Geographic Information Systems Applications in Transit (2004)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

5 CHAPTER TWO GEOGRAPHIC INFORMATION SYSTEMS TECHNOLOGY DEVELOPMENT viewer website from TriMet (Portland, Oregon). The bus stops and bus routes are overlaid on an aerial photograph of the area. The data integration and management is pro- vided by GIS, which inventories and spatially indexes the bus stops, routes, and images. As this example shows, the power of GIS is now available through the Internet to the general public and is no longer the domain of the special- ist. OVERVIEW The GIS is one of the most innovative advances in the study of geography. Since its development in the 1970s, GIS has had a major impact on geographic analysis and on business practice in government and the private sector. Most transportation agencies now use GIS and Geospatial Information Systems for Transportation (GIS-T) is one of the largest users of GIS technology. The significant innova- tion that GIS provides is the ability to manage data spa- tially in layers and then overlay these layers to perform spa- tial analyses (1). Therefore, a roads layer can be integrated with a land use layer enabling a buffer analysis of the land uses within a given distance of the road. The capabilities of GIS have improved over the past three decades, and GIS now provide a wide range of tools for data management and analysis. In the early 1990s, GIS added specific tools for linear data management of transportation data that has proved to be extremely successful among transportation organizations (2). These capabilities enable transit agencies to georeference their bus routes, stops, timepoints, and other features to a digital street centerline file, and keep all these data in synch. Figure 1 shows an example of a transit There have been examples of the use of GIS in transit from the early days of GIS deployment in the 1980s (3). Some of these projects involved home-grown GIS products that were developed to take advantage of personal com- puters and advances in computer graphics (4). However, by the early 1990s, the GIS market consolidated around a small group of GIS vendors including ESRI, Inc. (Envi- ronmental Systems Research Institute), Intergraph, Inc., MapInfo Inc., and Caliper Corp. (Note: These represent the firms known to the author at the time of this research. Any error or omission is unintentional and no endorsement of these firms is implied.) Generally, the GIS vendors provide the GIS software and rely on third party developers or the users themselves to develop transit tools and applications. Successful examples of transit GIS toolboxes developed on GIS platforms were found in the literature review and case studies. Many of these are located outside the United States, which demonstrates the success of GIS as a univer- sal technology. Examples include the TOP (Transit Opera- tions Toolbox) program developed in Copenhagen, Den- mark; the ROMANSE (Road Management System for Europe) GIS-enabled trip planning system implemented in Southampton, United Kingdom, as part of a European Un- ion-funded transportation infrastructure project; and the In- tegrated Transportation Management System in Singapore that includes GIS-based transit operations and planning. Also in Europe an economic interest group called TRUST (TRANSMODEL Users Support Team) has developed TRANSMODEL as a reference data model for public trans- port. Supported by the French government, TRANSMODEL was accepted in 1997 by CEN (the European Standards In- stitute) as an experimental standard. TRANSMODEL pro- vides a detailed data model of public transport functions and transit data, but it does not include spatial data. TRANSMODEL has been linked to GIS in a test site in Salzburg, Austria, using the Geographic Data File (GDF) format (see section later in this chapter on Standards Initia- tives). The small number of U.S. examples is documented in the literature review and case studies that follow. FIGURE 1 Web–GIS transit data viewer (TriMet).

6 The growing popularity of GIS has attracted the interest of transit software vendors who provide scheduling, vehicle tracking, and trip itinerary planning programs. In some cases, these vendors have developed their own mapping in- terfaces with GIS-type functionality. In other cases, they provide import and export programs to convert data into compatible GIS formats. These developments reflect in part the demands from the customers for mapping inter- faces, but also that spatial data are needed in scheduling. This raises issues for transit agencies in how they integrate proprietary spatial data formats with their GIS programs. These proprietary formats have influenced the way that GIS is used in transit. Indeed, one of the issues that emerged from this synthesis project was the incompatibil- ity between traditional transit programs and the newer GIS products. This issue also interfaces with transit data stan- dards development and the institutional development of GIS programs vis-à-vis traditional centers of transit opera- tions (e.g., scheduling) within the organization. These is- sues are discussed in more detail later in this report. They are mentioned here because of the way they affect transit GIS practice and, as apparent in the literature review, his- torically provide something of a fault line in GIS applica- tions development. Although the barriers and tensions this presents are real, there are successful examples of how GIS and other programs have been integrated, as evident in some of the case studies. The GIS vendors have also historically been somewhat reluctant to implement open standards for their products. More recently they have moved toward more open stan- dards and most of the leading vendors are now members of the Open GIS Consortium (OGC), which promotes open GIS standards, including the Web Feature Standard (WFS) and Map Feature Standard (MFS). As web-based services become more prominent, mapping services over the web using XML (Extensible Markup Language) or GML (Geo- graphic Markup Language), SOAP (Simple Object Access Protocol), Java2EE, and similar protocols that can work with WFS and MFS may open up a new era in GIS devel- opment. Already there are web–GIS applications that are independent of specific GIS formats, which have the po- tential to assist transit agencies, especially those that can- not afford a full GIS implementation. For example, it is now feasible to store geometry as a spatial object in some database management systems (DBMS), which replaces one of the key functions of GIS. These trends are important because, too often, GIS developments are evaluated from a vendor-specific viewpoint rather than considering the broader domain of GIS and IT capabilities to support tran- sit business processes. If the next generation of GIS fol- lows the trend toward web-based services, then services like mapping, geocoding, and transit analysis tools may be available on-line by means of a transit Internet Service Provider, thus weakening the dependency on specific GIS platforms. (Note: This is not a prediction