Ground Access to Major Airports by Public Transportation (2008)

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.

The airport manager faces a wide variety of challenges in the creation of a successful ground transportation strategy, which almost certainly will comprise several separate services to respond to the needs of several separate market segments. This chapter reviews the key steps for improving public transportation access to U.S. airports and presents some information that is further devel- oped in later chapters. Six steps are in the process outlined in this chapter: 1. Establish the public policy goals for airport ground access (a theme that is further developed in Chapter 2). • Form the collaborative effort that will be needed for implementation. • Understand the travel behavior of the longer distance traveler. 2. Undertake the program for data gathering and system monitoring (a theme that is further developed in Chapter 6). • Design the survey to reveal key market characteristics. • Emphasize accurate geography and market segmentation for both air passengers and airport employees. 3. Understand the markets revealed and their relationship to candidate solutions (a theme that is further developed in Chapter 6). • Understand the composition of the overall airport market. • Establish the target markets at several levels of trip-end density. • Understand the precedents for market support of various modes and services. 4. Design a program of services and strategies for airport ground access (a theme that is further developed in Chapters 3 and 4). • Understand the quality attributes achieved by successful services. • Match modes with markets. • Acknowledge the role for dedicated, higher cost services. 5. Manage the airport to encourage rather than discourage higher occupancy use (a theme that is further developed in Chapter 7). • Examine priorities and implications of curbside allocation and pricing. • Evaluate the level of amenity experienced by the public mode user. 6. Present the ground access services to the traveler (a theme that is further developed in Chapter 9). • Provide basic service description to the users. • Develop programs for integrated passenger information and ticketing. 15 C H A P T E R 1 Six Steps in a Market-Based Strategy for Improving Airport Ground Access

Step 1: Establish the Public Policy Goals for Airport Ground Access Define the Stakeholders and Get Them to the Table In the first step of this six-step process, it is essential to establish a collaborative initiative to implement improved public transportation services for airport access. Such establishment will require locating the key players, bringing them to the table, gaining agreement on the public pol- icy goals of the proposed policies, and establishing a basic understanding of the nature of the problem being faced. This step establishes a regional context for decision making. Preparing to address airport ground access involves many stakeholders including managers of airports, operators of public transportation, operators of private transportation, managers of the roadway system, and managers of the regional transportation planning process. In addition to the transportation agencies, other organizations are critical to the improvement of public transportation access to airports. These agencies—including those with environmental approval powers, the power to change taxi regulations, and the ability to subsidize transit services designed to link workers with jobs—all have a role to play in a coordinated strategy to improve airport ground access. The early involvement of the agencies with environmental review power cannot be overstated, as results from the planning process are often integrated into key environmental documents. One transportation leader recently told Congress: “. . . we have begun to realize that no insti- tution ‘owns’ the congestion or safety problem at the local level or state level, and no institution has the right players around the table such that they could be accountable for the daily per- formance of the system.” This observation is particularly true for the subject of improved airport ground access; yet, someone has to get the right players around the table, and someone has to be accountable for the performance of the system. In some cases, leadership can be provided by a strong regional plan- ning agency, such as the Metropolitan Transportation Commission (MTC) in the San Francisco Bay Area or the Metropolitan Washington Council of Governments (MWCOG) in Washington, D.C., both of which have strong roles in airport planning. In most cases, leadership must come from the managers of the airports themselves. Coordinate with the Regional Planning Process The parties need to define the extent to which the ground access issues are regional in nature, as this will affect the number of stakeholders needed at the table. Many on-airport improvements can be managed at a very local level, but others will require a broader based coalition to deal with the issues that are clearly regional in nature. For those issues that require a multiagency response, it is critical to involve the managers of the regional planning process, usually the regional met- ropolitan planning organization (MPO). Failure to do this will result in serious problems in obtaining funding and needed environmental clearances. The Role of the Congestion Management System Within the established metropolitan transportation planning process, there are several pro- cedures that are critical for the successful integration between the project-specific activities and the regional requirements. Many metropolitan areas, particularly those with air pollution issues of non-attainment, require the creation of a Congestion Management System (CMS) by the region’s MPO. The role of the CMS is to document significant sources of congestion and low system performance and to examine a wide variety of strategic solutions to the problem, only the 16 Ground Access to Major Airports by Public Transportation Key Challenges in Step 1 • Identify the key stakeholders and get them to the table • Determine the extent to which the problem requires a regional solution • Directly involve the managers of the regional transportation planning process • Undertake early planning activities to allow for later incorporation into environmental documentation • Understand the travel demand behavior of the longer distance traveler

last of which is the addition of roadway capacity. Indeed, in areas of non-attainment, federal funding can only be used for roadway capacity increases that result from the completion of the CMS. At the very least, the managers of the airport access improvement strategy should be work- ing closely with regional managers of the CMS. At this point, the regional planning must focus on the unique demands that will be placed on the data collection and analysis process for improving public transportation access to an airport. Usually, the travel demand forecasting process used in the metropolitan planning organization is focused on the needs of the peak-hour commuting period. The existing databases may or may not be structured to deal with the needs of the longer distance traveler. Traditional forms of U.S. Census journey-to-work data will be of only limited value to the analysis of airport access. MPOs may or may not be prepared to analyze the transportation behavior patterns of the longer distance traveler, in this case the air traveler. Preparation for Major Investments In the event that the planning process may result in a major capital investment, the early planning should be undertaken in a manner consistent with the requirements of the later cre- ation of either an Environmental Impact Statement or a Finding of No Significant Impact. In either case, the rules for formal scoping and for the public participatory process must be established in the earliest phase of the planning process. In particular, the early examination and narrowing of alternatives must be undertaken consistent with the requirements of the National Environmental Policy Act, as part of a publicly visible process; lack of attention to the legal requirements of process at this point risks the invalidation of later results from court challenges. For the reasons discussed in the preceding paragraphs, clearly any major attempt at applying regional resources to improving public mode services to airports must be either initiated by the regional planning body or closely coordinated with others in the region having the statutory authority for transportation planning. The planning effort to improve public transportation serv- ices to the airport should be included in the Unified Planning Work Program approved by the MPO, regardless of whether federal funds are proposed in the planning or implementation efforts. Indeed, recent funding legislation requires that the operators of airports be members of the MPO. Design Analysis Tools for the Longer Distance Trip The tools of analysis must be applied to understand the particular travel demand behavior of the traveler taking a longer distance, multimodal, multisegment trip. From the outset, the ana- lysts need to see the problem in terms of the full trip of the traveler. The choice of a mode to or from an airport is part of a larger set of decisions made in the process of going from the door of origin to the door of destination of the full trip. It is critically important to establish early in the process that the needs of the long-distance traveler most probably will require solutions that are not simply extensions and elaborations on service concepts already provided for the metropol- itan context. The operation of traditional, low-fare, multistop street bus service to major airports may be a critically important element of a program to get workers to jobs, but such services only rarely have the ability to attract air travelers. The long-distance traveler makes logical and rational economic decisions, and those decisions are different from those made in daily commuting. The longer distance traveler is making a different set of decisions from those of the metropolitan-scale traveler. These decisions are different in terms of uncertainty and lack of knowledge about the non-home end of the trip. The decisions are different because of the amount of baggage being carried by the traveler, the trav- eler’s sense of apprehension about the reliability of the trip and arriving on time, and the total trip costs. Six Steps in a Market-Based Strategy for Improving Airport Ground Access 17

