National Academies Press: OpenBook

Ground Access to Major Airports by Public Transportation (2008)

Chapter: Chapter 3 - Attributes of Successful Ground Access Systems

« Previous: Chapter 2 - The Context for Public Transportation to Major Airports
Page 52
Suggested Citation:"Chapter 3 - Attributes of Successful Ground Access Systems." National Academies of Sciences, Engineering, and Medicine. 2008. Ground Access to Major Airports by Public Transportation. Washington, DC: The National Academies Press. doi: 10.17226/13918.
×
Page 52
Page 53
Suggested Citation:"Chapter 3 - Attributes of Successful Ground Access Systems." National Academies of Sciences, Engineering, and Medicine. 2008. Ground Access to Major Airports by Public Transportation. Washington, DC: The National Academies Press. doi: 10.17226/13918.
×
Page 53
Page 54
Suggested Citation:"Chapter 3 - Attributes of Successful Ground Access Systems." National Academies of Sciences, Engineering, and Medicine. 2008. Ground Access to Major Airports by Public Transportation. Washington, DC: The National Academies Press. doi: 10.17226/13918.
×
Page 54
Page 55
Suggested Citation:"Chapter 3 - Attributes of Successful Ground Access Systems." National Academies of Sciences, Engineering, and Medicine. 2008. Ground Access to Major Airports by Public Transportation. Washington, DC: The National Academies Press. doi: 10.17226/13918.
×
Page 55
Page 56
Suggested Citation:"Chapter 3 - Attributes of Successful Ground Access Systems." National Academies of Sciences, Engineering, and Medicine. 2008. Ground Access to Major Airports by Public Transportation. Washington, DC: The National Academies Press. doi: 10.17226/13918.
×
Page 56
Page 57
Suggested Citation:"Chapter 3 - Attributes of Successful Ground Access Systems." National Academies of Sciences, Engineering, and Medicine. 2008. Ground Access to Major Airports by Public Transportation. Washington, DC: The National Academies Press. doi: 10.17226/13918.
×
Page 57
Page 58
Suggested Citation:"Chapter 3 - Attributes of Successful Ground Access Systems." National Academies of Sciences, Engineering, and Medicine. 2008. Ground Access to Major Airports by Public Transportation. Washington, DC: The National Academies Press. doi: 10.17226/13918.
×
Page 58
Page 59
Suggested Citation:"Chapter 3 - Attributes of Successful Ground Access Systems." National Academies of Sciences, Engineering, and Medicine. 2008. Ground Access to Major Airports by Public Transportation. Washington, DC: The National Academies Press. doi: 10.17226/13918.
×
Page 59
Page 60
Suggested Citation:"Chapter 3 - Attributes of Successful Ground Access Systems." National Academies of Sciences, Engineering, and Medicine. 2008. Ground Access to Major Airports by Public Transportation. Washington, DC: The National Academies Press. doi: 10.17226/13918.
×
Page 60
Page 61
Suggested Citation:"Chapter 3 - Attributes of Successful Ground Access Systems." National Academies of Sciences, Engineering, and Medicine. 2008. Ground Access to Major Airports by Public Transportation. Washington, DC: The National Academies Press. doi: 10.17226/13918.
×
Page 61
Page 62
Suggested Citation:"Chapter 3 - Attributes of Successful Ground Access Systems." National Academies of Sciences, Engineering, and Medicine. 2008. Ground Access to Major Airports by Public Transportation. Washington, DC: The National Academies Press. doi: 10.17226/13918.
×
Page 62
Page 63
Suggested Citation:"Chapter 3 - Attributes of Successful Ground Access Systems." National Academies of Sciences, Engineering, and Medicine. 2008. Ground Access to Major Airports by Public Transportation. Washington, DC: The National Academies Press. doi: 10.17226/13918.
×
Page 63
Page 64
Suggested Citation:"Chapter 3 - Attributes of Successful Ground Access Systems." National Academies of Sciences, Engineering, and Medicine. 2008. Ground Access to Major Airports by Public Transportation. Washington, DC: The National Academies Press. doi: 10.17226/13918.
×
Page 64
Page 65
Suggested Citation:"Chapter 3 - Attributes of Successful Ground Access Systems." National Academies of Sciences, Engineering, and Medicine. 2008. Ground Access to Major Airports by Public Transportation. Washington, DC: The National Academies Press. doi: 10.17226/13918.
×
Page 65
Page 66
Suggested Citation:"Chapter 3 - Attributes of Successful Ground Access Systems." National Academies of Sciences, Engineering, and Medicine. 2008. Ground Access to Major Airports by Public Transportation. Washington, DC: The National Academies Press. doi: 10.17226/13918.
×
Page 66
Page 67
Suggested Citation:"Chapter 3 - Attributes of Successful Ground Access Systems." National Academies of Sciences, Engineering, and Medicine. 2008. Ground Access to Major Airports by Public Transportation. Washington, DC: The National Academies Press. doi: 10.17226/13918.
×
Page 67

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.

