations, database development, and ArcInfo and ArcView 2.0 training. This presentation will introduce the scopes of work, systems designed for each site, the applications for customization and automation, how GIS is used in criminal analysis, and the significance of GIS to both the departments and their communities.
The current administration has developed several community initiatives to deal with the problems of violent crime in America. One such initiative is the Community Oriented Police Services (COPS) program. It was developed by the Department of Justice under the Violent Crime Control and Law Enforcement Act of 1994 to increase deployment of law enforcement officers devoted to community policing on the streets and rural routes in this nation. Under this national program, the federal government plans to put 100,000 officers out on the street over a six-year period at a total cost of $8.8 billion. The program provides awards to sites of all sizes, including large urban centers, mid- size cities, and small rural towns. The philosophy driving the COPS program is that an increase in the number of officers on the street should lead to an increase in the number of arrests. Another complimentary approach to enhance law enforcement is the introduction and implementation of advanced technologies, such as GIS. The Department of Justice's recent Drug Market Analysis Program (DMAP) serves as an excellent example of the value of GIS to law enforcement. Between 1989 - 1992, the Office of Justice Programs (OJP) in conjunction with the National Institute of Justice (NIJ) funded a GIS pilot project to map and analyze drug market activity in five large cities (Hartford, Connecticut; Jersey City, New Jersey; Kansas City, Missouri; Pittsburgh, Pennsylvania; and San Diego, California). The data netted by this study enabled city police to locate drug markets and specific hot spots, address gun problems, track gang areas, and target areas for specific programs, such as Weed and Seed. The Criminal Division believes that the kind of success demonstrated by the GIS-based DMAP project in large urban centers can also be achieved at the local level. In order to prove the potential advantages of GIS in local law enforcement, we wanted to develop several real-life models in which our belief could be tested and critically evaluated. The Criminal Division initiated a community outreach program for law enforcement agencies near the Washington D.C. area to test the viability of using GIS to assist in their enforcement efforts. The purpose of the outreach program is to study the effectiveness of advanced technologies to boost criminal analysis capabilities, to facilitate the sharing of data--within the department and between the police and the community--and to enhance law enforcement performance at the local level.
Several law enforcement agencies in communities ranging from small rural incorporations to major cities and diversified counties have been targeted to participate in the law enforcement GIS outreach pilot-projects. Two very dissimilar sites were chosen: the Warrenton Police Department (WPD) in Warrenton, Virginia, and the Montgomery County Police Department (MCPD) which has its headquarters in Rockville, Maryland.
The small town of Warrenton, located near the center of many famous civil war battlefields, has not historically been associated with violent crime. According to the chief of police, up until a short while ago, the most heinous crime in the town was the abandonment of a jalopy in front of an elderly woman's residence. Recently however, there have been a series of violent drive-by shootings, robberies along the highway, and the murder of a local high school athlete. This increase in violent crime is seen as a spill-over from the greater Washington D.C. communities, and has had a disturbing effect on the town of Warrenton. The Warrenton community, searching to understand the causes of growing violence in its midst, has turned to the Police Department for answers.
The Warrenton Police Department comprises one chief, one lieutenant, three sergeants, eleven officers, one administrative assistant, and one data entry clerk. Only one of the staff is a confident computer user. At the time of selection for participation in the pilot project, the Department had no criminal analysis capabilities, tabular or spatial. All reports were submitted in either written or filled-in format. Since standard reports to the city council and uniform crime reports (UCRs) to the state did not answer the questions being posed by the community, the Police Department decided it would have to implement an analysis program that would reveal and summarize trends in crime. The situation in Warrenton and its proximity to Department of Justice offices in DC made it a practical choice for one of the GIS pilot-projects.
The Montgomery County Police Department (MCPD) was chosen as the site for the second GIS implementation. The MCPD provides an outstanding laboratory in which to test the use of GIS in the field of criminal analysis because it is already using an ArcInfo application and has been investigating Automated Vehicle Location (AVL), Global Positioning Systems (GPS), GIS, and other technologies for some time. The staff of over 900 includes one chief, one deputy chief, three bureau chiefs, a special operations division, and a special investigations division.