nor is it being ad- vocated, but it reflects trends already occurring in the wider IT community that offer more choice as to how software and data are managed and delivered to the user.) Another interesting trend has been the convergence be- tween geospatial technologies comprising GIS, GPS (global positioning system), and remote sensing technolo- gies such as satellite images, LIDAR (Light Detection and Ranging), and products that orthorectify remote sensed data. This convergence is occurring in part because of IT compatibility and the overlap and complementarities be- tween the technologies. Many users prefer the term “geo- spatial” to “geographic” information systems for these rea- sons. Within the academic community, GIS is seen as a technology application within the realm of geoinformation science. In the broader IT community, GIS is often referred to as geospatial information technologies (GIT rather than GIS), which has a harder IT edge to it; and sometimes as geomatics, which denotes geospatial data and processes as well as the technology. These terms may be interchange- able and somewhat duplicative and reflect the particular perspective of a community of interest. Nevertheless, their growing use signifies IT convergence and, at the same time, diversification in the use of geospatial technologies. Users are beginning to mix and match technologies to meet specific requirements. The challenge for GIS, especially the GIS vendors, has been to keep up with these demands and provide a one-stop-fits-all GIS package. In doing so they have inexorably become more “IS” and less “G”; hence, the distancing of some academics from the technol- ogy products and the return to fundamentals of “GIscience.” As with all software tools there is a perceived danger in re- lying on the tool versus the analytical ability of the user. This raises concerns in some quarters, especially the re- search community, and could be an interesting topic for further research. GEOGRAPHIC INFORMATION SYSTEMS BUSINESS ORGANIZATION Broadly speaking, GIS has been applied at three levels within transit organizations. At the first level, when the technology is first introduced into the organization, its ap- plication is project based in areas like ridership analysis or bus stop inventory. At level two, the GIS technology ma- tures in the organization and becomes more widely used as a departmental resource, supporting a broader range of functions in business areas such as route planning. Finally, at level three, it becomes a mainstream enterprise system that is part of the agency’s IT architecture. Although some users have progressed through all three levels, the majority of transit GIS users are still at levels one or two. The

7 TABLE 1 L EVELS OF GIS BUSINESS ORGANIZATION GIS Application Level Business Model Transit Archetype Organization GIS Programs Staffing Project Ad hoc opportunistic implementation, not a cost center or budget area. Focus on specific short-term transit activities Small- to medium-sized agency with no central IT/GIS support Management of base layers for transit data. Map-based query and display via standard GIS tools. Ad hoc desktop applications for geocoding, bus stop, and transit data analysis; simple demographic nalyses a 1–2 self-taught GIS specialists and a small number of other ad hoc users Departmental Part of department’s business plan. GIS budget. May serve as a GIS service center to other departments Medium-to-large transit agency with GIS unit or specialists within planning or operations departments Broader use of GIS in the above areas plus programs to develop more sophisticated applications and tools for transit planning and operations 3–5 GIS specialists with regular users within the department and from other business units. Some training in GIS may be part of the business plan Enterprise Part of agency’s IT infrastructure. Corporate planning and budgeting. Corporate service center (even if located within a department) Large transit agency with R&D programs, planning and operations that utilize GIS. GIS provided by IT or specialist GIS unit GIS data management and applications development integration with other transit software and information systems 5 or more GIS/IT specialists with broad range of skills from GIS programming to GIS/IT systems integration. GIS staff supported by other corporate IT resources. Large number of users across the agency Notes: IT = information technology; R&D = research and development. chances of progressing to level three are greater in a larger agency with more resources, but size is not the only factor that influences GIS development. Table 1 summarizes the GIS development levels and correlates these with the gen- eral business models that determine the role of GIS in the agency. Although no two transit agencies are alike, GIS ar- chetypes can be generated. GEOGRAPHIC INFORMATION SYSTEMS RESOURCES FOR TRANSIT There are some key resources to assist transit agencies in implementing GIS programs and a number of case studies to provide guidance on the do’s and don’ts for transit prac- tice. Vendors Research publications often cite information provided by GIS vendors, such as case study examples of transit appli- cations. The vendors can also provide information and con- tacts at their success sites. In some cases, they provide White Papers and data models as templates for transit GIS services, offering advice on implementation steps as well as indications of cost, technical resources, and data sources. They also provide training and consultant services directly or through their business partners. Further infor- mation is available from the following sources: • • • • ESRI: GIS for Public Transit Management—Case study examples are available on the ESRI website: http://www.esri.com/industries/transport/transit.html and in a publication on GIS-T (5). ESRI has also published the data architecture for their transporta- tion data model, called UNETRANS (Unified Trans- portation and Network System). Intergraph—See website http://imgs.intergraph.com/ transportation, where a copy of their Geotrans White Paper is available. Caliper TransCAD Transit Analysis—TransCAD GIS has specific extensions for public transit. (See website http://www.caliper.com/tcovu.htm#Transit%20Analysis.) MapInfo, Inc.—See website http://www.mapinfo. com, where transit examples are shown. The spread of GIS across the nation means that GIS technical support services are available locally from a number of vendors and GIS consulting firms. In addition, GIS user groups that can offer support and advice on GIS issues have been formed in many regions. Generally, the level of technical resources available through these channels is high and reflects the maturity of the market for GIS services. These resources, however, come with a price tag and for smaller agencies just starting out in GIS the costs and level of effort to develop a