What makes a public transportation access system to a major airport successful? The breadth of travel patterns to specific airports (detailed in Chapter 4) shows the wide variety of experience around the world in the design and implementation of public transportation strategies to major airports. Those patterns range from the remarkable public transportation share in Oslo to the specialized role played by public transportation to most U.S. airports. This chapter interprets best practice and attempts to draw out lessons learned from this wide variety of experience. This chapter will examine the implications of certain attributes of successful services, whether those services are in operation in the United States, Europe, or Asia. One lesson is clear at the outset—no particular modal solution is optimal everywhere: a simple focus on line-haul speed of the vehicle does not produce a high mode share to public transportation, as revealed in Shanghai; the adoption of high-cost, high-quality rail design does not convince more Hong Kong travelers to ride the train rather than the bus; direct on-airport rail connections to an advanced regional rail system do not attract more travelers to choose the rail transit to the San Francisco International Airport than the less direct connections in operation at nearby Oakland Interna- tional Airport. It is a central theme of this report that the services offered must be based on an analysis of the needs of the traveler, not the adoption of one particular mode (usually rail) as the “world class” standard. However, most of the highest mode shares to transit reported in this study do come from European and Asian systems that use rail services as a major and dominant strategic com- ponent. This chapter seeks to look at service attributes attained in successful systems without regard to the dominant mode that resulted in those high mode shares to public transportation. After this examination of service attributes, Chapter 5 will explore the question of the integra- tion of ground access services into larger national systems, and the role of integration of baggage and ticketing systems. Then, Chapter 6 will present a discussion of the application of market research techniques to a planning process based on the needs of the traveler, including of the roles of geographic and demographic market segmentation. Understanding Successful Airport Ground Access Systems This section will focus on the attributes associated with the success of the rail projects that form the principal mode of most of the successful systems to be detailed on an airport-by-airport basis in Chapter 4. It will quickly become clear that no single attribute—such as the speed of the vehicle, the directness of the on-airport connections, or the connectivity to the rest of the pub- lic transportation system—can by itself explain the propensity for high market shares. Rather, it 52 C H A P T E R 3 Attributes of Successful Ground Access Systems

is clear that a successful ground access system will need to combine various attributes from sep- arate services designed to meet the needs of the separate market segments. As noted in previous chapters, most U.S. airports have at least three market areas: a dense downtown/inner market area; a distant set of dispersed origins, for which dedicated express buses can carry travelers collected by other modes; and a mid-suburban area, where door-to-door shuttle services can be supported. A quick summary of possible explanations of high mode share is presented in the following sections. Does Airport Size Explain Ridership? Does an airport have to be extremely large to justify and support an exemplary ground access system? As shown in Tables 2-1 and 3-1, the ranking of public transportation use cannot be explained by the location or the size of the airport. MAP is not a good predictor of total public mode market share: the largest airports, Heathrow and Frankfurt, rank in the middle of the sam- ple in terms of ground access market share; the smaller airports rank both higher and lower than the largest. The sheer size of an airport does not explain the mode share to public transportation services. Table 3-1 shows that London Stansted (smaller) has a higher mode share to public transporta- tion than does London Heathrow (larger). Oslo and Zurich are relatively smaller airports but have high mode shares. Oakland (smaller) has a higher mode share than Dallas/Fort Worth (larger). On the other hand, Paris de Gaulle (larger) has a higher mode share than does Paris Orly (smaller) and New York JFK (larger) has a higher mode share than New York LaGuardia Attributes of Successful Ground Access Systems 53 Rank Airport Public transport market share Size of airport Distance to CBD 1 Oslo 64% 16 MAP 30 2 Hong Kong 63% 44 MAP 21 3 Narita 59% 31 MAP 40 4 Shanghai 51% 21 MAP 18 5 Zurich 47% 19 MAP 7 6 Vienna 41% 17 MAP 12 7 London Stansted 40% 21 MAP 35 8 Paris Charles de Gaulle 40% 56 MAP 15 9 Amsterdam 37% 44 MAP 12 10 Copenhagen 37% 20 MAP 7 11 Munich 36% 31 MAP 17 12 London Heathrow 36% 67 MAP 15 13 Stockholm 34% 15 MAP 25 14 Frankfurt 33% 52 MAP 6 15 London Gatwick 31% 34 MAP 30 16 Geneva 28% 9 MAP 3 17 Brussels 26% 16 MAP 7 18 Paris Orly 26% 25 MAP 9 19 Düsseldorf 22% 15 MAP 5 SOURCE: M. A. Coogan, based on airport information. Table 3-1. Market share by size and location.

(smaller). In general, while airports need a certain size to support public transportation services, size alone does not explain high ridership. Distance traveled to the airport is worthy of more attention. Does Distance from Downtown Explain Ridership? Most airports serve one dominant downtown (e.g., Boston), or at least a set of dominant downtowns (San Francisco and Oakland/Berkeley). What is the influence of line-haul distance to the downtown mode share? Some trades-offs are clear: with close-in service, the taxi provides a cost-effective alternative to the public transportation trip, whereas with a distant airport it does not. For example, the sheer distance involved in a trip to Narita airport (located approximately 37 miles from downtown Tokyo) or Oslo airport (located approximately 30 miles outside Oslo) makes the taxi a weak competitor. Thus, airports that are relatively close to downtown, such as Reagan Washington National, tend to have a high taxi share to the airport. At the same time, the close-in airport can offer many destinations by public transportation with only a moderate amount of transferring (e.g., in Washington, D.C.). However there are some complexities to consider. High rail mode shares exist when the dis- tance is long, the taxi fare is high, and travel time can be gained on the line-haul segment to com- pensate for the non-directness of access at the non-airport end of the trip. At first glance, the high rail mode shares for Zurich and Copenhagen may seem to be an exception to this rule, as they are relatively close to the downtown. In fact, each of these airports is tied into an unusual nation- wide (and sometimes multi-country) feeder system. For example, the mode share to distant Swiss regions is very high, while that to the center of Zurich is low, because the taxi is a feasible alternative. But such programs as that in Copenhagen, with its new tunnel/bridge directly from Copenhagen airport to Sweden, are a part of a longer distance national feeder system, not just a local one. As a general rule, the longer the ground access trip, the less competitive is the taxi, and the less attractive is the casual kiss-ride drop-off trip. Does the Quality of the Airport Connection Explain Ridership? Looking at the connections on the airport, most of the public transportation services included in the sample of European/Asian airports have direct rail service to the airline terminals on the airport grounds. A major exception to this is Paris Orly airport, which operates a people mover over a 3-mile guideway to transfer travelers to the regional rail line that also serves Paris de Gaulle airport to the north. Thus, with both the quality of the line-haul service and the connectivity with the rest of the system constant, the Paris airport with the direct connection can be seen to have a higher market share to rail than the airport without the direct connection. At face value, a service with no change of vehicle at the airport should be expected to capture a higher market share than a service with a transfer at/near the airport, all other things being equal. For example, a traveler using rail from either downtown Dallas or Fort Worth would have to transfer once at the rail station, and a second time at a remote parking lot before getting a bus to any one of the five airline terminals. A low market share would be expected when compared with a bus or van that goes directly from major hotels in those two downtowns to the airports. However, in the United States, airports with direct rail service to the terminal area do not nec- essarily attain a higher share to public modes than those that do not. Of the ten U.S. airports with the highest mode shares to public transportation shown in Table 2-1, only two airports (Atlanta and Reagan Washington National) have rail service direct to the terminal complex; seven airports do not have rail service direct to the terminal; and San Francisco has direct rail service only to one terminal. In the latter category, the exceptionally high mode share attained by the 3-mile bus con- nection at Oakland International Airport needs some explanation other than minimization of 54 Ground Access to Major Airports by Public Transportation