Montgomery County is home to a large number of high-profile diplomats, elected officials, and business leaders who might easily become targets of an individual or group with special interests. It is also a very ethnically mixed community with pockets of minority populations that have at various times been targets of hate crimes. Altogether, there are five districts in the county, most centered around large urban communities: Bethesda, Rockville, Silver Spring, Wheaton, and Germantown. Four of the districts are situated along the corridor of Washington D.C.'s mass-transit subway, Metro. During the past five to ten years, the county has witnessed soaring crime rates along the corridor communities as the metro-rail expanded its services to meet needs of growing populations.
The MCPD has responded to the rising number of crimes by staffing up and by actively looking for technological alternatives. About two years ago, MCPD persuaded the county to install ArcInfo on two SUN workstations at its Rockville headquarters. But the use of ArcInfo housed on the SUNs was limited to preprogrammed applications for redistricting and deployment because no funds were available for additional training and development.
Earlier attempts to provide criminal analysts at each district with GIS capabilities failed, but the idea has not withered. Each district analyst provides daily and weekly incident data summaries to post in the officer call room and sent to the central analyst at headquarters. Although the central analyst receives the information from each district, the data are rarely appended and looked at analytically countywide. The data are never exchanged or redistributed back among the districts. The problem at MCPD is not the lack of skills for criminal analysis, but the lack of better tools (GIS) and a method for sharing data among the districts, to understand what is happening countywide. The Department has a countywide network that connects all districts, but the lack of funding and technical support hinders progress in data communications. Despite the disappointments of early attempts to implement technological advancements, the Department has persistently sought alternative sources for GIS development.
The Department of Information Systems and Technologies (DIST) is a county office that is largely responsible for technological improvements made on a countywide basis. DIST set up and maintains the network for the county that ties together the Department of Transportation, the county GIS, and the police department, including each police district office. DIST also created and maintains the GIS for the county and is the resource for all data used at MCPD excluding the crime and incident data collected by officers in the field. The Criminal Division chose the MCPD to participate in the GIS pilot project because of its great need and desire to play a more active role in finding and implementing technological solutions that could benefit the whole community.
Several meetings and interviews were held with both the Warrenton and Montgomery County Police Departments to assess the needs of each agency. Although the GIS and technological requirements of the two departments vary greatly, due to substantial differences in their size, complexity, and administrative histories, the applications and benefits of GIS for criminal analysis are very similar at both installations. Both Police Departments were posed questions aimed at determining what requirements needed to be met to ensure a successful installation of a geographic information system. In order to meet the GIS and management needs of the Police Departments, several categories of tasks and various material requirements were outlined.
At the time of the assessment, the Warrenton Police Department was outfitted with a single IBM compatible computer, one 24-pin dot matrix printer, and one outdated database application for records management. There was nothing in place to help automate record entry, management, or report generation. At another building just two blocks away, the County Sheriff's Office was making plans to open a technically advanced communications center that could accommodate police and emergency 911 dispatch calls for both the town of Warrenton and the entire county. The WPD's most immediate needs are outlined below.
Provide State-of-the-Art Hardware and Softare: Keeping up with a constantly growing number of records in any database requires a faster, larger computer than the 286 that was in place at the start of the project. The WPD also required a more sophisticated records management program to help manage the incident data more efficiently. The staff needed a reliable power backup, archival device and software, a letter-quality printer with color printing capability, user-friendly GIS and desktop mapping software, and most importantly, something that would allow them to take the GIS with them to public meetings.
Network with Communications Center: With the completion of the Fauquier County and Warrenton Joint Communications Center, the WPD's major concern was to link-up to the computer-aided dispatch system, ARGUS CAD, which also provides a complete array of programs for records management. ARGUS CAD as an online program replaces the older DOS version of records management software that WPD used on its 286 pc.
Automate Reports to Community: If the WPD could produce more quickly the standard monthly reports required by the town of Warrenton and city council, there would be time to explore the data. Each month the police chief and administrative assistant produced the reports by taking tallies from paper copies of incidents and arrest records and then writing a summary or filling out forms. The process was tedious, time consuming and greatly dreaded by the police chief. The fact that piles of paper had to be manually sorted through in order to produce summaries of each month's data underlines the WPD's need for automated generation of reports.