To the extent possible, those crafting new strategies to divert air travelers away from low- occupancy vehicle strategies should familiarize themselves with the experience of others around the world who have created successful airport ground access services. Chapters 2, 3, and 4 were created to help transfer the lessons learned from major airport ground access systems around the world for application by the U.S. airport manager. Best Practices in the United States: Establishing the Process The following practices are some of the many good examples of coordination with the regional transportation planning organizations that exist in the United States: • The role of the San Francisco Bay MTC in the planning of airport access improvements in the Bay Area and in continued management of the ground access surveying process. • The role of the Denver Regional Council of Governments in undertaking a comprehensive examination of ground access issues for Denver’s new airport. • The role of the MWCOG in the analysis of the implications of continued and expected airport growth, expressed in terms of projected ground access flows. • The role of the Southern California Association of Governments (SCAG) in the formulation of aviation policy in the Los Angeles region. Step 2: Undertake the Program of Data Gathering and System Monitoring In Step 2, the airport manager must create a database upon which to plan and monitor the services and facilities for improved airport access. This step is critical because the improvements to airport access must be based on a clear understanding of the market behavior of the several submarkets for airport ground access services. The airport ground access survey is the primary tool used to gain the information needed for a market-driven, traveler-oriented process. Decisions can then be made on a modally unbiased basis stemming from the analysis of the needs of the traveler. This process cannot be commenced without high-quality data describing just who those travelers are and where they are coming from. The evaluation of a given service should be examined in terms of its performance in its own logical catchment area, not in terms of mode share for an entire airport. As described in Chapter 6, it is important to establish a market description of that subset of travelers for whom the proposed service is relevant. Targeted market segments should be defined and services designed for their particular needs; success or failure of those services should be established in terms of the capture rate within the targeted market group. A specialized van service from a hos- pital complex to an airport, for example, should be evaluated on the basis of how well it attracts riders from its specified market area, not on its performance in the entire airport ground access market. For any given service under evaluation, there will be a geographic area where that ser- vice makes sense as a logical choice and a geographic area where that service makes no sense at all. The airport ground access survey is the essential backbone of the market-driven planning process. Such a survey can be expected to cost between $100,000 and $300,000. Without this information, the process of matching services to market needs cannot be undertaken. Data Collection for the Airport Ground Access Survey The application of market research methods to airport ground access, including survey procedures, is presented in detail in Chapter 6. Key issues for data collection include the exact 18 Ground Access to Major Airports by Public Transportation Key Challenges in Step 2 • Develop the data-gathering instrument • Document the geographic segmentation for the ground access trips • Document the demographic segmentation for the ground access trips • Commit to an ongoing program to monitor the performance of the system • Develop measures of performance for the airport ground access system

geographic origin of the ground access trip, time of day, the trip purpose, and the resi- dency status of the traveler. TCRP Report 62 (16) describes the use of additional market research techniques, including focus groups. A comprehensive process of market research can include both survey methods that rely on “stated preference” and methods that rely on “revealed preference.” Demographic Elements Categories of Trip Purpose. The survey must be designed to support geographic segmen- tation and demographic segmentation. The point of origin must be defined with enough clarity that it can be integrated with geographic information systems. The origin of the ground access trip can be determined by either the zip code of origin or an address specific enough to support geocoding in the data entry process. The designer of the survey must deal with a basic trade-off between the amount of data desired and the need to keep the survey short. Specific trip purposes such as medical, personal business, school, or vacation are not needed for analyses of airport access. For the airport access survey, the most important trip purpose differentiation is simply “business” versus “non-business.” Categories of Residential Status. The second element of the demographic segmentation concerns the residential status of the traveler. As documented in Chapter 2, the mode choice decision of the traveler at the non-home end of the full trip is fundamentally different than the mode choice decision in the geographic area in which the traveler resides. The level of automo- bile availability (whether for the drop-off mode or the drive–park mode) is substantially higher at the home end than at the non-home end of the trip. In addition, the level of familiarity with the details of the public transportation system is usually much lower at the non-home end of the trip. For these reasons, the survey must be designed to properly differentiate between the trav- eler commencing the ground access trip in his/her own residential area and the traveler com- mencing the trip in the non-home end of the journey. With these two elements of information, all travelers can be easily categorized into four clearly defined market segments, sometimes referred to as “the four-cell matrix.” The market research process recommended in this project requires the creation of these demographic market segments: • Resident business • Resident non-business • Non-resident business • Non-resident non-business Why Look at Separate Market Segments? These four separate market segments can be applied to a wide cross section of U.S. and European/ Asian airport ground access markets. Importantly, none of these categories can be applied as a “cookie cutter” approach to predicting behavior. The four market segments allow several subsets of the market to be observed separately. Successful strategies offer a variety of public mode services, at a variety of prices. At a given airport, a multistop bus service at less than $2 will appeal to a different market than a door-to-door shared-ride service for $15. At Baltimore/Washington International Airport during peak hours, travelers are offered multistop MARC commuter rail services to Union Station for $5 or Amtrak Acela service for more than $30. Some travelers will choose the first train out (at the higher cost), while others will wait for the lower priced rail service. Their choice is influ- enced by their demographic market segment. Danger Areas in Data Collection The designers of the survey should be aware of the particular data collection pitfalls that exist for airport access. For the analysis of traffic flow, a category called “bus/limousine/van” may be Six Steps in a Market-Based Strategy for Improving Airport Ground Access 19