transfer! The Bay Area case study is similar to the Paris case study in that both airports connect to the same regional rail system. With the quality of the rail system held constant, the 9% mode share in Oakland, compared with 7% from San Francisco, cannot be explained simply in terms of the ease of airport transfer. In the same vein, the 8% mode share to rail at New York JFK airport (no direct rail) compares favorably with direct on-airport rail connections in Chicago (Midway and O’Hare); Portland, Oregon; St. Louis; Minneapolis–St. Paul (determined from interviews with airport personnel); Philadelphia; Cleveland; and Baltimore. Düsseldorf airport provides another case study: it offers both a direct on-airport rail connection and an indirect connection via people mover to a nearby station; travelers choose the indirect connection over the direct connection by two to one. In short, directness of the connections on the airport cannot explain the wide variation in mode shares reported, although there is strong anecdotal data to support the idea that fewer transfers are better than more transfers. Does Line-Haul Speed Explain High Ridership? Without question, the speed of the line-haul vehicle between the airport and the downtown area is important. Table 3-2 shows the relationship between overall speed of the train and the mode share attained. Average speeds of more than 40 mph are attained in Zurich, Oslo, Narita, Attributes of Successful Ground Access Systems 55 Airport Market share by rail Rail travel time (min) Distance from CBD (miles) Implied rail speed to CBD (mph) Zurich 42% 10 7 42 Oslo 39% 19 30 95 Narita 36% 55 40 44 Amsterdam 35% 17 12 42 Copenhagen 33% 13 7 32 Munich 31% 40 17 26 Vienna 30% 16 12 45 London Stansted 29% 40 35 53 Paris Charles de Gaulle 28% 35 15 26 Hong Kong 28% 23 21 55 Frankfurt 27% 12 6 30 Express 9% 15 15 60 London Heathrow (23%) Tube 14% 45 15 20 Geneva 21% 10 3 18 London Gatwick 20% 30 30 60 Stockholm 18% 20 25 75 Düsseldorf 18% 12 5 25 Brussels 16% 14 7 30 Paris Orly via People Mover 14% 35 9 15 Shanghai Maglev 6% 8 18 135 SOURCE: M. A. Coogan. Table 3-2. Market share by time and speed.

and Hong Kong and contribute to strong rail mode shares in those cities. But, Table 3-2 shows that line-haul speed alone does not explain the propensity to attain high market share. High-Speed Service and High Market Share: Oslo Airport Express The Oslo Airport Express train (Figure 3-1), which has the second highest mode share to rail in the sample, is an example of a strategy based on a determination to attain high running speeds and low terminal-to-terminal travel times. From the beginning, the running time of the train to the new airport was to be no longer than the running time of the bus from the existing airport— 19 minutes. For this investment, the government set the following policy goal: the airport rail system would attract 50% of the market, a mode share considerably higher than any system had attained to date. Of this desired share, 42% was set as the goal for the Oslo Airport Express ser- vice, with an 8% goal established for the traditional national train service. In Oslo, the strategy to provide high-speed service to the downtown and additional direct service beyond has resulted in a 39% market share for the dedicated Airport Express train and another 13% mode share to the slower, lower priced Norwegian Railway. High-Speed Service and Low Market Share: Shanghai Maglev A dramatic example of a strategy to build a market based on the speed of the line-haul vehicle comes from the Shanghai maglev project (8). On first look, the service characteristics of the maglev are impressive. While the bus takes about 60 minutes and the taxi takes 50 min- utes, the maglev makes the line-haul segment of the trip in just 8 minutes. The headway of the super high-speed train is 15 minutes. A good connection is available at the airport: the maglev station is connected by a pedestrian bridge (see Figure 3-2); no people mover or shuttle bus is needed to access the service. However, it was not possible to get a maglev directly into the center of the city, so a terminal was built on the edge of the downtown next to an existing metro stop. 56 Ground Access to Major Airports by Public Transportation PHOTO: M. A. Coogan. Figure 3-1. The Oslo Airport Express train was specifically designed for high speeds on this service.

