Figure 1: Pilot Project Area
The City of Dallas is experiencing economic and population growth that places an increased demand on its transportation infrastructure. The City’s Transportation Engineering Division is attempting to utilize GIS as one of the tools to manage existing transportation resources and enhance the efficiency of transportation operations.
This paper discusses the Transportation Engineering Division’s GIS pilot project that was implemented to test the feasibility and effectiveness of a GIS. The objective of the GIS program is to build a transportation-related spatial database and GIS applications. Data from a variety of sources is integrated into one database accessible by the transportation staff through the use of ArcView. Data on traffic volumes, speed limits, location of traffic accidents, characteristics of major thoroughfares, location of signalized intersections and school zones is processed and made available to the staff. A combination of data sets is used to conduct spatial analyses. Future enhancements of the system include geocoding information from the citizen request database (requests for traffic signals, signs, and pavement markings) and information from the street cut permit database.
The City of Dallas has been operating a citywide GIS system for the past several years through the GIS Division of the Planning and Development Department. As a result of the Planning Department’s efforts, other departmental leaders began to see the potential benefits of having a GIS system for their own departments and began to provide funding for GIS implementation.
The overall mission statement and guiding principles of the Transportation GIS are the same as the citywide GIS. However, the goals of the Transportation GIS are more division-specific. The long-term goals include using GIS as a tool in transportation operation and planning, providing accurate and timely spatial information to the transportation staff, and publishing transportation data on the Internet and Intranet.
The Transportation Engineering Division was among the first groups in the City to begin implementation of a GIS. Due to limited funding available for the GIS project, a pilot project was started. The pilot project allowed the division GIS analyst and administrative leaders to estimate the total cost of implementing a citywide Transportation GIS.
Work on the pilot project started in December 1996. Several focus group meetings were conducted and a comprehensive survey of transportation staff GIS needs was performed. Then a data inventory was begun and data sets for the pilot project were initially developed. The goals of the pilot project were:
Figure 1 shows the city area and pilot project area.
Figure 1: Pilot Project Area
As a result of group discussions and surveys completed by the employees, the following table was generated showing what information should be made available in the Transportation GIS. Table 1 also shows which section of the Transportation Division would be responsible for supplying a particular data set to the GIS analyst and the format of the existing data source.
|
Section Responsible for Data Set |
GIS Data Set |
Existing Data Source |
|
District Engineering |
|
|
|
Signal Engineering |
|
|
|
Street Lighting |
|
|
|
Transportation Planning |
|
|
|
Field Operations |
|
|
|
Interagency Coordination |
|
|
Table 1: Results of Data Inventory
The GIS Division of the Planning Department develops and maintains a comprehensive spatial database of the City. The database includes the following data sets: city limits, census tracts and blocks, council districts, service districts, parcels, street centerlines, hydrography, locations of schools, fire stations, police stations, digital orthophotographs of the City, etc. All of these data sets are available to the transportation staff and can be used for analyses together with the transportation data sets.
Most of the information used in the Transportation Division deals with streets; therefore, the street centerline file was used as a base for developing all transportation-related data sets. The GIS Division of the Planning Department is responsible for the maintenance of the City’s street centerline file (adding new streets as they are built, changing street names, etc).
The street centerline coverage contains all the streets within the city limits with attributes such as street name, type, prefix, suffix, and address range. However, most of the attribute information in the Transportation Division is collected for major streets only. From the street centerline coverage, the major streets were selected and a new coverage, which is used as the transportation base map, was created. Having a subset of all the streets speeds up the display and analysis process.
Unfortunately, by separating the two coverages, a maintenance burden was added – changes in the street centerline coverage have to be manually added to the transportation base map. The added maintenance should not be an issue since major streets do not change often. Most changes affect smaller residential streets, which are not included in the transportation base map. It seems that the reduction in coverage size is worth the extra work.
Any additional data collected in the Transportation Division is based on and linked to either the street centerline coverage or the transportation base map. The major challenge was to find ways to link the transportation attribute data with the two coverages. Additional information is linked to arcs and nodes by using unique identifiers. When a new street centerline coverage replaces the old one, the added information is not lost.