Automate Reports to State: The WPD is also mandated by federal and state law to file uniform crime reports (UCRs) with the state of Virginia. A recent change in the requirements calls for the more detailed submission of reports in a format for the National Incident Based Reporting System (NIBRS). A process to automate the generation of these reports in a digital format for transfer is another requirement.
Increase Data Quality: A sample of 480 incident reports spanning a one-year period was scanned and converted with an optical character recognition (OCR) program and compiled into a database for testing in the GIS. Almost 20% of those records did not have addresses that could be matched with the address database to create point locations. A number of crimes occurred along a particular stretch of highway within the town's jurisdiction, and often the only location given was the highway route number. To increase the accuracy and content of crime records included in the geographic database, the WPD had to find a way to record all incident information for geocoding or coordinate assignment.
Provide Tools for Analysis of Criminal Activity: The WPD eagerly anticipated the chance to implement GIS for the purpose of analyzing crime in Warrenton, so it could more effectively deploy its staff and make a greater difference in the war on crime. With GIS to display the events of particular crimes by type, time of day, suspect, method of operation, etc., certain trends or hot spots could be watched and information leading to arrests could be shared with the County Sheriff's office or neighboring towns.
Several MCPD and county offices were visited, specific applications identified, existing procedures reviewed, and existing or potential data sources and procedural alternatives were examined. The GIS needs assessment involved on-site visits to MCPD headquarters, the five districts, the Special Operations Division (SOD), the Montgomery County Department of Transportation (MCDOT), and the Montgomery County Department of Information Systems and Telecommunications (MCDIST). It included interviews with staff involved in the records management process from the "ground-up": police on-the-beat, clerical and data personnel, criminal analysts, and Technical Services Division staff members. These interviews and meetings provided the information necessary to produce an insightful and comprehensive summary of existing capabilities and future requirements for the GIS design.
Findings of the needs assessment indicate an enormity of administrative, management and field procedures that had to be modified in order to implement GIS effectively countywide. The primary needs for GIS at MCPD are as follows.
Provide GIS Capabilities to Criminal Analysts: The MCPD purchased and installed several copies of a desktop mapping product at each of the five districts two years ago. Although the criminal analysts were given general software instruction, the GIS was never put to use because the analysts were facing a high learning curve and needed more application-specific training to get started. As a result, the analysts continued their jobs using the tabular approach and did not produce a single map or graphical representation of crime data. The solution to their problem requires the installation of a menu-driven, user-friendly GIS program; the development of very specific applications to query and represent results of criminal analysis; and the provision of task-specific training that details the procedures to accomplish repetitive daily and weekly GIS tasks.
Initiate Data Sharing Among Districts: Although the MCPD is connected to the countywide network and its own local network, none of the analysts in each district was employing the network to transfer data or communicate with each other. Staff at remote sites from headquarters in Rockville were never given procedural training for the use and maintenance of the network software, IBM LanServer. Weekly database files compiled by the district analysts were transferred on 3.5" disks to the central analyst. There was no exchange of data among district analysts; the central analyst was the only person to see all reports, which usually remained in separate formats and were rarely compiled into one database for countywide analysis. Understandably, the MCPD was seeking a way to distribute and make available all analyzed data to all appropriate personnel.
Initiate Data Sharing with the Community: In response to the requests for more information from local government entities and the business community, the MCPD was looking for a way to present its data as clearly and meaningfully as possible. There was a growing need to address audiences with diversified interests. At the same time, the MCPD realized the need to cooperate with county organizations that could exchange or use data with the Police Department. For instance, the MCDOT provides traffic and accident reports as well as information on traffic flow and trends; the county GIS at DIST provides a complete resource of demographic, infrastructure, natural resource, and parcel mapping including digital orthophotos and other images to anyone requesting them. The MCPD was not taking advantage of these resources even though a countywide network was in place.
Automate Incident Reporting Procedures: The MCPD does have a computer aided dispatch (CAD) system; however, it does not produce an accessible account or record of incidents. The calls received and recorded by the CAD are loaded onto 9-track tapes and archived. The CAD program does not initiate a digital crime record that can be viewed, selected, or extracted for any purpose unless the archive tape is loaded again and accessed through individual expertise. The CAD is completely non-interactive with the records management unit that manages and keeps all incident reports. The criminal analysts were experiencing backlogs of more than one week and could not get their hands on current data unless they keyed-in the data themselves. Consequently, the MCPD was tremendously interested in finding methods of automating field reporting and record management.