From Shanghai Airport, the exclusive airport bus follows a strategy of serving several areas directly. Seven separate airport bus lines are operated to such destinations as the main train station and the City Air Terminal. Headways for the separate bus services range from 15 to 30 minutes. At a cost of around $7, the maglev service is roughly twice the cost of the airport bus, while still somewhat cheaper than a taxi for one. However, with a party of two, the taxi becomes cheaper than the maglev and directly competitive with the airport bus. The faster maglev attracts only about 6% of the market, compared to 43% for the more direct (and cheaper) airport buses. Market research undertaken in Shanghai shows that people traveling on business had a lower than average use of the maglev, while their use of taxi (25%) was the high- est of any market segment. Indeed, the business travelers also had the highest use of the airport bus of any market segment, at 48% mode share. Retired persons had no recorded use of the maglev, presumably because of the price differential. Highest use of the maglev came from “tours” and “vis- iting friends.” About half of the trips by arriving air travelers involved only one mode; about an equal number involved two modes, the most popular being airport bus and taxi (about 15% of all trips). Without question the low market share gained by the high-speed maglev is surprising. The analysts noted that the higher income markets, like those traveling on business, chose the taxi in spite of the obviously longer travel time to the city edge, at 60 minutes versus 8 minutes. Clearly, the lower income travelers selected the cheaper buses, while the business travelers went for the no-transfer service offered by the taxi. The lack of selection of the maglev-plus-taxi option is puzzling. The implications are clear: the analyst and service designer must be concerned with the door-to- door travel times and the directness of public mode services rather than with the highest speed of the vehicle (reported at 450 km [~280 mph] per hour for the Shanghai maglev). These conclusions are consistent with the Hong Kong experience of the market response to one high-speed rail line compared to a wide variety of more direct bus lines, as discussed below. In both cases, the resident (who is aware of the local options) has a greater propensity to choose the directly routed bus than does the visitor (who is less aware of local options). Attributes of Successful Ground Access Systems 57 PHOTO: http://home.wangjianshuo.com/archives/20030809_pudong_airport_maglev_in_depth.htm. Figure 3-2. The Shanghai Airport maglev station (left) is directly connected to the air terminal (right) by this pedestrian bridge.

Is Higher Speed or Directness of Service More Important? In the case studies of successful rail services to downtown, two strategies for service design emerge: (1) focus on the line speed to the terminal or on the quality of distribution services, and (2) minimize the headway that comes from joint operation with regularly scheduled services. Both strategies seek to produce a door-to-door travel time that is competitive with the taxi and the private vehicle. In the comparison of the two strategies, the Oslo Airport Express can be used as a prototype of the high-speed dedicated-service strategy (in which services are designed specif- ically for air travelers) and Munich’s standard S-Bahn can be a prototype of the lower speed shared-service strategy (in which air travelers share public transportation services designed for commuters and others). In the evolutions of these systems, service was improved in Oslo by decreasing the line time, while service in Munich was improved by doubling the number of trains, thus lowering the waiting time by 50%. An Example of Low-Speed, Shared Service: Munich Although several cities have chosen to create dedicated express airport services, most of the airports in the sample are served by rail lines that are also used by daily commuters. Munich can be used as an example of a local strategy, because, as shown in Figure 3-3, the airport station is served by only conventional metropolitan railway equipment, with no direct national service. In the 1990s, the Munich S-Bahn system made a major improvement to airport service with the addition of a second local rail line, making no change in the basic strategy to serve the airport with the existing metropolitan rail system. In 1998, the Munich system doubled the amount of service to the airport with standard local equipment providing service that is shared with the other users of the system. A new line was extended for 4 miles from an existing route, the S-1 (shown at the left end of the dotted line on Figure 3-3), at a cost of DM 220 million (US $121 million). In the first months of the new service, ridership from the airport station increased by 7%, with air traveler mode share rising from 28% to 31%. This increase in ridership is notable because the actual travel time by either of the two lines to downtown remains about 40 minutes. This travel time is similar to that of the London Underground from Heathrow airport but worse than that of most other local airport services. 58 Ground Access to Major Airports by Public Transportation SOURCE: Copyright Münchner Verkehrs- und Tarifverbund GmBH (MVV) Munich. Figure 3-3. Munich Airport (upper right) is served by two local train lines, with good network coverage.