The Table 2 shows the data sets that were chosen to be included in the pilot project:
|
Data Set |
Coverage |
Feature Type |
|
Major streets |
Transportation Base Map |
Arc |
|
Signalized intersections |
Transportation Base Map |
Node |
|
Traffic volume |
Transportation Base Map |
Attribute |
|
Speed limits |
Transportation Base Map |
Attribute |
|
Thoroughfare plan |
Thoroughfare Plan |
Arc, Route |
|
Bike routes |
Street centerline |
Route |
|
Bus routes |
Street centerline |
Route |
|
Road humps |
Road Humps |
Arc |
|
School zones |
School Zones |
Arc |
|
Traffic accidents |
Traffic Accidents |
Point |
Table 2: Pilot Project Data Sets
Each section of the Division is responsible for collection and update of the attribute data related to its work. Many transportation-related databases were developed over the years (e.g. traffic count book, street inventory database, school zone flasher database, or traffic signal database), while others had to be converted from a hardcopy into electronic format (e.g. school zone extent information and speed limit information).
It was necessary to establish clearly defined guidelines for the entry of data into the databases. Standard data-entry procedures include a list and description of fields that are required for linking attributes with their spatial representation. Every record in the database must have a unique identifier that is being used to link the attribute information with a feature in the coverage. The common field in the attribute table and coverage allows the user to join one data table to another and perform analyses on the new combined set. Initially, the GIS analyst is responsible for creating spatial representation of the records in the database. For example, it is GIS analyst’s responsibility to create a coverage of school zones from the street centerline coverage and the school zone flasher database supplied by the Field Operations section. However, the responsible section is still required to maintain its own non-spatial data.
By making each section responsible for its own data, the attribute maintenance burden is deferred from the GIS analyst. This approach also ensures better accuracy of the data sets. An employee associated with each of the data sets is responsible for helping the GIS analyst to gather and update that particular data set. The GIS analyst is responsible for the actual GIS data conversions and updating of the existing coverages.
When a new feature is added (e.g. a new school zone is installed), the database administrator is responsible for delivering a new version of the database and identifying the new records. The GIS analyst adds the features to the appropriate coverage. The same procedure is applied for record removals. Since the attribute data is linked to coverage features by using unique identifiers, updated values can be automatically displayed on a map.
Figure 2: Data Update
Each section is provided with a PC workstation and ArcView software. All ArcView users received training and have a ‘read-only’ access to the GIS data on the LAN. Therefore, when someone in the department is in need of GIS data, that person can have the GIS trained personnel in their section perform the necessary analysis or deliver the necessary data.
Traffic count information is valuable for traffic modeling and for determining the need for improvements along roadways. Using GIS, one can display traffic counts for many different thoroughfares and compare traffic volumes. This comparative analysis can be used to choose the location for the next signal or intersection improvement project.
The transportation base map contains major city streets, which are included in the Traffic Count Book. The Traffic Count Book contains information on average daily traffic volumes, morning and afternoon peak volumes, and peak time. Traffic volume information is collected and updated regularly and is stored in an MS Access database. Each street segment in the database is described by the street name, from street, to street, and has a unique identifier, which is used to link the attribute table with the transportation base map coverage.
Figure 3 shows the information from the City Traffic Count Book on a map.