a reasonable definition. However, for the analysis of public transportation patterns, it is critical to separate publicly available buses and vans from limousine service not available for shared-ride purposes. Similarly, the question “What mode do you usually take to the airport?” gets a differ- ent response from that of the preferred formation, “On your last trip to the airport—and only that trip—what mode did you take?” A survey bias towards socially desired behavior patterns occurs on the first question. Data Collection to Monitor the Performance of the System The measurement of performance of the system is a very important output from the data collection process. A classic example of a commitment to measurement exists in the contrac- tual relationship between the Massachusetts Port Authority (Massport) and the Conservation Law Foundation, a non-profit environmental organization. The simplest, and most basic, commitment is to the continual monitoring and measurement of mode share to the airport and to the volume of vehicle miles traveled (VMT) associated with airport access. For such a program, it is critical to monitor the actual vehicle volumes throughout the airport roadway system; the accurate calculation of VMT will require both traffic counts by vehicle classifica- tion and the kind of origin-destination information only made available by the Ground Access Survey. A very basic example of a system of performance measurement was developed by Massport in the mid-1980s. Table 1-1 shows the number of vehicle trips on the roadway created by one air traveler gaining access to the Boston airport on the ground system by various modes. Each of the values was calculated empirically from observed occupancy and load factors for each of the modes. In the evaluation of the performance of the system, a given strategy was considered to be beneficial if it moved the traveler to a more efficient mode (i.e., down the rows of the table) rather than to a less efficient mode (i.e., up the rows of the table). The implications of some changes in travel behavior are intuitively obvious; a new express bus service that diverts a traveler away from his/her former drive–park mode is a more efficient mode and is evaluated positively. But not all implications of mode changes are intuitively obvi- ous. If, for example, on-airport parking rates are set extremely high to discourage the use of drive–park, the implications of the resulting mode change are not so clear. If that trip is diverted to pick-up/drop-off mode, the implications for vehicle miles traveled are highly negative, and the candidate practice is evaluated negatively. The program of monitoring performance must be designed to record such subtle changes in travel behavior. 20 Ground Access to Major Airports by Public Transportation Mode Vehicle trips per air traveler trips Pick-up/Drop-off 1.29 Taxi 1.09 Drive–park 0.74 Rental car 0.69 Door-to-door van 0.33 Scheduled bus 0.10 Rapid transit 0.00 SOURCE: Massachusetts Port Authority, “Logan International Airport, Ground Access Non-Pricing Study, Second Report to the Conservation Law Foundation,” 1991. Table 1-1. Measures of effectiveness in Massport program: ground access vehicle trips per air traveler trip.

Best Practices in the United States: Continuing Survey Programs Some of the most comprehensive survey programs in operation at the world’s airports are located in the United States. The following programs are good examples of commitments to monitoring the performance of the system through surveys: • The Air Passenger Survey Program of the Port Authority of New York and New Jersey, which operates the largest and most comprehensive data-gathering program for airport ground access in the country. • The historic role of Massport, the Executive Office of Environmental Affairs, and the Conservation Law Foundation to establish a commitment to continuous monitoring of the performance of the ground access system to the Boston airport. • The airport passenger survey program of the MWCOG, which integrates air survey data into the regional transportation planning process in Washington, D.C., and parts of Maryland and Virginia. Step 3: Interpret the Markets and their Relationship to Candidate Modes The most basic question in market research for airport ground access is “Where are they com- ing from?”. Different airports have different fingerprints that identify their ground characteris- tics. Some have highly dispersed origins. Some have highly dense ground access origins. Forming an understanding of those patterns is a critical step in the development of solutions for airport ground access. Geographic information systems now allow analysts to interact with the data and create locally derived categories of trip-end density. With these tools, each analyst can develop methods of revealing natural market patterns appropriate to the needs of the analysis. Logically, analysts examining distribution patterns in Manhattan would select different breaking points for data categorization than analysts examining Denver. Geographic Scale of the Airport Ground Access Markets Some airports attract most of their patrons from a relatively compact geographic area, while others draw their patrons from vast geographic areas. The geographic scale of the airport’s catchment area provides an early indication of the nature of the density patterns to be dealt with in the development of successful ground access services. This report defines the primary market area for the airport as a whole as that area composed of zones with more than 5 airport trips per square mile, by all modes. This definition has proven to be an effective way of focus- ing attention on those areas where empirically some 70% of the airport’s ground transporta- tion customers originate. In the densely developed area served by Reagan Washington National Airport, the geographic area composed of zones with at least 5 trip ends per square mile covers only 484 square miles. In the highly suburbanized geography of Los Angeles Inter- national Airport, the area composed of zones with at least 5 trip ends per square mile spreads over 1,500 square miles. These highly aggregated observations about the overall nature of the ground transportation market can be made early in the process and reveal much about the nature of the challenge of pairing airport access services to market segments. However, to understand the ability of mar- kets to support specific services, the total airport market must be disaggregated into at least three categories of trip-end density. Six Steps in a Market-Based Strategy for Improving Airport Ground Access 21 Key Challenges in Step 3 • Determine the density charac- teristics of the overall ground access market • Define a geo- graphic area with more than 50 trip ends per square mile and under- stand the nature of the market within this area • Define a geo- graphic area with less than 5 trip ends per square mile and under- stand the nature of the market within this area • Define a geo- graphic area with between 5 and 50 trip ends and understand the nature of the market within this area • Analyze how each of the three market areas may require different kinds of services