The Role of Distribution and Connectivity With shared services, the line-haul travel speeds from the airport to the CBD are slow, but the service is well integrated with local distribution systems. At both London Heathrow and Munich airports, the local rail service, with its shared services, captures more of the market than does any other service. An example can be observed in London: dedicated service on the Heathrow Express takes about 17 minutes to Paddington Station (central London), leaving every 15 min- utes. The Underground’s Piccadilly Line to central London takes about 40 minutes, leaving every 4 minutes. The Express traveler waits an average of 7.5 minutes and travels 17 minutes, for a total travel time to Paddington Station of about 25 minutes. The walk from the express rail platform, through the Paddington Station complex to the specific underground platform takes about 7 minutes. The headway of the connecting service may add another 5 minutes of waiting time. Examination of total trip times shows that there are only a small number of Underground sta- tions (the immediately adjacent stations on lines connecting from Paddington) at which the total travel times for the Heathrow Express plus Underground are superior to the Underground plus other Underground travel times. Shared services make the traveler endure whatever level of overcrowding exists on the rail vehicle during rush hour, which, in London, can be a serious problem. Dedicated services pro- vide guaranteed quality of service on the line-haul segment, leaving the traveler with the need to find adequate distribution from the rail terminal. Case Study: Fast Service versus Slower, More Direct Service Planners at the Hong Kong Mass Transit Railway Corporation (MTRC) have been examining the competitive market position of the fast rail and the slower bus services available to the air traveler. High-quality air-conditioned buses, often double-decked, provide direct service to many urban destinations. Looking only at travel from the airport to downtown (Central Station), the fast train provides service in 23 minutes, at a fare of more than $10 US; the Airbus A route takes 48 minutes and charges about half as much; while the standard city bus takes 53 minutes and charges much less. At the time of the analysis reported here (1998), the rail gained 21% of the market; the Airbus, 16%; and the city bus, 20%. The factors that result in this high mode share to bus seem to include more than price mini- mization, because MTRC also provides good lower priced rail service to the airport complex. From the beginning, planners designed the rail system to operate with two price points. While the Airport Express Line train to downtown operates directly from the airport terminal, a second standard train, reached by shuttle bus, operates from a nearby station. The entire trip (shuttle plus train) on the standard train is about one third the cost of the express, making the shuttle plus train option directly comparable with the cost of the city buses. In fact, the air traveler who uses this lower priced rail connection can get to Central Station in only 39 minutes, compared with 53 minutes on the city bus. But for this price-sensitive market, the shuttle bus–to–rail con- nection is capturing only 3% of air travelers; the direct city bus captures 20%. The bus system serves many area destinations directly, with no change of mode required for the trip. For the air traveler, directness of service may be more important than price minimization or line-haul speed to the terminal point. To understand the motivation for mode choice—and to explore the attribute of directness of service—MTRC managers conducted market research. Of those travelers on the direct bus routes, an expected 55% said that the lower fare was a reason for choosing the bus; importantly, 51% stated that directness of service (i.e., no need to transfer) was a reason for their choice of mode. Directness of service was considered a factor by only 18% of rail travelers, presumably those with destinations convenient to the terminals. Attributes of Successful Ground Access Systems 59

Of all travelers on the Airport Express, an expected 63% stated that speed was the reason for choosing the rail. Some 13% mentioned the fare as the reason, which is lower than the fare for either taxi or airport door-to-door bus service. In an important conclusion, one of the original architects of the Hong Kong Airport Express writes: “It is apparent that even with a good design and well-integrated railway service, the Airport Express does not have inherent advantages over more direct single-mode bus travel. In other words, the speed advantage of rail versus single-mode road competitors when traveling over distances of only up to 34 km [21 mi] do not result in significant enough time savings to compensate for the necessary transfer.” (9, emphasis added) Lessons Learned: The Importance of Line-Haul Speed and Directness of Service The examination of relative line-haul speeds in the database of successful European/Asian air- port rail operations has several key implications for the U.S. practitioner. The first implication, and by far the most important, is the difference that exists in the basic travel-time conditions, largely associated with the existence of fast highway connections in the United States. Four of the airports in the sample offer service to downtown that is twice as fast as automobile service. Table 3-3 shows that automobile travel times in Oslo are more than twice as long as the rail line- haul time. Table 3-3 shows many examples in which the automobile travel times are significantly 60 Ground Access to Major Airports by Public Transportation Airport Market share by rail Auto travel time (min) Rail travel time (min) Ratio of auto time to rail time Distance from CBD (miles) Dedicated service? Zurich 42% 20 10 2.0 7 No Oslo 39% 50 19 2.6 30 Yes Narita 36% 90 55 1.6 40 Yes Amsterdam 35% 30 17 1.8 12 No Copenhagen 33% 13 13 1.0 7 No Munich 31% 35 40 1.1 17 No Vienna 30% 17 16 1.0 12 Yes London Stansted 29% 70 40 1.7 35 Yes Paris Charles de Gaulle 28% 45 35 1.3 15 No Hong Kong 28% 35 23 1.5 21 Yes Frankfurt 27% 20 12 1.7 6 No Express 9% 45 15 3.0 15 Yes London Heathrow Tube 14% 45 45 1.0 15 No Geneva 21% 10 10 1.0 3 No London Gatwick 20% 80 30 2.7 30 Yes Stockholm 18% 41 20 2.0 25 Yes Düsseldorf 18% 12 12 1.0 5 No Brussels 16% 20 14 1.4 7 No Paris Orly via People Mover 14% 25 35 0.7 9 No Shanghai Maglev 6% 50 8 6.2 18 Yes SOURCE: M. A. Coogan, based on airport and rail information. Table 3-3. Market share by comparative times.