8 GIS program can appear daunting. For these reasons, many transit agencies have created GIS programs that use a mix- ture of in-house and contractor resources. There is an informal GIS-T community comprising GIS users, vendors, and academics, which meets at conferences such as the Annual GIS-T Symposium, those sponsored by TRB, vendor conferences, and others. There are also pro- fessional groupings around specific issues such as stan- dards development (described later) and transit research. There is no formal transit GIS group or forum. Transit GIS is therefore poorly represented among the GIS-T commu- nity, and one of the challenges for transit is raising its pro- file to ensure that it gets the attention it deserves and the GIS technology it seeks. There is much going on in the transit GIS arena, as evident in the findings of this synthesis, but few beyond the transit GIS community are aware of it. Universities There are several universities that provide courses and re- search in transit GIS including • • • • • University of South Florida, Center for Urban Trans- portation Research (CUTR), which has an active transit GIS research program and was one of the sponsors of the National GIS in Transit conference (the third conference was last held at CUTR in 1999). Oakley GeoGraphics Laboratory, Bridgewater State College, Massachusetts, which maintains the National Transit Database (NTD) in GIS formats and in 1992 and 1993 performed surveys of GIS use in transit. The National Transit Institute at Rutgers University, which has conducted GIS research and workshops. Some of the universities that are members of the Na- tional Center for Geographic Information and Analy- sis have conducted transit GIS research, although most of its focus is on highways and aviation Other universities that have undertaken significant research include the Massachusetts Institute of Tech- nology, Iowa State University, University of Wiscon- sin, Portland State University, and the University of Illinois at Chicago. Universities have created many techniques for GIS in transportation including the widely referenced NCHRP Report 460: Guidelines for the Implementation of Multi- modal Transportation Location Referencing Systems (6). They have cooperative research programs and internships that can benefit transit agencies. FTA The FTA has sponsored a number of initiatives in GIS. The National Transit GIS is a representative inventory of the country’s public transit assets. Creation of this national sys- tem is an ongoing and collaborative effort on the part of many within the transportation industry. Use of these tran- sit data can facilitate the exchange of information within the U.S. Department of Transportation (U.S.DOT) and throughout the transit industry. This effort supports the mission requirements of the U.S.DOT, particularly as estab- lished by the Intermodal Surface Transportation Efficiency Act of 1991 (ISTEA). Most notably, this includes the de- velopment of a GIS-based National Transportation System for transit routes as a major element of the National Spatial Database Infrastructure (NSDI). This spatially referenced database will provide transit planning and operations data such as population served, ridership, passenger miles, and route/rail miles for all modes of public transit. The systems and facilities include rural and urban bus systems, com- muter rail, subways, light rail, people mover systems, high- occupancy vehicle systems, ferry terminals, and transit ways. The Standards, Guidelines and Recommended Practices establishes a framework for maintaining the NTD, ensuring data integrity, interoperability, and consistency. The meth- ods and quality control used in creating, storing, exchang- ing, and documenting the data in the National Transit GIS is known by recommending feature type definitions, for- mats, file formats, update procedures, and other standards. The document outlines feature type definitions and de- scriptions, addressing and street naming conventions, fea- ture type automation and conversion guidelines, transfer formats, and update and maintenance procedures. In 1992 and 1993, the FTA sponsored the GIS in transit surveys conducted by Bridgewater State College, where the spatial data sets for the NTD are compiled and managed (see http://www.fta.dot.gov/library/technology/GIS/ntgistds/ NTGISTDS.HTM). Transit GIS publications can also be found on the Bureau of Transportation Statistics and the FHWA GIS websites http://www.bts.gov/gis/index.html and http://www.gis.fhwa.dot.gov/. TCRP TCRP previously prepared TCRP Report 60: Using Geo- graphic Information Systems for Welfare to Work Transpor- tation Planning and Service Delivery (7). The objective of that research was to develop a handbook providing guid- ance on the use of GIS for Welfare to Work transportation planning and service delivery. The handbook includes a brief review of current practices and recommended model approaches for applications of GIS to Welfare to Work. Supplementing the handbook is a CD-ROM that provides graphic examples of the program. The case studies in- cluded on the CD-ROM provide examples of the capabili- ties of GIS and are an excellent resource for the wider tran-

9 sit community, not just those involved in Welfare to Work programs. Standards Initiatives The FTA is one of a number of organizations that has been actively involved in developing data standards for the tran- sit industry, including geospatial data standards. A driving factor behind this has been the Geospatial One-Stop (GOS) program initiated by the federal government. Other factors include the recognition from within the industry that in- compatible standards result in inefficiency, duplication of data and applications, and unnecessary redundancy. Con- sequently, a number of industry forums have developed that, along with the GOS, have contributed to the develop- ment of transit data standards. The most important of these are the Transit Communications Information Profile (TCIP) (8), Bus Stop Inventory Best Practices and Recommended Procedures (9), Location Referencing Guidebook (10), and in- telligent transportation systems (ITS) program. GOS The GOS initiative is a federal e-government initiative de- signed to expedite the creation of seven framework layers, one of which is transportation (11). The framework layer for transportation is being developed under the auspices of the NSDI project following guidelines laid down by the Federal Geographic Data Committee. In support of the trans- portation framework layer, a data content standard for transit has been created by a committee of experts known as the Transit Modeling Advisory Team coordinated by the BTS. It is being pilot tested in a few places. Transit agencies have some concerns about the incompatibility of the National Transportation Communication for ITS Profile (NTCIP) and GOS. The BTS has taken the lead in developing the transpor- tation GOS standard. The primary purpose of the standard is to support the exchange of transportation data related to transit systems. In doing so, it aims to establish a common baseline for content of transit databases for public agencies and private enterprises. The content will be organized in metadata formats that will be supported by the vendors and user communities. For example, the Federal Geographic