Three Categories of Trip-End Density For the purpose of this research project, three basic categories of trip-end density have been created: • Less than 5 airport trip ends per square mile • Between 5 and 50 airport trip ends per square mile • More than 50 airport trip ends per square mile Each of the three categories has its own challenges. As will be discussed in Step 4, the empiri- cal data suggest that providing services from door to door at trip-end densities of less than 5 trip ends per square mile is extremely difficult and may result in shared-ride services producing basi- cally low-occupancy taxi services under a different name. The examination of geographic areas composed of zones with at least 50 airport trip ends per square mile provides a point of departure for further analysis concerning possible markets for traditional fixed-route and -schedule service. The existence of geographic areas with more than 50 trip ends per square mile is necessary but not sufficient to support these services. Having defined the geographic area of more than 50 trip ends per square mile, the analyst can further explore the characteristics of density within this geographic area, which vary considerably among U.S. airports. Table 1-2 ranks 10 of the 27 most transit-oriented U.S. airports in order of the portion of their ground transportation markets originating in zones with densities greater than 50 trip ends per square mile. Airports Ranked by Orientation to Areas of High Trip-End Density Fixed-route and -schedule service requires a certain density of trip ends to operate at reason- able headways. Table 1-2 shows that, of U.S. airports, only San Francisco International Airport and Reagan Washington National Airport have a majority of trip origins coming from the dens- est category, those areas with more than 50 trip ends per square mile. The use of the category “more than 50 trip ends per square mile” is a surrogate to describe the market areas most susceptible to higher occupancy public mode solutions. It is a first step in the process of identifying specific service proposals, ranging from scheduled hotel loop service (appro- priate to most large airports) to full-scale regional rail transit coverage (applicable to a small number of airports), such as Washington Metropolitan Area Transit Authority (WMATA). Whereas the first data column of Table 1-2 summarizes the extent to which an airport is ori- ented to the highest category of trip density, the second column provides more information about the trip-end density within that geographic area. This information is needed to assess the ability of the market to support fixed-route and -schedule services and can be used as an indica- tor of the potential for high-capacity service to be successful. By far the airports with the great- est concentration of trip ends are in New York with more than 400 trip ends per square mile for this analysis area. At the other extreme, the low trip-end densities for analysis areas in Los Angeles are particularly cautionary in the context of markets to support fixed-route and -schedule services throughout the defined area. Density and Market Support Associated with Specific Modes Next, the analyst should review the existing data concerning the trip-end densities that are sup- portive of various forms of airport ground access services. Looking at the existing services and market support conditions, what do we know about the correlation between trip-end density and specific modal service? What mode shares can be expected within specifically targeted geographic areas? While many factors contribute, clearly volume (and density) of trip ends are critical ele- ments in understanding the ability of specific markets to support specific modal services. 22 Ground Access to Major Airports by Public Transportation

Six Steps in a Market-Based Strategy for Improving Airport Ground Access 23 Airport Percentage of airport ground origins from zones with more than 50 trip ends per square mile Trip-end density from these zones, as trip ends per square mile Daily air travelers from these zones San Francisco 57% 225 18,000 Reagan National 52% 216 9,840 New York LaGuardia 49% 409 11,700 New York JFK 44% 310 10,450 Boston 35% 210 9,300 Los Angeles 33% 77 12,970 Washington Dulles 30% 110 4,280 Denver 29% 100 8,600 Seattle 28% 126 4,700 Tampa 25% 126 3,025 SOURCE: TCRP Report 83, MarketSense. Table 1-2. U.S. airports ranked by orientation to dense urban market. A key conclusion of TCRP Report 62 (16) and TCRP Report 83 (47) is that the overall mode share for an entire airport does not reveal the extent to which a given strategy may or may not be working; it does not provide the basis on which to analyze the performance of specific services. Rather, each candidate service needs to be examined in terms of a catchment area in which the serv- ice is a logical choice for the traveler. Using this market research technique, Chapter 6 reviews a set of specific services in the Washington, D.C., area in the context of their logical catchment area. Air Traveler Markets Supportive of Rail Services TCRP Report 83 calculated that the primary geographic market for rail services for air travelers to Boston’s airport is characterized by a density of 150 total airport trip ends per square mile. Within this logical catchment area, the Massachusetts Bay Transportation Authority (MBTA) rail services attain a mode share of 16% of the air travelers to Boston’s airport. The same analysis process has determined that the prime geographic market for rail services to Reagan Washington National Airport is characterized by a density of 125 total airport trip ends per square mile. Within this logical catchment area, WMATA rail services attain a mode share of 13% of air travelers to Reagan Washington National Airport. Air Traveler Markets Supportive of Regional Collection Points Primary geographic markets were calculated for airport express bus services from regional col- lection points serving airports in Boston, San Francisco, and Los Angeles. Support for these dedicated airport bus services was found in geographic market areas with less than 5 trip ends per square mile. The Van Nuys FlyAway bus service to Los Angeles International Airport was supported by a market area with 8 trip ends per square mile. Express bus services from regional collection points to Boston’s airport attained more than 20% mode share in their markets, while the Marin Airporter (San Francisco) captured more than 30% of its primary market area. Similar strong markets are reported from other data sources for longer distance bus and van services serving New York John F. Kennedy (JFK) and Boston airports. Air Traveler Markets Supportive of Door-to-Door Services In both Seattle and Oakland, the logical catchment areas for door-to-door van services were char- acterized by airport trip-end densities averaging about 15 trip ends per square mile. A market area