higher than the rail travel times. Given the extent of roadway investment in the United States, attaining similar relative travel-time advantages for rail services will be difficult in most U.S. applications. The second implication is that the rankings of services by relative travel times to downtown do not correlate linearly with the rankings by mode share performance. The data reveal that it is the comparative travel time on a door-to-door basis that seems to influence choice. The data presented in Table 3-3 show that the focus on travel time to one location may be unproductive. For example, there are many points in central London where the slower mode (i.e., the Under- ground) gets the traveler to the destination without the negative experience of the transfer. Likewise, there are many points in Hong Kong where the slower mode (i.e., the direct bus) serves the traveler more directly than the faster mode. The third implication is that the travel-time characteristics to downtown may not be a good surrogate for the travel-time characteristics to the actual destinations of the users. The travel time to downtown Geneva is an interesting piece of information, but 75% of those leaving the Geneva airport are not going to the city of Geneva. The ratios of comparative travel times to Lausanne or to Bern are considerably more favorable to rail. The service must be designed based on the understanding of the needs of the travelers and must reflect the actual spatial distribution of trip- end destinations. The Implications of Dedicated Premium Service Dedicated versus Shared Service Public transportation services to airports can be categorized as either a dedicated service or a shared service. In the United States, there are no examples of rail service dedicated only to air trav- elers, but the Logan Express (Boston), the Van Nuys FlyAway (Los Angeles), and other airporter buses in major U.S. airports are all examples of service designed specifically for the air traveler. In Scandinavia, cities such as Helsinki and Gothenburg that have dedicated bus services attract higher levels of market share than do many cities with rail connections. Under the dedicated con- cept, services and vehicles designed specifically for the needs of the air traveler are provided. With shared service, air travelers use the same vehicles as other public transportation passengers in the corridor of service. European and Asian airports have many examples of rail services operated for air travelers only. Of the nineteen European/Asian airports in the sample, nine have dedicated rail services (shown in the last column of Table 3-3). In London, both the Gatwick Express and the Heathrow Express rail services are examples of dedicated service, with vehicles designed for the air traveler. Service to Heathrow Airport on the London Underground’s Piccadilly Line and other commuter rail services stopping at Gatwick Airport are examples of shared service. Many dedicated services market their high-quality line-haul times with fast service to only one downtown terminal. Most shared services, such as the Piccadilly Line to Heathrow, provide rel- atively slow speeds into the city, but with distribution to many points in downtown. In many cases, the dedicated service (e.g., Gatwick Express, Heathrow Express) utilizes a vehicle (origi- nally) designed to accommodate checked baggage. In most shared services, such as Munich’s S-Bahn service, no specialized vehicle is used, resulting in vehicles that may not serve travelers’ need for extra baggage space. The ten airports without dedicated service have chosen to provide public transportation that is designed primarily for commuters and the rest of the system. A characteristic of the dedicated-service strategy is the ability to provide minimized travel times between the airport and the downtown. However, the most successful overall mode share is gained by airports that offer a variety of strategies. Attributes of Successful Ground Access Systems 61

Table 3-3 shows that, in general, providing dedicated service does not itself guarantee high market share to rail. Looking at the 14 airports with rail mode share of 20% or higher, seven do provide dedicated service and seven do not. Increase in Mode Share Due to Dedicated Premium Service Given that every airport needs lower priced shared ground access services, the following ques- tion is raised: how much increase in rail market share would result from the addition of dedi- cated service to the existing shared service? This question is currently being examined by ground access planners in Chicago, New York City, and Paris. As summarized below, planners are designing higher priced rail services to O’Hare, Midway, JFK, and de Gaulle airports to serve in addition to existing lower quality rail services to those airports. Ridership data that document the experience of Heathrow airport can help answer this ques- tion. Longitudinal data have been created that describe the change in overall rail market share between Heathrow and central London resulting from the addition of highly specialized dedi- cated services to a system that already offered one-seat, non-dedicated services shared with all other rail system travelers. Data from before and after the addition of Heathrow dedicated ser- vices have been examined and, from these data, an expansion factor for each of the four market segments has been calculated to represent the growth in market share attributable to the addi- tion of dedicated rail service. Importantly, the market segment most impacted by the premium service is the resident business segment, which experienced a 60% growth in market share. By contrast, resident non-business, more concerned about cost minimization, grew only by 13%. Non-resident market share to rail grew by about 40%. All in all, the addition of premium rail service to the existing shared rail service resulted in a 33% growth in rail mode share to Heathrow. Service Attributes of Proposed Projects In four cities around the world, major capital investments to improve rail services to major airports are being considered. In the case of Berlin, the decision was made in connection with the decision to focus all airport activity on one new regional airport, phasing out older closer-in facilities. In the other three cities—Paris, Chicago, and New York, planning is underway to pro- vide high-quality service that is dedicated to the needs of the air traveler. In addition, there have been plans for several years to build a maglev train between Munich Airport and the downtown Main rail station (Figure 3-4); the political future of that project is not clear after a tragic acci- dent on the maglev test facility in Northern Germany. It is important to note that in each of these four examples, local decision makers are consid- ering dedicated, premium service concepts at this time. The actual form of the two U.S. projects, however, is still under active debate. Berlin Brandenburg Airport In 1999, German Railways announced its decision to develop a dedicated train to operate express service to the new Berlin Brandenburg International Airport, which will consolidate and replace the existing airports in Berlin. An S-Bahn suburban rail line already serves the site for the new airport, currently known as Schonefeld Airport, with a 25-minute service to downtown. The S-Bahn division of German Railways will develop a new dedicated express line that will connect with Berlin’s new central rail station, called “Berlin-Lehrter Bahnhof,” with only two intermediate stations. The specially designed trains will be capable of 100 mph service and will reduce the running time to downtown to 18 minutes. 62 Ground Access to Major Airports by Public Transportation

Paris Charles de Gaulle Airport For years, the access strategy between Charles de Gaulle airport and downtown Paris has been based on the use of standard regional rail services, which are shared with commuters. No spe- cialized service to the downtown was planned. Over the past decade, the French National Railways (SNCF) and Aéroports de Paris (ADP) worked to develop a new dedicated high-speed service to the Gare de l’Est. Responsibility for leading the planning for the new service has recently shifted entirely to the national railway system. Thus, the French authorities are planning for Charles de Gaulle airport to have two services available at two separate price points: a shared commuter service along the RER-B with distribution services through downtown Paris and the suburbs to the south, and a second, non-stop dedicated train service to a terminal at Gare de l’Est, where a transfer to taxi or other mode would be needed to continue the onward journey. In one routing option, an expensive new tunnel was proposed between the airport and the city; more recently the rail authorities are examining more efficient use of the existing rail right-of- way currently used by the RER trains. At present, no downtown baggage check-in services are planned for the new dedicated rail service. While funding commitment for the project has been lacking, the ADP website reports that the French government is now committed to the project. Chicago Midway and O’Hare Airports The Chicago Transit Authority developed an ambitious plan to operate a dedicated airport train service from O’Hare International Airport to downtown and continuing on to Midway Interna- tional Airport. A central feature of the plan was the creation of bypass tracks at key stations along the Kennedy Expressway, which would allow for an elaborate skip-stop operation that would reduce travel time from about 45 minutes to perhaps 25 minutes. The plan proposes the creation of a new station at a downtown location (called “Block 37” on State Street) where a connector could be built from the present tunnel of the Blue Line (serving O’Hare) to that of the Orange Line (serv- ing Midway.) At this point a single station would be dedicated to the Blue/Orange train, with the Attributes of Successful Ground Access Systems 63 PHOTO: M. A. Coogan. Figure 3-4. The future of the proposed Maglev from Munich Airport (shown here in a demonstration mock up at Terminal Two) is unclear.