Data Committee has de- veloped a metadata template for spatial data that at least one GIS vendor has incorporated into its GIS program. Bene- fits of adopting the standard include reducing the cost of acquiring and exchanging data; improvements in the geo- spatial transportation base data; improved integration of safety, emergency response, and enforcement data; and streamlined maintenance procedures. At the time of this report, the standard had been submit- ted to the American National Standards Institute for review and comment. It complies with related standards developed by the International Standards Organization (ISO), specifi- cally the Technical Committee on Geographic Informa- tion/Geomatics (TC 211), which produced ISO 19133 Tracking and Navigation Draft International Standard. The standard can be implemented using a variety of software packages and is designed to accommodate data encoded with- out geometry, as well as support the exchange of data encoded in a variety of GIS. It is also designed to be able to depict the complete transit system at all levels of service and all func- tional classes that may be defined by the transit agency. Thus, it provides a comprehensive set of transit features including bus stops, routes, patterns, segments, timepoints, fares, land- marks, facilities, amenities, block, trip, and geographic fea- tures. These features or entities of the transit system are re- lated to one another in a transit system data model that describes the data content or base attributes of each feature. As indicated, the transit standard appears to include most if not all types of transit data that are managed by transit agencies and outlines the relationship of these to the spatial data; that is, how the transit data can be referenced to the geospatial data to create a framework layer. However, there is concern that the standard will not be followed by commercial vendors, who code their applications to sup- port specific business processes, such as scheduling or trip itinerary planning. So, although the content may be com- plete, how it is implemented may vary across applications and between agencies. How problematic this will be re- mains to be seen. According to the data model, features are logically related to one another based on real-world prac- tice and should be robust enough to accommodate variants. Even so, there are transit programs that do not connect fea- tures logically or omit relationships that seem peripheral to their specific application. In these cases, the transit stan- dard can serve as a checklist to identify gaps; but who is responsible to fix any omissions? And, are there any penal- ties for noncompliance? These are questions that need an- swering if the transit standard is to be successful. Another ISO Technical Committee, TC 204, which de- velops and reviews standards for Transportation Informa- tion and Control Systems, has created the GDF standard mentioned in the Introduction. GDF is currently a pub- lished standard by ISO awaiting final review and comment prior to formal adoption, which could come as early as 2005. GDF is a detailed geospatial data standard for trans- portation including transit. It was originally developed to support ITS navigation services and has been incorporated by map vendors and some GIS vendors in their software. Therefore, it appears there are at least three standards that transit agencies need to be aware of (GDF, TC 211/GOS, and TCIP), which may affect their use of geospatial data as well as applications that use geospatial data. It is not sur- prising that confusion exists among transit operators as to which standard to adopt. There needs to be some coordina- tion among the different standards bodies and transit-

10 industry leadership to sort out some of the confusion. Part of the problem is that beyond the technical experts who participate on these standards initiative little information is known or circulated among the transit community. Most transit agencies, therefore, are unaware that these standards exist or are in development, and even fewer of them have been consulted as to their impact. Bus Stop Inventory Although not a standard per se, this is an example of an- other initiative sponsored by the FTA to provide guidelines to transit agencies. The bus stop inventory is a core data management tool for supporting planning, operations, maintenance, and marketing functions throughout an agency. It supports the deployment of advanced technology systems such as GIS, itinerary planning, APC, and auto- matic vehicle location (AVL). The guidelines describe col- lection, storage, and maintenance procedures as recom- mended by agencies and vendors who develop, implement, and use stop inventories. The Bus Stop Inventory Best Practices and Recommended Procedures report provides examples and templates on data content and design of the inventory including examples from field practice (9). TCIP TCIP is part of the NTCIP, which is a joint standardization project of AASHTO, the Institute of Transportation Engi- neers (ITE), and the National Electronic Manufacturers Association (NEMA), with funding from the FHWA. The TCIP development effort began under the auspices of the Institute of Transportation Engineers in cooperation with APTA, FTA, and FHWA. The TCIP family of recom- mended standards addresses Advanced Public Transporta- tion Systems (APTS) data interfaces, related automated transit tools, and data. The standard, NTCIP 1400, TCIP Framework Standard, also address the business require- ments of the APTS data interfaces. As the name implies, the focus of TCIP activities is on communications of transit data, such as data packets be- tween vehicles and roadside devices. With the increasing use of GPS and other location devices, the need to com- municate location along with other information is critical. Examples include vehicle tracking and bus annunciation systems. As part of this program a Location Referencing Message Specification standard has been proposed (SAE J2266) that provides a message packet for transmitting lo- cation data. At the time of this synthesis, the standard was to go forward in 2004 to reballot by the Society of Auto- motive Engineers (SAE) Advanced Traveler Information System Committee, then to the SAE ITS council for final adoption. NTCIP/TCIP has already adopted standards for defining location referencing methods, for example, for points, lines, polygons, and routes, which a GIS needs to follow if exchanging data between TCIP compliant appli- cations (see NTCIP 1405:2000, Standard on Spatial Repre- sentation Objects, Version 1.03). The ITE, AASHTO, NEMA, NTCIP Joint Committee announced in September 2004, that management of the TCIP program was being transferred to APTA. In the meantime, the existing TCIP standards have been re- scinded, effective September 30, 2004. This move appears to have been the result of APTA’s refusal to sign the NTCIP Memorandum of Understanding; preferring to pursue its own TCIP standards development. This change reflects some of the confusion and overlap in transit standards de- velopment. As a result