south of the San Francisco International Airport supported door-to-door van service with a trip-end density of 24 trips per square mile, while the Los Angeles primary market supported door-to-door services with an average of 27 trips ends per square mile. Door-to-door vans capture a variety of mode shares from their respective logical catchment areas. Mode shares of less than 10% are attained in Los Angeles, Boston, Seattle, and the market area south of San Francisco International Airport. Mode shares of about 20% are attained in the City of San Francisco, and in the Oakland market. While there are clearly densities below which door-to-door van services cannot be supported, they are able to serve in areas of high density. Van services operate with strong market performance in the City of San Francisco in a market area with more than 300 trip ends per square mile. Markets Supporting Exclusive Airport Buses to Downtown Examples of airport-oriented bus services from downtown hotel and major activity centers have existed in most major U.S. airports, serving a wide variety of downtown trip-end densities. While these buses serve central business district (CBD) densities as high as 500 trip ends per square mile in Boston or New York, they also serve the smallest of downtowns. As buses have consider- able flexibility in their operating patterns, this research effort has not established a lower level sup- port threshold under which services cannot operate. Advanced downtown bus services, such as the Airport Express in New Orleans, have shown exceptionally strong market capture rates. The Need for a Composite Approach The market analysis process examines the strength of specific markets to support airport ground access services and provides hints as to the modes best matched to those markets. While the details of effective market segmentation will vary from airport to airport, it is fair to say that a comprehensive strategy to deal with U.S. airport ground access must deal with at least three geographic submarkets. • A Dense Urban Market. Clearly, there is a geographic area of highest trip-end density, some portions of which may support fixed-route and -schedule services. There is no empirical evi- dence that zones with less than 50 trip ends per square mile can support such services on their own. Successful rail services have been observed in market areas of far more than 100 trip ends per square mile. Hotel loop buses serve small geographic areas, with highly compact markets: Seattle’s Gray Line Express serves a hotel-oriented concentration of more than 400 trip ends per square mile. Boston’s CBD generates more than 500 trip ends per square mile, support- ing both rail and hotel loop services. • An Exurban Market. Clearly, significant portions of the overall airport market come from large geographic areas where collection services need to be provided by means other than the vehicle providing line-haul services to the airport. Express services dedicated to the needs of the air traveler are supported by immediate market areas with trip-end densities less than 10 trip ends per square mile and provide park-and-ride availability to those coming from areas of very low trip-end density. • A Middle Market. Finally, there is a category for which upper and lower boundaries are less clear. It is the largest of the three categories for U.S. airport ground access: zones of origin generally more than 5 and less than 50 trip ends per square mile. As discussed in Step 4, this market may be the most difficult to serve. Best Practices in the United States: Examples of Market Types at U.S. Airports The wide variety of market types in the United States serves to illustrate the importance of designing a cross section of services. In the United States, airport markets cannot be characterized 24 Ground Access to Major Airports by Public Transportation

as either all exurban in nature or all urban in nature. The following airports are good illustrative examples of three types of markets: • An Airport Oriented to a Dense Urban Market. To San Francisco International Airport, the majority of trips come from areas in which airport trip ends are densely concentrated: about 18,000 air travelers come from zones with more than 50 trip ends per square mile; this area has an overall average of about 225 trip ends per square mile. San Francisco has the nation’s single largest market for airport trips from the kind of highly concentrated trip ends that can be served by a variety of fixed-route and -schedule modes, including rail. • An Airport Oriented to an Exurban Market. To Denver International Airport, more than 9,000 air travelers come from zones that have trip densities of less than 5 trips per square mile. Of the 27 most transit-oriented U.S. airports, Denver’s airport had the highest volume of “exurban” trip ends, which come from highly dispersed zones of origin. • An Airport Oriented to a Middle Market. To Los Angeles International Airport, the major- ity of airport trips come from market areas that are neither dense nor exurban in nature: about 21,000 air travelers originate in areas with less than 50 trip ends per square mile but more than 5 trip ends per square mile; this area has an overall average of about 15 trip ends per square mile. This area represents the United States’ largest market for medium-density modes, such as door-to-door vans. Step 4: Design a Program of Services and Strategies for Airport Ground Access Having established an understanding of the nature of the markets for airport access services, a ground access strategy can be developed to include a set of services appropriate to the submarkets revealed. During this step, a set of candidate modal services must be selected, determined by the needs of the travelers and by the ability of the markets to support specific services. At this point, decisions must be made between investment in rail versus bus systems. The decision about whether to build a rail system to a U.S. airport may be driven more by the overall public transportation strategy of the region rather than by airport access needs in isolation. When a region, such as San Francisco, has invested heavily in downtown rail distribution services and other regional connections through the system, extension of that system to cover the airport can be seen as part of a regional transportation strategy. By contrast, when the rail services do not currently serve a major role in a bigger network of collection and distribution, the investment in a stand-alone rail system to the airport may not make sense. In this phase of the process of improving public modes to major airports, services must be designed to achieve certain service quality attributes revealed in the analysis of successful systems around the world. Chapter 3 summarizes a set of attributes that are important for services. Those attributes are not specifically tied to the choice of bus versus rail but rather describe the needs of the traveler without regard to mode or technology. Lessons Learned from Successful Systems The key lessons from the analysis of international systems presented in Chapter 3 do not form an argument for or against rail solutions in the United States. The key issue is to understand the attributes of service from the European experience and to design services that deal with those attributes. Each of the four attribute areas defined in that chapter can be reviewed for the impli- cations for a choice of mode in the United States. Six Steps in a Market-Based Strategy for Improving Airport Ground Access 25 Key Challenges in Step 4 • Design a set of services for – a dense urban market – an exurban market – a middle market • Incorporate the attributes of the successful sys- tems, including quality of – Line-haul service to CBD – connection at the airport – service beyond the CBD – appropriate baggage strategy • Design a set of services to appeal to four market segments: – Resident business – Resident non-business – Non-resident business – Non-resident non-business

Quality of the Line-Haul Connection to the CBD Finding an available right-of-way is a problem for the designer of a bus access system and for the designer of a rail system. Finding an available express track has been determined to be a prob- lem throughout Europe. Multistop rail transit service in London was perceived to be so slow that new, non-stop rail was created. Planners at Munich’s airport are looking at magnetic levitation (maglev) alternatives to deal with the historically slow rail travel times there. Universally, buses stalled in general-purpose traffic cannot provide a competitive advantage over the automobile. By contrast, volumes on the Braintree Logan Express bus service (Boston) increased by 50% when a bus lane was added to the system. If the metropolitan system can provide free-flowing bus lanes, total travel times may well be lower by bus. Simply extending multistop local service to include the airport is a formula doomed to failure. Quality of Connection at the Airport The selection of the rail mode does not ensure a good quality connection from the baggage pick- up location, nor does the selection of bus preclude a good connection. In Europe, some rail stations are located immediately adjacent to a common baggage pick-up location, while other rail stations require clumsy, uncomfortable connections by bus shuttle vehicles. In the United States, connecting charter buses leave from the Las Vegas airport from within a unified terminal complex adjacent to a common baggage pick-up area, while many U.S. rail services operate from locations far from major baggage pick-up areas. This issue of the high-quality connection between airline operations and the ground access vehicle needs to be solved for whatever ground mode is selected. On the other hand, the new data from Oakland challenges the assumption that directness of connection is more important than underlying market conditions. Certain market segments, such as resident non-business, may be willing to put up with lower levels of service amenity in a trade-off with more important trip-making objectives. Quality of the Connecting Service Beyond the Terminal Providing high-quality services to areas beyond the traditional downtown is a problem for both rail and bus systems. Connections between the major rail terminals in downtown London are difficult, and the mode share for Heathrow air travelers to connecting national rail service is low. By contrast, trains from Zurich Airport rail station are totally integrated into the national rail system, and mode share to national destinations is extremely high. The Newark Liberty International Airport rail station provides a case study of the appeal of longer distance rail services as a mode of airport ground access; at the present time, the market patterns are not showing the expected growth in ridership there. The Existence of a Strategy for Baggage While the designers of airport ground access systems must deal with the impediment of bag- gage and its negative impact on the choice of public modes, this report has created a compre- hensive discussion of the failure—through much of the world—of downtown airport check-in terminals operated by airline personnel. Chapter 5 documents problems at downtown terminals serving London Heathrow, London Gatwick, Munich, Newark, and Madrid airports, while reporting more positive market experiences in Hong Kong, Vienna, Moscow, and Kuala Lumpur. Systems operating national, longer distance rail equipment, such as that in use in Copenhagen, can allow for the use of existing baggage storage areas. For rail systems operating standard commuter and rapid transit equipment, the problem is only rarely solved in a manner satisfactory to the traveler with large baggage. Generically, the accommodation of baggage is not an issue between bus and rail, but rather is an attribute to be sought by the service designer. Dealing with the baggage issue tends to argue for 26 Ground Access to Major Airports by Public Transportation