possibility of a baggage check-in station. New train equipment would be purchased that could accommodate both the air travelers and their baggage. A 2006 consultant report examined a less costly proposal, in which a dedicated service would be provided, but the investment in the bypass tracks postponed (10). This service would run “closed door” on the existing rails, in effect waiting as the preceding train stopped at each station but not serving those stations itself. This concept was developed in the original “Train to the Plane” service operated to JFK in New York for several years. Over time, bypass tracks could be added incrementally to allow the dedicated train to overtake local trains at key locations. Oper- ational details for the new Chicago dedicated service will be developed over time. New York JFK Airport The concept of a new dedicated rail service for air travelers between the general site of the World Trade Center and JFK airport was proposed in the aftermath of the September 11 attack on Lower Manhattan. For this project, a major engineering study recommended the creation of an entirely new tunnel between Brooklyn and Lower Manhattan, whether for commuter rail service to a new terminal or for extension of rapid transit north into Manhattan. In either case, a new air traveler terminal with full airline baggage check-in was proposed near the World Trade Center site. In the proposed concept, a hybrid form of the existing JFK AirTrain people mover would operate counter-clockwise around the existing air terminal loop and proceed to Jamaica Station. There the vehicle would switch from the existing linear induction propulsion system to a third-rail propulsion system. The vehicle would then reverse direction, traveling from the Jamaica Station to a new connection to the existing Long Island Railroad (LIRR) Atlantic Branch elevated system to a point in downtown Brooklyn. In the preferred plan a new tunnel would be built from there to the World Trade Center site. In an alternative, the new ser- vice would be connected to the existing Montague rapid transit tunnel, allowing stops at Broad Street, Fulton Street Transit Center, and Chambers Street, according to the project press release of May 5, 2004. Since the conclusion of that 2004 feasibility study, the prime proponent of the project, the Lower Manhattan Development Corporation, has ceased operations. A planning study has since been undertaken that “will evaluate all reasonable rail and non-rail alternatives which potentially address the project’s goal and objectives” (11). The project created a list of 47 options for study. No funding commitment has been made to the project, which is being reviewed by the new Governor of New York, particularly in the context of competing trans- portation proposals. Summing It Up This chapter has reviewed a series of attributes associated with successful airport ground access systems. It has established that no single attribute or characteristic can be used alone to predict the level of market share attained by public transportation services to airports. Rather, the total travel time—which includes the efficiency of the connection on the airport, the speed of the vehi- cle to the terminal, the quality of distribution services experienced after the line-haul trip, and the provision of services meeting the unique needs of the air traveler—all interact in determin- ing the marketability of the trip. No mode emerges as perfectly matched to all trips. The experience of fast express time can be marred by the lack of a taxi at the arrival terminal. The experience of a van trip operating directly to a hotel can be damaged by long in-vehicle times serving the needs of three or four other pas- sengers and by long waits to assemble the trip at the airport curb. 64 Ground Access to Major Airports by Public Transportation

Desired Attributes of Rail Service to U.S. Airports The two previous TCRP studies reviewed a wide variety of factors associated with the success or lack of success of airport rail services around the world. The following key factors have been shown to affect the use of rail service: • Proportion of air travelers with trip ends in downtown or the transit-rich core areas. For example, at Reagan Washington National airport about 33% of all air travelers have trip ends in the downtown area. Other airports where large proportions of travelers have downtown trip ends include those serving Boston, Chicago, New York, and San Francisco. At most air- ports, fewer than 15% of all travelers have trip ends in the downtown area. Thus, in most com- munities, the geographic service area directly served by a downtown rail service represents a relatively small percentage of the total air traveler market. • Characteristics of air traveler market. Air travelers with few or no checked bags are more likely to use rail service. Large family groups are less likely to use rail. Thus, airports serving a high proportion of business trips (e.g., Atlanta and Reagan Washington National airports where more than 40% of the travelers are making business-related trips) are more likely to attract rail users than those serving tourist destinations (e.g., Las Vegas and Orlando where less than 30% of the travelers are making business-related trips). The proportion of passen- gers familiar with regional transit systems (i.e., who understand the schedules and how to purchase a ticket) is also important. • Regional travel time. The availability of direct service between the airport and downtown (or major activity centers) allowing travelers to avoid transfers or multiple stops is important. Travelers going between the airport and downtown encounter 6 to 9 station stops at Reagan Washington National airport versus 15 or more stops on less successful rail systems. As evi- denced by the data, travelers tend to use rail service when they are concerned about (1) unre- liable travel times on access roadways or encountering traffic delays en route to the airport and (2) the lack of convenient parking at the airport and the need to search for an available space. • Ability to walk between station and destination. Air travelers may find using rail service more attractive if their final destination is within walking distance of the station, and less attractive (and less convenient) if they must transfer to a second mode (e.g., a bus or taxicab) to travel to/from the station. The need for travelers using rail service to wait for and transfer to a sec- ond mode may provide a travel time advantage for door-to-door services. • Extent of regional coverage. A comprehensive rail network, serving a large catchment area, will serve a larger potential market and provide travelers with more travel opportunities (e.g., those who may wish to leave from their place of work and return to their home) than does a rail system consisting of a single line between downtown and the airport. • On-airport travel time. The time (and distance) airline passengers are required to travel between the station and their gate is also important. Convenient rail service is easier to provide at airports that have a single terminal (e.g., Atlanta or Chicago Midway airports) than those that have multiple terminal buildings (e.g., New York JFK, Boston, or Paris de Gaulle airports) where travelers must use intermediate shuttle buses or people movers to get to the rail station. • Frequency of service. Waiting times of 10 minutes or less are preferred. The rail service at one U.S. airport operates on 30-minute headways, while a taxi ride downtown at the same airport requires a wait of only 15 to 30 minutes. The availability of late-night and weekend service is also important. Desired Attributes of Van and Bus Service to U.S. Airports The TCRP studies documented that air travelers represent a unique market that differs from traditional daily commuters. Compared to daily commuters, air travelers are typically more time sensitive and less cost sensitive, have more baggage, use the transit system less often, and are more Attributes of Successful Ground Access Systems 65