Figure 3: Average Daily Traffic Map GIS allows the users to query and compare locations and characteristics of traffic signals throughout the City. Conversely, GIS can be used to keep very detailed information about each individual signal by linking the location to other files and data sets. Traffic signals are stored as nodes in the transportation base map. Each signalized intersection has a unique identifier, which is used to link the spatial data set with attribute information. The following information is available for each signalized intersection in the pilot project: End-users are able to access any of these sources of information about the signalized intersections through ArcView. Avenue scripts have been developed that allow users to link an intersection on the map with its corresponding AutoCAD as-built drawing or digital photograph. The location of school zones and the accompanying signs, flashers, and speed zones are an important component in transportation planning and allocation of resources to maintain roadways and equipment. It is important that potential problems with school zone markings and equipment be prevented and GIS can be used to monitor conditions, such as age and date of last replacement. This information can be used to effectively schedule preventative maintenance to markings and equipment. To develop a school zone data set, information from two sources was combined: school zone flasher database and the City Ordinance. The school zone flasher database contains information on hardware type, date of installation, and update of the flashers, etc. The City Ordinance contains an exact description of the school zone extents (e.g., from 60 feet south of Allen St & Cole Ave to 60 feet south of Allen St & Howell St). The combination of these two sources allowed discrepancies between the two sources to be discovered and corrected. School zones are stored in a separate arc coverage. Street segments designated as school zones were selected from the street centerline file based on a block range. The coverage was then edited according to the school zone extent information. After the coverage was completed, a map of schools and school zones was created. The map helped visualize the information from the database and immediately identify problematic locations, e.g. schools without the school zones. Figure 4 shows locations of the schools and school zones in the pilot project area. Figure 4: Schools and School Zones in the Far North Dallas Currently, the transportation engineers are using the Traser database to access and analyze the information on traffic accidents. This proprietary database, maintained by the Dallas Police Department, contains all traffic accidents that have occurred in the City since 1992. Information from the Traser database has been downloaded and imported into MS Access. Locations of traffic accidents were then geocoded and point coverage of traffic accidents was created. Each point in the coverage represents one traffic accident. Attribute information associated with each accident includes: date, time, block number, primary street, reference street, direction and distance from the reference street, manner of collision, light conditions, weather conditions, road class, road conditions, road surface, traffic control device, number of vehicles involved in the accident, injuries, and fatalities. Figure 5 shows the intersection traffic accidents in the pilot project area. Graphic selection of the intersection of Preston Road and Frankford Road resulted in displaying 16 traffic accidents that happened there in 1996. Figure 5: 1996 Intersection Accidents The compiled GIS database is stored on an MS Windows NT server and can be accessed through ArcView. By storing the data at a centralized location, each section benefits from being able to access data from all other sections. The use of the pilot project database by the employees has been quite limited because it includes only a small area of the city, and consequently, the database cannot be used to perform any citywide analysis. Therefore, data conversion methodology designed for the pilot project has been used to begin compilation of a citywide transportation database. This database is still under development, but some of the preliminary GIS data sets have already been used in several projects. Some of the projects are: Traffic accident data for the last three years were used to generate maps of accidents involving pedestrians and bicyclists. The maps showing the accidents together with locations of schools allowed determining the proximity and frequency of these accidents for each school. The map of bicyclist accidents and bicycle routes allowed determining whether there is a need for improvement or additions to the City’s bike route system. Locations with a higher number of accidents involving pedestrians and bicyclists are then studied in more detail. The transportation engineers check the adequacy of sidewalks, streetlights, traffic signs, and markings in problematic areas. Results of these studies will be used to suggest solutions for improving road safety. To find out which capital improvement projects are located on bus routes, a coverage of capital improvement projects was created and overlaid with a bus route coverage. Dallas Area Rapid Transit (DART) uses its own GIS system and agreed to exchange GIS data with the City. DART provided a bus route coverage and an MS Access database that includes information on the number of trips per day and the number of people per day for each bus route. Since DART buses use City streets and accelerate their deterioration, this information was used to create a list of projects for which the City will try to secure additional funding from DART. The projects were prioritized based on the number of bus routes running on a particular street, number of bus trips per day and the number of people using a particular bus route. A map showing average daily traffic on City streets has helped transportation staff with their work on the Intelligent Transportation System Master Plan. The map was used to find the best locations for surveillance cameras and changeable message signs to maximize the benefits to the motoring public. Locations of these new devices are stored as a point coverage in the GIS database. GIS was also used to determine how many cameras and changeable message signs are planned for installation in each City Council District and Service District. This type of analysis was very difficult to perform before GIS was in place. GIS has proved to be a valuable tool for signal timing projects. Information on