the adoption of dedicated services (by whatever mode) rather than shared service of traditional mul- tistop transit (by whatever mode). Whether a dedicated train or a dedicated airport bus service is being considered, baggage handling can be designed in from the outset. Summary: Designing to Deal with Revealed Attributes For each of the four design areas specified previously in “Lessons Learned from Successful Sys- tems,” U.S. designers can strive to attain the attributes revealed in the successful international systems not by mimicking the choice of mode but rather through careful regional systems design that finds solutions for the issues defined by the four attribute areas. Design Airport Ground Access Services for the Three Geographic Areas At this point in the planning process, candidate markets for services can be defined. Within the contour for the market area of more than 50 trip ends per square mile, submarkets can be sought at significantly higher market concentration. With knowledge of the location of these strong market segments, rail and other services can be considered. The market research method in this report advocates defining a targeted geographic area for a given candidate service and understanding the airport trip-end density (all modes) from that geographic area to better understand the contribution that service can play. At least three geographic areas should be examined for the service most likely to meet the needs of the customer: • Services for the Dense Urban Market. High-quality line-haul service to the highest trip-end density should be developed, whether by rail or by dedicated airport bus. Examples of high- quality rail services include Washington, D.C.’s Metrorail system and Metropolitan Atlanta Rapid Transit Authority (MARTA) service to Hartsfield-Jackson Atlanta International Airport. Examples of high-quality bus service to the CBD include the specialized airport bus service from New Orleans International Airport, as well as those in Denver and downtown Seattle. • Services for the Exurban Market. Dedicated airport bus service from specially designed regional parking facilities should be examined to offer services to those areas where airport trip-end densities cannot justify or make feasible collection services. Both the Los Angeles International Airport FlyAway program and the (Boston) Logan Express programs are adding additional lines and services. • Services for the Middle Market. A variety of strategies should be explored for the majority of U.S. airport travelers who come from outside of the densest downtown areas, but within the principal market area of the airport, defined here as the area with more than 5 airport trip ends per square mile. Within this area, a wide variety of combinations of door-to-door, fixed-route, and, most importantly, combinations thereof, can be considered. Understanding Demographic Segments Within Each Geographic Market In the design of candidate services for each of the geographic areas, the market research–based planning process requires information beyond the density of trip ends. This report strongly rec- ommends that each geographic area be examined in terms of the four demographic segments: resident business, resident non-business, non-resident business, and non-resident non-business. In many cases, the support of a high-fare, high-quality premium service (such as the Heathrow Express) is dependent upon the strength of the business market. In other cases, the support of multistop transit service (such as the Blue Line in Boston) is dependent on a strong non- business market, including students and vacationers. In many cases, airport buses from regional collection points are very attractive to the resident market (who find lower parking charges) and Six Steps in a Market-Based Strategy for Improving Airport Ground Access 27

not at all attractive to the non-resident market (who find it more convenient to get rental cars on the airport than in outlying areas). The knowledge of demographic characteristics gained from the ground access survey will also become critical at the time of marketing and pricing the services. For example, to increase ridership on days of low business travel, a marketing strategy might offer low fares for families via local newspapers only. The incoming businessperson would not be aware of the existence of these fares and would continue to pay the higher basic fares. Such a marketing strategy would be designed to lower fares for that portion of the market that is elastic to fare change and not to lower fares for that portion of the market that is inelastic in relation to price. Best Practices in the United States: Service Based on Markets Examples of best practices can be found for all three of the submarkets, ranging from dense urban conditions to areas of dispersed origins. Best Practices for the Dense Urban Market • A good example of best case practices for service to areas with a high density of airport trip ends is the Airport Express bus service in New Orleans, which captured about 15% of the entire airport market before Hurricane Katrina. Its mode share rate for its primary market area (downtown) may be the highest of any U.S. airport. • The high-frequency AirBART bus operated by Oakland International Airport to the BART rail station captures about 9% mode share. It can be argued that this service is well matched with the needs of this airport dominated by a low-cost carrier. • To Reagan Washington National Airport, the Metrorail service covers the geographic area where most airport trips originate. This match between the origins of the riders and the location of the rail service in that area results in an airport-wide mode share of more than 12%. • An unusual best practice is the extension of the FlyAway express bus service concept to a new terminal within the Los Angeles Union Station, providing an exceptional level of urban intermodal connections. Best Practices for the Exurban Market • The Logan Express system serving Boston airport continues to grow as more services are added. These services capture an estimated 20% of their catchment areas. At the time of data collection, airport buses from three parking lots attracted more airport riders than the entire fixed-route and -schedule public transportation system. • The Marin Airporter is a privately owned service noted for its understanding of the market needs of its customers. The Marin Airporter has captured 30% of the travelers in its market area of San Francisco. • The Van Nuys FlyAway is a mature dedicated airport bus operation, capturing an estimated 17% of the travelers from its catchment area. Best Practices for the Middle Market While the dedicated express bus and the longer distance specialized van service are character- ized by line-haul trips of more than 10 miles, the middle market is marked by shorter trip lengths. Service operated in middle markets experiences competition from the pick-up/drop-off mode and the taxi mode. • In Oakland, door-to-door vans capture nearly 20% of their logical catchment area in a mid- dle market of less than 20 airport trip ends per square mile. Door-to-door services in an area immediately south of San Francisco International Airport, with much shorter trip distances, attract about 7% of their logical catchment area. 28 Ground Access to Major Airports by Public Transportation