of the transfer, APTA, which repre- sents the transit industry, has agreed to coordinate with the NTCIP and will assume intellectual property of the TCIP standards developed under the auspices of ITE/AASHTO/ NEMA. It is hoped that this will lead to clarification of standards, roles, and organizational responsibilities. Best Practices for Using Geographic Data in Transit: A Location Referencing Guidebook Sponsored under a cooperative agreement with the FTA, this guidebook was developed at the request of the transit industry. It provides best practices for both transit manag- ers and technical staff with respect to planning, implement- ing, and using geographic data in transit. The guidebook discusses issues and best practices for defining and using geographic locations of bus stops, routes and other map data that are needed for successfully implementing ITS and GIS, as well as for obtaining operational efficiencies. The first phase of the project involved a feasibility study to as- sess transit needs and available standards. The second phase focused on producing the guidebook to summarize and synthesize standards for using GIS and location refer- encing. Published in October 2003, the guidebook provides a comprehensive overview, and in some areas a detailed description, of existing standards and practices (10). It in- cludes 10 technical appendices and a detailed glossary. Readers who wish to review GIS and transit in a single publication should refer to this document, which when published could become the “text book” for transit GIS implementation, including guidance on how to move from a project-level GIS to a department- and enterprise-level implementation. This synthesis uses some of the same in- formation and information sources as that report. ITS for Transit ITS comprise a range of advanced technologies that collec- tively aim to improve the safety and performance of trans-

11 portation. Transit is one of the core areas of ITS. Much of the focus on ITS is in the arena of standards development, such as NTCIP, and the development of regional architec- tures that are mandated by 2005. Failure to create a re- gional ITS architecture may jeopardize federal funding of ITS projects. The following brief discussion summarizes the transit elements of the transit ITS architecture. A full description is available on the ITS architecture website: http://itsarch.iteris.com/itsarch/. The National ITS Architecture provides a common framework for planning, defining, and integrating ITS. It is a mature product that reflects the contributions of a broad cross section of the ITS community (transportation practi- tioners, systems engineers, system developers, technology specialists, consultants, etc.). The architecture defines • • • The functions (e.g., gathering traffic information or requesting a route) that are required for ITS. The physical entities or subsystems where these func- tions reside (e.g., the field or the vehicle). The information flows and data flows that connect these functions and physical subsystems together into an integrated system. Table 2 lists the transit functions and associated subsys- tems within the ITS architecture. The geospatial data man- agement and map update services are part of the archived data subsystem. TABLE 2 I TS TRANSIT USER SERVICES Function Subsystem Public Transportation Management Public transportation management En-route transit information Personalized public transit Public travel security Information Management Archived data function (including geospatial data and map update) To fully maximize the potential of ITS technologies, system design solutions must be compatible at the system interface level to share data; provide coordinated, area- wide integrated operations; and support interoperable equipment and services where appropriate. The National ITS Architecture provides this overall guidance to ensure system, product, and service compatibility and interopera- bility, without limiting the design options of the stake- holder (Figure 2). The ITS architecture illustrates at a high level how the transit functions and subsystems—travelers, vehicles, transit management, and safety—are linked to- gether by means of the communications protocols. In deployments, the character of a subsystem deploy- ment is determined by the specific equipment packages chosen. For example, a municipal deployment of a Transit Management Subsystem may select Trip Itinerary Planning and Vehicle Scheduling equipment packages, whereas a state traffic management center may select Trip Itinerary Planning and Automatic Vehicle Location packages. In ad- dition, subsystems may be deployed individually or in aggregations or combinations that will vary by geography and time based on local deployment choices. A traffic management center may include a Trip Planning Subsys- tem, Transit Information Provider Subsystem, and Emer- gency Management Subsystem, all within one building, whereas another traffic management center may concen- trate only on the management of traffic with the Traffic Management Subsystem. Although GIS may not be one of the subsystems or identified elements of the ITS architecture, Map Update Provider is a recognized key process for managing loca- tion-based information and location-based services for transit operators and their customers, and it feeds a number of subsystems and services such as AVL. GIS can play an important role in implementing ITS, especially where location data needs to be exchanged between the different subsys- tems or where the subsystems need to share a common base map and location referencing system. ITS provide something of an umbrella framework for many of the other standards development discussed previously above. TRANSIT GEOGRAPHIC INFORMATION SYSTEMS APPLICATIONS The literature review discovered more than 130 publica- tions on GIS in transit. Table 3 summarizes these by appli- cation categories. Not surprisingly, some publications cover more than one category; however, as far as possible they are assigned to the category that is the primary focus of the publi- cation. Furthermore, some articles are not entirely focused on transit but include transit alongside other transportation modes. For these reasons, the transit orientation of the ar- ticle is also recorded as being high, medium, or low. High means it is focused entirely on transit, medium indicates equal consideration of transit alongside other modes, and low includes transit, but not prominently. The discussion that follows will focus on those publica- tions with a high transit orientation. An annotated bibliog- raphy of some of these publications is provided following the references. Not surprisingly, the categories with the most publications are planning and IT, which between them account for 77% of the literature. For each category, exam- ples of transit applications in the three levels—project, de- partmental, and enterprise—are provided where available. These examples are representative of the types of uses of GIS among transit agencies and give a flavor of the wide range of applications that are capable with GIS.