link distances between signalized intersections is easily accessible. Traffic volume data for the pilot project include: average daily traffic, morning peak volumes, afternoon peak volumes, turning movement counts, and hourly traffic volume counts, which can be used to create traffic volume diagrams. Division managers, staff members, and GIS personnel have found the Transportation GIS pilot project to be successful. The goals and objectives of the pilot project were accomplished. The most important benefits include: The size of the pilot project area was one of the limiting factors in using the compiled database for citywide analysis. The pilot project database compilation was the first step in the effort of implementing the GIS in the Transportation Division. Methods for integrating the data from a variety of sources into the GIS database were designed and tested. These methods have been used to start developing the citywide Transportation GIS database. Another problem faced during the implementation of the pilot project was the limited use of the pilot project information by the end-users. This limited use was mainly attributable to two factors: lack of knowledge of GIS capabilities and little time to use ArcView. Although several end-users did receive introductory ArcView training, they did not have time to experiment with the pilot project data. It is expected that as more citywide Transportation GIS data becomes available and customized user interfaces and applications are developed, there will be less staff training required. Customized user interfaces should also help make GIS a more mainstream application available to an average user. The users need to be regularly informed of what data sets are available. This information is kept in the data dictionary and in the metadata documentation. The data dictionary includes information on what data is available, were it is stored, how it was created, how often it is updated, when is the next scheduled update, etc. The metadata documentation contains detailed description of each data set loosely based on Federal Geographic Data Committee metadata standards. It will take more time and effort to educate the users about the GIS capabilities. Several presentations have been given to the transportation staff about what GIS data is already available and how it can be used. As the users see examples of data and spatial analysis, they might suggest other ways that GIS can help them do their job. It is also important to cooperate with other groups within the organization to avoid duplication of efforts and to take advantage of existing data. As the Transportation GIS continues to develop, more data sets will be included in the GIS database, customized applications will be developed that would allow the end-users easy retrieval of the information, and Transportation GIS data will be made available to the general public over the Internet. Some of the future data sets might include: The preventative maintenance of street signs and markings before they become public hazards is a task of the Field Operations Division. A database that contains information about the condition, age, etc., of signs and markings can be used with GIS to develop a schedule of preventative work that will help eliminate signs and markings that require repair or replacement. In the case of a signal malfunction or controller cabinet damage, field crews can access the information from the traffic signal hardware database, CAD as-built drawings and digital photographs, before going to the field. This prevents field crews from having to visit the location to determine what type of equipment is needed and then go back to the location for repairs. The street lighting inventory would allow GIS users to analyze various aspects of the more than 73,000 streetlights along Dallas streets. This would help coordinate efforts with TU Electric (the local electric company) and with citizen petition groups to better allocate City resources. Eventually, street light failures and bulb expirations could be predicted and a replacement/repair schedule developed to drastically reduce the occurrence of street light outages before they pose a threat to public safety. The service request database contains citizen requests for installation or removal of traffic signals, signs, information about signal malfunctions, and other transportation-related problems. Each request is identified by its location. This location information can be used to geocode the requests and find where there are consistent problems. After locations are identified, appropriate action can be taken to eliminate the source of the problems. Since streets closed for construction become barriers to traffic, having information from the street cut permit database in the GIS would be very helpful especially for routing emergency response vehicles. In the future, the City plans to make the Transportation GIS data sets available on the City’s Internet Web page. Access to information from other City departments would also be helpful. This data (already converted to GIS format) would be available over the citywide LAN from the respective departmental GIS such as Water Department, Police Department, Fire Department, Planning Department, and other Public Works Divisions. The use of GIS to manage information resources has many benefits, including increased access to many different data sets and types of information, powerful analysis tools, and the benefit of quickly displaying the results of complex analyses in a graphical format. These capabilities make GIS the tool for better-informed decision making. However, in order to utilize the many benefits of GIS, databases must be designed, developed, and maintained. The pilot project is a good way to start implementing a GIS if limited resources are available. As the data is developed and made available to the end-users, they will recognize the advantages of having all of the work-related information at their desktops, which will hopefully spawn further development of the GIS. 
Traffic Signals
School Zones
Traffic Accidents
IV. Application of Compiled Data in Spatial Analysis
Road Safety Improvements
Capital Improvement Project Tracking
Intelligent Transportation System Master Plan
Signal Timing Projects
V. Findings and Recommendations
Recommendations:
VI. Future Plans for Transportation GIS
Public Access to GIS Data
VII. Conclusions
Maria Durcanska
GIS Analyst, City of Dallas
Department of Public Works and Transportation
1500 Marilla Street, Room L1-A-S
Dallas, TX 75201
Phone: 214-670-4831
Fax: 214-670-3292
E-mail: mdurcans@ci.dallas.tx.us