• Chapter 4 documents similar markets in Las Vegas and Orlando, in which door-to-door vans capture more than 10% of the airport-wide ground access market. • In Seattle and Los Angeles, door-to-door vans capture more than 5% of their respective market areas in areas of middle market density. The middle market, described in this analysis as the zones of more than 5 and less than 50 air- port trip ends per square mile, is the largest of the three and the most difficult to serve. With low reported market capture rates and with occupancy levels approaching those of single-party taxis, this market segment needs the most research. Step 5: Manage the Airport to Encourage Higher Occupancy Vehicle Use Several elements of a comprehensive strategy to improve public mode airport ground access can be implemented within the boundaries of the airport itself. The manner in which the airport is managed can have a significant effect on the quality of the experience for those travelers who have chosen to access the airport by more efficient, higher occupancy modes. The various strategies for improving public transportation access to airports are set against the context that most U.S. airports are not managed to encourage the use of higher occupancy modes; in many cases, the opposite is the case. Airports are primarily seen as transfer facilities between various forms of automobile use and the air services operated at the airport. In many cases, the motivation for the creation of new strategies for managing ground transportation vehicles (such as peripherally located Ground Transportation Centers) is to remove the larger vehicles from the primary roadway, which is freed up to devote more capacity to private auto- mobile pick-up and drop-off. Encouraging the Use of High-Occupancy Service In the United States, currently only one airport has rail transportation that carries more air travelers for ground access than do the bus and van options: Reagan Washington National Air- port. Even at airports with new rail services, such as Portland, Oregon, more passengers depend upon bus and van services than upon the rail option to the downtown. And yet, in sharp con- trast to the recent advances in design for the airport/rail interface, there has been very little coor- dinated attempt to determine the potential of improving the connection between the bus and the airport activities. Ironically, the modes most successful at most U.S. airports—buses and vans—have received the least amount of attention in terms of functional priority at key airport transfer points. In the allocation of curb space, the lanes closest to the terminals (those with the shortest walking distance) can be allocated to the most efficient modes, rather than a traditional pattern of allo- cating these lanes to private vehicles. All too frequently, the traveler who chooses more efficient, higher occupancy modes from the airport is sent to an outer curb, unprotected from weather, with little in the way of accurate infor- mation or services. In many airports, the task of choosing a van operator, for example, occurs outside with no protection from rain or snow, or heat. In many cases, critical connections with long-headway regional services are made from an isolated curb, with no accurate real-time infor- mation informing the traveler that the bus is on time, is late, or has already departed. Often, trav- elers waiting at the curb for a shuttle bus to a regional rail system are not given information about the arrival time of the bus or the rail system it is serving. Six Steps in a Market-Based Strategy for Improving Airport Ground Access 29 Key Challenges in Step 5 • Manage the air- port to encour- age transit first • Allocate curb space to give pri- ority to those arriving by higher occupancy mode • Improve the architectural standards experi- ence of the pub- lic mode traveler • Build transfer facilities for bus and van modes to the design standards attained for rail projects • Modify regula- tions that make it difficult for the traveler to pur- chase public mode services at the airport curb • Modify regula- tions that make it difficult to use higher occupancy services to the airport without prior reservation

Learning from Recent U.S. Airport Designs Recent U.S. design experience at key rail projects can point the way towards the adoption of higher standards for transfer facilities for bus and van. The traveler inside the Newark airport terminals is offered real-time information screens that show the next departures from Newark airport rail station for both Amtrak and New Jersey Transit. The departure schedules of the two rail operators are displayed in chronological order on one screen, consistent with the needs of the traveler. Armed with this connecting mode information, the rail user can proceed upstairs to the Newark AirTrain people mover. All connections to the people mover are made within the interior spaces of the airport terminal. Accessing the AirTrain platform is simpler and quicker than getting to the major parking facilities. At the Newark airport rail station, the pedestrian paths are clear and the information about connecting services is abundant. For major transit investments in Washington, D.C., and San Francisco, high-quality architectural solutions have been designed for the transferring public mode traveler. At the reconstructed Reagan Washington National Airport, the MetroRail station is located closer to the terminal than is the major parking garage facility; travelers walk through the rail station lobby to get to the parking garage. The public transportation terminals built by WMATA in Washington and BART in San Francisco can be used as case studies in the improvement of the condition of the arriving passenger connecting on pub- lic modes. In Washington, D.C., the walkway bridges are heated and air conditioned and brightly lit. In San Francisco, the arriving traveler on ground transportation at the new International Terminal disembarks from the BART train at the same level as the airport check-in function: no bridges, no ele- vators, and no escalators will impede traveler flow from the three-track station. The operation of both a Delta Air Lines and an AirTran Airways check-in facility at the MARTA station within the Atlanta landside terminal is another example of high-quality archi- tectural integration. That rail station is located immediately adjacent to the common baggage claim facility for the entire airport, allowing the seamless connection from baggage pick-up to the rail platform overhead. Standards for the Ground Transportation Transfer Experience The architectural treatment at recently constructed rail stations establishes that the transfer experience to public modes at an airport can be positive. The question is then raised about the quality of transfer to buses and vans. It is not a question that can be solved quickly, or with only one solution. In some airports, a shared Ground Transportation Center is the optimum solution, and in others it is not. Clearly, if there is a guiding public policy to encourage the use of higher occupancy modes, the level of amenity offered to the connecting public transportation traveler should be as good as or better than that offered to the traveler connecting onward by private mode. Some of the strategies required by a comprehensive public policy are best carried out by the public sector, and some of the strategies are best implemented by the private sector. In theory at least, it is immaterial whether the onward connecting service is operated by the public or private sector; the public mode traveler should experience the same level of architectural amenity in the transfer act as comparable portions of the airport. At several large airports, bus and van passengers often board their vehicles at parking lots, dead end locations, outer curbs and other facilities with no traveler support services. Designs to Integrate Bus Systems into Airports Baltimore/Washington International Airport has adopted a managed strategy for authorized van service, with specific companies authorized for specific geographic areas. In design terms, this strategy makes possible the creation of a single departure point for all door-to-door services, located inside the airport terminal at the center of the terminal complex. The multiparty groups are formed inside this area with all waiting occurring inside with access to information. 30 Ground Access to Major Airports by Public Transportation