Provide Tactical Support: One of the key interests of MCPD is the technological advancement of the SOD's SWAT team. The Department has been investigating several technologies to aid tactical maneuvers; in addition, the Department keeps an indexed threat assessment handbook with information on sites determined to be vulnerable. Although the handbook is helpful, it is neither digital nor does it provide spatial information other than addresses. One of the specific tasks identified by the MCPD was the creation of a digital, spatial database of all potentially threatened sites within the county, including VIP and diplomat residences, school clusters, census tracts, political boundaries, hazardous material storage areas, religious and administrative edifices, elevations, public safety and health facilities, and emergency routing scenarios for disaster planning.
Support Technical Division/GIS Administrative Staff: The duties of the few Technical Division staff at MCPD are so greatly concentrated on maintaining the records management unit that there is no time to train or assist others outside the unit. The analysts need LanServer network and file transfer training, and the officer and central analyst need UNIX and Workstation ArcInfo training. The MCPD was looking to the Department of Justice to provide some of these training services.
The systems designed for the WPD and MCPD consist of hardware, GIS software, data, program applications, and services based on the findings of the needs assessments conducted by the Criminal Division's GIS Staff. The following designs outline specific materials and services that were or are to be provided by the Criminal Division.
The general expectations of using GIS at local enforcement sites are summarized by the goals outlined below; reaching these goals is the focus of the two outreach programs at Warrenton and Montgomery County:
Initially the value of GIS will be demonstrated by enhancing the collection and reporting of data regarding the incidences and categories of criminal activity within the WPD and MCPD jurisdictions. Through geographic analysis of these data, hot spots and patterns of criminal activity can emerge. Identification and analysis of times, locations, frequencies, and similarities in methods of operation will indicate areas of trafficking or vulnerability. This information can then be applied to staffing and deployment of police officers, resulting in more proactive responses to criminal activity.
Areas of secondary impacts are expected to be just as valuable and visible. These secondary impacts include more effective long-range financial and human resource planning, and development and distribution of public information to draw attention to legitimate safeguards for protecting the local populace and inhibiting criminal behavior.
The study of GIS within the Warrenton Police Department and the Montgomery County Police Department will span two and a half years, with a final report being issued in the latter part of fiscal year 1997. Interim progress reports, issued three months after system installation and again twelve months later, will monitor the continued viability of the two models. The final reports will present anecdotal information describing specific incidents or cases gathered from interviews with the police chiefs and staff; statistical summaries regarding data management, reports, and incidents; and examples of thematic maps used in criminal analysis. If project evaluation reports are positive, these reports can be used to generate greater understanding for and support of GIS as a technological resource for local law enforcement. Publication of such documents within the Department of Justice will help to promote and stimulate interest in similar cooperative projects.
It is still too early to evaluate the cumulative progress each of the Departments is making with GIS, the two sites are in the early phases of the project implementation. However, we are already convinced that GIS can be a powerful management information tool within the law enforcement arena. Utilizing the Warrenton and the Montgomery County Police Departments as evaluation models, we believe that we can demonstrate this in a concrete way. We are optimistic that the final reports will indicate that GIS can positively influence both Departments' abilities to understand the parameters of criminal activity in their jurisdictions, establish priorities for public safety, and focus resources effectively to respond in the best interests of their communities.
The level of the Criminal Division's GIS staff involvement with WPD and MCPD demonstrates a commitment to the application of new technologies as an effective alternative to the manpower-only solutions at most law enforcement agencies. The Department of Justice will continue to promote implementation of GIS as a long- term solution to combat crime.
Joseph R. Lake, Jr.
Associate Executive Officer
Criminal Division
U.S. Dept. of Justice
10th and Constitution, NW
Washington, DC 20530
Tel: (202) 616-0762
Theresa Whistler
GIS Project Manager
Criminal Division
U.S. Dept. of Justice
1400 New York Ave., NW
Washington, DC 20005
Tel: (202) 616-2956