12 FIGURE 2 ITS high-level architecture diagram. TABLE 3 TRANSIT GIS PUBLICATIONS BY CATEGORY Category No. of Publications Transit Orientation (1 = high, 2 = medium, 3 = low) Planninga 59 (46%) 1: 35 2: 15 3: 9 Information Technologyb 40 (31%) 1: 20 2: 12 3: 8 Operationsc 13 (10%) 1: 7 2: 3 3: 3 Managementd 12 (9%) 1: 6 2: 1 3: 5 Customer Servicee 5 (4%) 1: 5 2: 0 3: 0 aPlanning, including route and facility planning, automatic passenger counting systems, ridership reporting, demographic analysis, and modeling tools. bInformation technology, including hardware, software, custom tools, and standards. cOperations, including vehicle and facility maintenance, vehicle location, routing and scheduling, and real-time traffic information. dManagement, including safety, security and incident response, system performance and reporting, asset management, and finance. eCustomer service, including trip itinerary planning, customer relations, real-time customer information, public information, and marketing. Planning Planning, including route and facility planning, APC sys- tems, ridership reporting, demographic analysis, and mod- eling tools has traditionally been a strong area for GIS applications and includes the foundation infrastructure for GIS such as the base map, transit network, bus stop, and route inventories. There are examples of GIS applications in all sizes of transit agencies, from small rural to large ur- ban operators. Typical of the rural applications is the Shen- andoah Valley Public Mobility Project (Virginia), which is using GIS to coordinate the transportation of human ser- vice organizations (12). They have been assisted by the Smart Travel Lab at the University of Virginia to use GIS to visualize the current transportation services provided by these agencies and to look for possible overlaps. The routes that agencies run on a regular basis are represented in Arc- View by line segments. These routes are then layered on a

13 map of the Lord Fairfax Planning District and some simple analysis is done on the routes, such as buffering. One pos- sible outcome is a web-based system that contains all the routes and allows a user to search for the best route for a potential client between two points. This project-level ap- plication demonstrates the usefulness of GIS in compiling and visualizing transit information, and also shows the benefit of getting support from another organization with GIS experience. In this case, it was a local university, but elsewhere transit agencies have often sought support from their colleagues in another local agency. There are several other examples of how GIS is being used to provide mapping and analysis tools. In Corpus Christi, Texas, the Regional Transit Authority and Texas A&M University–Corpus Christi have developed a GIS that includes the street maps for the three county service regions, the route system, and the bus stop locations. These maps are used together with U.S. Census Bureau block and block group information to perform communication, analy- sis, planning, and service assurance. A GPS could also be used to support AVL and data collection (13). At the state- wide level, GIS has been used to compile information on transit services and for evaluating levels of service for planning purposes. Maryland and Florida are two states that have created statewide transit databases. The Maryland Transit Administration developed their database to support the NTD program mentioned earlier (14). In Florida, they created a statewide transit GIS for a Transit Technical As- sistance Program to local systems (15). Many of these sys- tems are limited in their ability to hire experienced GIS transportation professionals. The technical assistance will enhance the work performed by the agency’s GIS profes- sionals and will introduce transit planners to the potential uses of GIS. At the department level there are examples of the broader use of GIS to support many agency functions. GRTC Transit, the public transportation agency serving Richmond, Virginia, and Chesterfield County, Virginia, has created a GIS to improve its route planning process and to track assets (16). The system has helped the agency adjust its routes to serve the rapidly growing population in central Virginia and to keep its asset inventory current. GIS and GPS capabilities allow for new perspectives on the plan- ning process for transit applications and the analytical tools that such technology provides. GIS technology is therefore helping to integrate decision making at GRTC. This is a good example of the development and implementation of GIS technology in a mid-sized transit agency. At the enterprise level, the Metropolitan Transportation Commission (MTC) in Oakland, California, is building a regional transit information system (RTIS) and regional transit database (RTD) for the San Francisco Bay area (17). The primary objective of developing this enterprise archi- tecture is to not only foster cooperation and information exchange and interoperability among transit operators, but to also provide the public with more comprehensive and easy access to the transit information. The San Francisco Bay area comprises 9 counties and 100 cities, with a com- bined population of more than 6 million, and it is served by 26 different transit operators, including the metro rail sys- tem—Bay Area Rapid Transit. A primary objective of this RTIS is to provide comprehensive and accurate transit in- formation to the user in the most efficient manner. A key element of this architecture is an RTD that will be a reposi- tory for all transit data and related spatial information for all regional applications. GIS technology is a core founda- tion for the RTIS and the Take Transit trip itinerary planner both for managing the geographic and transit data and pro- viding it over the Internet. Other examples of enterprise use of GIS can be found in King County Metro (KCM; see the case study in chapter five) and in publications on the Orange County Transporta- tion Authority (OCTA; California) (18), and the Tri-County Metropolitan Transportation District of Oregon (TriMet; Portland, Oregon) (19). These transit agencies were among the first to use GIS and have gradually expanded the scope of GIS within their respective agencies. TriMet is another of the case studies included in this synthesis (see chap- ter five). OCTA has pioneered a number of innovative pro- jects with GIS and in 1997 won an innovation award from the American Planning Association for its detailed analysis of ridership patterns. These types of analyses have helped OCTA adjust its bus service patterns to more accurately re- flect customer needs. The FTA and FHWA recognized OCTA’s pioneering use of GIS in transit planning and made a case study of the agency in 2001. There are also publications on how to create an enter- prise GIS-T. Ford and Widner (20), for instance, analyze ways of partnering to share transportation data among state and county governments, using Virginia as a case study. They define four levels of data partnership, from informal arrangements to cooperative agreements. They note that enterprise approaches provide the most comprehensive datasets capable of supporting multiple applications, but require formal agreements on the data model, formats, road and transit definitions, attribute sources, accuracy, and data security. This is quite a shopping list and needs the support of management as well as technical users if it is to be successful. Attanucci and Halvorsen (21) review the ca- pabilities provided by GIS and describe, by example, how a number of transit agencies are currently using these pro- grams. In addition, a hypothetical comprehensive GIS is envisioned to show how a service and operations planning unit can take full advantage of today’s GIS features. The re- sources required to establish a transit GIS are also dis- cussed and a candid assessment of various obstacles to es- tablishing a full-featured transit GIS is made.