Similar advances in quality of terminal design have been incorporated into the centrally located Ground Transportation Center at Minneapolis–St. Paul International Airport, which is accessed by underground walkways from the main baggage claim areas. The act of finding, pur- chasing, and accessing public modes of transportation occurs in a heated/air-conditioned interior space integrated into the airport terminal complex. Similar high-quality pedestrian connections are offered in the underground connections to the departure area at Portland Inter- national Airport’s redesigned terminal complex, where ground transportation information and ticketing is provided within the underground walkway system. All taxi, bus, and van departures from the Atlanta airport occur from a compact departure area located at the western edge of the terminal immediately adjacent to the common baggage claim area for the airport. At Chicago O’Hare International Airport, a City Bus Center has been built to improve the quality of transfer to the bus modes, located within the central structure with enclosed walkways from the domestic terminals. Considering Regulations to Encourage Higher Occupancy Mode Strategies Many local policies concerning the potential encouragement of higher occupancy patterns are determined by pre-existing regulations concerning the management of taxis. In some airports, a traveler standing at the curb seeking to purchase a shared-ride service is often not allowed to enter the vehicle unless he/she leaves the curb, goes back inside the terminal, calls a reservations line, and then comes back to the curb to wait for a subsequent dispatched vehicle. Going to the airport, similar inefficiencies exist in the system, especially for the traveler who would like to board a shared-ride vehicle to the airport but has not formally “pre-arranged” the trip. The public policy goal of getting greater levels of vehicle occupancy is often undercut by regulations designed for general-purpose management of taxis. Public policies should be explored that would serve to maximize the occupancy levels of public mode vehicles to the airport. Best Practices in the United States: Management and Amenity U.S. best practices in this category tend to include examples of good architectural treatment of amenities for transferring travelers, rather than any airport-wide strategy to encourage higher occupancy. Examples of such details include: • The revised Ground Transportation Center in the center of Minneapolis–St. Paul Interna- tional Airport is a rare example of improved amenity for the traveler on buses and vans. • The City Bus Center at Chicago O’Hare International Airport provides deplaning passen- gers a comfortable waiting area with seating, where they can purchase food, beverages, and newspapers/magazines, which is linked directly via an underground walkway to each terminal. • The location of the MARTA station in the Atlanta terminal and the location of the Delta Air Lines and AirTran Airways check-in at the MARTA station. Step 6: Present Information about Ground Access Services to the Traveler Assuming the markets have been analyzed and services have been established, the last step in the process requires the creation of a program to make the traveler aware of the public trans- Six Steps in a Market-Based Strategy for Improving Airport Ground Access 31 Key Challenges in Step 6 • Include ground transportation itinerary trip planning capa- bility on airport websites • Include ground transportation timetables in printed docu- ments describing airport services • Work in coordi- nation with local efforts to develop the national 511 traveler informa- tion system • Integrate the ticketing reserva- tion process between aviation and ground systems

portation services offered and to facilitate the purchase of these services. Fortunately, the technology to improve the quality of information sent to the traveler is being developed and implemented at a rapid pace. Building a Ground Transportation Information Strategy The traveler needs to be aware that public transportation options exist. Airport websites should include some form of automated trip planning for ground trips to and from the airport. For each city and town of destination, an airport information system should describe the services available, based on the actual schedules of each component segment of the trip for that particular hour of that particular day. These systems can now tie directly into the reservations systems of the ground transportation operators. A website managed, or at least approved, by the local airport should include automated itin- erary trip planning encompassing all public modes available to and from the airport, including public modes traditionally used in the public transportation system and public modes available only for airport services. Such a program would logically include estimated taxi fares and travel times, accurate by time of day. No currently available regional trip planning program includes a full description of all vans, limousines, and buses approved for airport use. Only the 2007 trip planning system in Amsterdam’s Schiphol Airport integrates both airport-specific and regional services on one screen. The BWI Ground Access Information Module currently under beta test- ing will also provide these integrated information services. Until automated services are ubiquitously and easily available at airports, printed material from simple brochures to elaborate ground transportation guides will continue to be the back- bone of traveler information strategies at airports. Good examples of such materials can be found at Baltimore/Washington International Airport and many other U.S. airports. Best Practices: Traveler Information While the Amsterdam Schiphol Airport operates the only website that fully integrates airport- specific modes and general regional services, other more limited examples of best practice can be noted: • The Transport Direct website in the United Kingdom describes all public transportation ground access options (and private automobile) from all airports in the UK to all destinations in the UK. • The 511.org website provides all general public transportation services by combinations of carriers in the San Francisco Bay Area. • The Trips123 website provides all general public transportation services from all New York City airports to all areas in the tri-state region. • The real-time Amtrak and New Jersey Transit train departure screens in key locations at Newark airport are a good example of the kind of traveler information that has to be devel- oped in the United States. Real-time airline departure information is presented within the train station mezzanine level. Conclusion A major theme that emerges from Chapters 1 and 2 is the need for some party to take leader- ship, and very often that happens at the level of the airport management. The professional ground access staffs at leading airports such as San Francisco and Baltimore/Washington take a proactive role in examining the extent of coverage and providing incentives (such as the grant- 32 Ground Access to Major Airports by Public Transportation

ing of exclusive rights to serve a given area). In each of these cases, it is understood that there are costs associated with the establishment of high-quality services; these costs are often associated with the continued subsidy of these services. In nearly all of the best practices, such as the ter- minal changes at Reagan Washington National Airport or the early development of the Logan Express, there have been financial costs to bear. There is no working assumption that the solu- tion(s) to these problems will occur without significant costs. Six Steps in a Market-Based Strategy for Improving Airport Ground Access 33