14 IT IT, encompassing hardware, software, custom tools, and standards, includes articles on how to implement GIS from a technical perspective. Within this category are many technical research papers from universities as well as documents and reports on GIS standards. At the project level, there are several publications that describe the GIS technology needed to address specific problems. Kratz- schmar and Zhou (22) describe one implementation of the infrastructure needed to facilitate the sharing of geographic information between data providers and service providers, using real-time bus locations as an example of using the Internet to deliver geographic content to the user’s browser and desktop. Web-based transit information systems are the subject of several research projects such as the develop- ment of a GIS architecture for a transit website in Montreal and Internet GIS approaches to transit information design (23–25). There is also some interesting research on GIS- based algorithms for transit scheduling and trip itinerary planning (12,26). There are several examples of GIS technology projects at the department level. The Utah Transit Authority has been making GIS and transit ITS technology work together (27). The integration of GIS analytical tools and transit ITS technologies has provided opportunities for changes in transit system design. The results have improved service and changed the political climate surrounding the devel- opment of transit services. The San Diego Association of Governments has created Estops, an on-line GIS-based transit stop inventory maintenance tool that is shared by agencies throughout the region (28). Previously, multiple inventories were used to maintain the same stop informa- tion. This not only increases the effort, but also increases the chance of error by increasing the redundancy of data. To reduce the stop inventory maintenance effort and cen- tralize the stop inventory database, the San Diego Associa- tion of Governments initiated this project to let all transit operators maintain their own stop inventory data in a cen- tralized database by means of a secured website. At the enterprise level, there are fewer examples of GIS systems architectures that integrate GIS with the agencies’ information systems. In the chapter on case studies, some of the latest examples are described. Because the enterprise approach to transit GIS is very recent the publications are few and focus on a high-level overview with only one or two practical demonstration projects. The RTD project at the MTC is one of the leading examples (29). In this pro- ject, the MTC is implementing a new architecture for the data integration and data management including spatial data. The information is being used for trip itinerary plan- ning as well as for building databases on bus stops, routes, and other transit features for planning and customer infor- mation applications. A unique feature of this project is the customization of a trip planning system to operate in a GIS-compatible environment. This is possible because the trip planning software—TranStar—was acquired from the Southern California Association of Governments rather than a commercial vendor. The MTC owns the source code and can therefore amend the program to embed GIS in its trip planning system, not simply link to an external GIS program. TriMet has also been using GPS technology to collect and maintain transportation data in collaboration with the agency’s GIS program (30). TriMet has recently increased the positional accuracy of 8,000 stop locations along bus routes and light rail. Digital images were also gathered along with stop amenity information and these are currently maintained using GPS. All TriMet buses are cur- rently equipped with GPS units that capture more than 600,000 observations of data on a daily basis. On-board computers and AVL equipment link with the bus dispatch center to improve communications with the bus operators, locate buses in real-time, and enhance data collection. Automatic passenger counters capture information while the GPS date and time stamps the record along with the XY coordinate location. This information is now routinely ag- gregated for passenger census at stop and route level. A centralized database allows users to access current data dy- namically for spatial and temporal analysis. For example, internal customers have access to real-time bus location information, and external customers can use the Transit Tracker application, which predicts the next arrival time of a bus, through the web or at selected bus stops. Finally, there are some reviews of systems integration issues in transit agencies (31). Transit agencies use many transportation accessory packages such as paratransit, scheduling, trip planning, and ride sharing and carpool software. There is a need to analyze data from these appli- cations in a GIS, but importing the proprietary data for- mats into GIS or other application packages can often be difficult and time consuming. Although many vendors of- fer packages of modules that are designed to meet all the needs of an agency, the reality is that no one set of applica- tions can satisfy the needs of an entire organization. To in- tegrate these separate but essential systems it is necessary to develop a cost-effective method to plan for the integra- tion of these products. Examination of integration issues in both DOT and transit agencies indicates the need for strategies that improve data access and reduce application maintenance costs. Operations As GIS technology has matured, it has moved beyond the display of static data to link with programs that are at the core of a transit agencies’ function, namely scheduling and vehicle monitoring. From the perspective of the transit op- erator, a basic unit of analysis is the bus trip. Although

15 varying in format, each transit property maintains a com- plete list of bus trips for each of the several regular sched- ules it accommodates (e.g., weekday, Saturday, Sunday, and holiday). This temporal aspect makes it more complex. Real-time adjustments to these assignments in response to unanticipated events (e.g., bus breakdowns and nonrecur- rent congestion) are performed by supervisors in the field and, increasingly, by means of centralized AVL and control systems. Longer-term and systemwide assessments of op- erational performance require the collection of vast amounts of data that must be organized to reflect complex temporal (e.g., morning peak, base, and evening peak) and spatial (e.g., in-bound by route or corridor) patterns. Automated data collection techniques are slowly replacing manual methods in this area as well. One consequence of automated data collection is the added volume of informa- tion in need of processing and analysis. Therefore, bus op- erations analysis is a data-intensive area that can be sub- stantially enhanced by spatial visualization in a GIS framework. Most of these applications occur at the enterprise level, but there are some approaches that include project-level and department-level perspectives. An example of the pro- ject approach is the application of GIS to the monitoring of bus operations. Bus operations analysis is a data-intensive area that can be substantially enhanced by the topological overlaying capabilities and spatial visualization afforded by a well-designed and methodically developed GIS frame- work. One study reviewed the major types of data usually collected by bus properties and the typical uses to which these data are put, identified spatial and attribute data or- ganization requirements that are of particular relevance to bus network structures, and developed a prototype GIS ap- plication to the monitoring of schedule adherence (32). There are examples in the literature of paratransit and rideshare services using GIS to geocode passenger trip ori- gins and destinations to try and match up the passengers to vehicles or to other passengers for ridematching. The de- velopment of on-line rideshare matching to provide con- venient choices for the commuter and reduced operating costs to transit operations are being developed for the Vir- ginia Department of Rail and Public Transit (33). The ap- plication uses web and GIS technologies. This agency pro- vides support to 15 local rideshare and transportation on demand management agencies through grant programs by conducting research, and by providing training and com- munications and marketing assistance. The program was built on GIS without which it would have been difficult to implement. At the enterprise level, some of the larger transit agen- cies have implemented AVL systems as part of a compre- hensive transportation management program. These sys- tems include GPS, CAD (computer-aided dispatch), and GIS, together with the communications devices. They en- able real-time monitoring of transit vehicle locations and are used to manage incidents, bus bunching, and other op- erational issues. Some of the case studies cited in chap- ter four are implementing these systems. Because of their complex systems integration, these systems are costly and mainly provided by large IT companies. An excellent re- view of the state of the art of these systems is provided by Casey et al. in Advanced Public Transportation Systems: The State of the Art Update 2000 (34). The report reviews AVL systems in Portland, Oregon; Essex County, New Jer- sey; Chicago, Illinois; Baltimore, Maryland; and Roches- ter, Pennsylvania, including their use of GIS. (A related re- port, Advanced Public Transportation Systems Deployment in the United States, contains a complete survey of AVL systems in use nationally in 1999) (35). Management This area includes safety, security, incident response, sys- tem performance and reporting, asset management, and fi- nance. It is the area of interest to managers and is also the area with the least focus on technical issues and details. There are examples of specific project-level applications of GIS especially in safety, incident response, and asset man- agement. In King County, Washington, for example, the transit division tracks security-related incidents on its tran- sit system. Its older systems were unable to combine these data with spatial analyses crucial in deploying security re- sources to the needed areas in a timely manner. A GIS was implemented, along with other DBMS, and proved effec- tive in supplying information needed by transit authorities and security personnel to decrease security incidents. The combined application was versatile enough to also assist other departments of the transit division (36). Kurt et al. (37) has described how GIS can be used to develop an in- tegrated asset management system for rural and small ur- ban transit. There are only a few publications on how to use GIS to monitor performance or provide decision-support tools for policy analysis. The FTA has used GIS in preparing its An- nual Performance Plans and GIS is promoted as an innova- tive technology to assist transit agencies improve planning and service delivery. Some research papers address the cur- rent state of transportation planning as related to GIS us- age. They seek to answer the question “How can geospatial data technology and GIScience contribute to improving our transportation system?” One example uses scenarios in transportation planning, including perspectives from a state DOT, a metropolitan planning organization (MPO), a tran- sit administration, and a small state-funded nonprofit han- dling ride-share information. It then addresses current models and considerations and goes on to outline future policy considerations (38).

16 et. Managers are very interested in the costs and benefits of GIS programs. There are few examples of this in the litera- ture; however, the Transportation Case Studies in GIS sponsored by the FHWA between 1999 and 2001, pre- sented an evaluation methodology and produced statistics to demonstrate the tremendous benefits derived from a well-thought-out and supported GIS program (39). Gener- ally, however, beyond the examples cited, there are few guidelines or reports that address management of GIS in transit agencies. There is a need for more studies in this area. For instance, managers need to know what the re- source requirements for an effective GIS program are and how to measure the benefits to the agency. Customer Service Among the very few publications on customer service uses of GIS in transit are two that highlight GIS applications in trip itinerary planning (40,41). Customers who inquire about transit service benefit from having maps of transit routes or walking directions to the bus stop, and there are some examples of these services being provided through the Internet. Customers also value real-time information on bus status. To provide transit information with GIS analysis functionality on the World Wide Web requires a system ar- chitecture that integrates web serving, GIS processing, and database management. It also requires an efficient path- finding algorithm to handle the unique features of the tran- sit network; for example, time-dependent services and multiple service routes serving the same stre At the project level there are descriptions of system architectures that link the web-based graphic user interface, the Web server, and the GIS server. A GIS server is composed of three distributed components including map server, transit network analysis, and relational DBMS. Some of these publications are the result of research rather than actual implementations; however, there are some re- ports that provide information and guidelines on how to implement GIS and trip itinerary planning systems. A 2002 state-of-the-practice review of trip planning systems in- cludes some implementation examples of GIS (42). These demonstrate both department- and enterprise-level imple- mentations of customer service with GIS. Another emerging area for GIS use is in the marketing of transit services. Zhou provides an example of how to apply GIS to a market segmentation analysis of transit rid- ers (43). GIS can also be used to plot customer responses to surveys on service quality and to analyze origin and destination data for improving services and identifying new markets. SUMMARY This literature review demonstrates the variety of uses of GIS in transit planning, operations, information systems, management, and customer service. It is evident that the use of GIS in transit is growing and the technology is now mature enough to be considered a core technology in tran- sit service delivery. GIS applications are moving beyond the traditional areas of planning and information systems, into operations, management, and customer service. The public is becoming more accustomed to on-line maps, and transit programs that do not include maps lack visual ap- peal. Even so, the growing use of GIS in transit is not merely cosmetic; customers are demanding real-time in- formation for trip planning and trip reliability. To remain competitive with other modes real-time data are important. There have been significant advances in GIS technol- ogy, in its user friendliness and capabilities to link to non- GIS programs such as scheduling, trip itinerary planning, and AVL. GIS enables these other technologies and can present the information to the public in an understandable visual manner. From an operational perspective there are benefits in being able to track and monitor transit assets. One of the fundamental applications of GIS is in managing the inventory of transit routes, stops, and schedule informa- tion together with the underlying street network. These elements can be synchronized in GIS so that if changes oc- cur in any one of them the others will simultaneously be updated. This ability to synchronize the geospatial data with transit data is a major benefit of GIS. Any data with a georeference can be added to the GIS, thus expanding the scope and scale of the GIS. As revealed in the literature re- view, the scalability and flexibility of GIS use in transit is one of its most important features. Recognizing the value of GIS in transit, the federal government and industry groups have been promoting the use of GIS technology and have created standards and guidelines for its implementation. A number of partner- ships have been formed between government, industry, and interest groups to provide advice and support to transit agencies implementing GIS. There are plenty of reasons, therefore, to take advantage of GIS, as the next chap- ter illustrates.