likely to use the system outside of normal commute hours. Often designing a special bus or van service to respond to this market is easier than trying to adapt a commuter-oriented, multistop bus (or rail) service to meet the needs of both daily commuters and air travelers. Door-to-door van and express bus services are examples of airport access modes that respond to the desire of air travelers for greater convenience and faster travel times than are typically offered by multi- stop bus services. Many operators of rail service prefer not to have airport-dedicated vehicles (e.g., with special baggage racks), because these special vehicles reduce their flexibility in the use of equipment. In the United States, specialized services have been developed to respond to specific markets not well served by traditional transit services. These services include the express bus services operated at the airports serving Boston (Logan Express), Denver (employee-oriented SkyRide), Los Angeles (Van Nuys FlyAway), and San Francisco (Marin Airporter). None of these services rely upon the general-purpose transit configuration of the metropolitan area. In each case, the specific needs of the target market segment were defined and provided for. In general, each of the transit services was able to attract about 20% market share in its immediate service area. Market conditions improved for the Logan Express’s Braintree service when both a new express bus lane and a new tunnel serving the airport were opened. Braintree Logan Express’s average daily ridership increased 50% as a result of the new radial bus lane, the commercial-vehicle-only tunnel, coordinated HOV policy, and other factors. As with rail systems, numerous studies have documented the requirements for a successful bus and van transportation service. At an airport, the following key factors affect the use of bus and van services: • Door-to-door transportation. Many air travelers are willing to pay additional fares for the con- venience offered by door-to-door services because they value travel time (particularly reliable travel time) more highly than travel costs. Such services also allow travelers to avoid transferring between airport access modes. • Express bus service. Express bus services, particularly those that offer travel time savings and service from intercept lots near regional access roads, have proven attractive to specific air traveler market segments. • On-airport travel time. The time (and distance) airline passengers are required to travel between the terminal and the boarding area is an important consideration. As with rail systems, an airport with a single terminal building allows better levels of service (i.e., fewer stops and faster travel time) than does an airport with multiple terminals or bus stops. • Pick-up/drop-off locations. To best serve the needs of travelers, drop-off locations should be located immediately adjacent to ticket counters and pick-up should occur next to baggage claim areas, preferably in areas reserved for buses, vans, and other commercial vehicles. • Frequency of service. The availability of off-peak, late-night, and weekend service is also important as many airline passengers travel during non-commuter hours (e.g., the peak hours at many airports are 11 a.m. to 1 p.m. on weekdays). • Regional travel time. The availability of HOV lanes on airport access routes can allow bus and van services to offer a travel time savings compared to private vehicles. The ability to stop at major activity centers, thereby allowing the traveler to avoid the need to use a second, connect- ing travel mode at the non-airport end of the trip, is an advantage. • Form of competition. The measures used to control competition between bus, van, and other rubber-tired services (e.g., taxis and limousines) are important. In an open market, a legiti- mate operator offering high-quality service will find it difficult to compete financially with an operator who (1) uses vehicles that are improperly maintained and lack proper insurance and (2) uses owner-operator drivers who lack proper training and are encouraged or required to improperly solicit business. 66 Ground Access to Major Airports by Public Transportation

• Regional coverage/traveler characteristics. The proportion of air travelers whose trip end is near the bus stops/stations is important. The degree of population density and automobile ownership may also influence the use of door-to-door service. For example, the proportion of travelers using shared-ride vans at San Francisco International Airport is much higher than the proportions at Oakland or San Jose airports, perhaps because of the greater population densities and lower automobile ownership rate in San Francisco. What’s Next? Understanding the logic of high market shares requires a case-by-case examination of both systems that are performing well and systems that are performing poorly. Using the most basic measure of performance—market share to public transportation, 46 airports around the world are examined in detail in Chapter 4. Chapter 5 then reviews known ridership impacts of strate- gies to integrate baggage check-in/handling and ticketing across modes. Documentation of the desired attributes of good service is clearly important. However, once documented, the service attributes per se clearly cannot explain the variations in the public trans- portation market, particularly when examined on the basis of the total airport market. A process to document both the geographic distribution of market segments and the demographic distri- bution of market segments is needed to understand the extent to which a given service is succeeding or failing in terms of the market for which it was designed. This task will be addressed in Chapter 6. Attributes of Successful Ground Access Systems 67

Next: Chapter 4 - Public Transportation Market Share by Airport »
Ground Access to Major Airports by Public Transportation Get This Book
×
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

TRB’s Airport Cooperative Research Program (ACRP) Report 4: Ground Access to Major Airports by Public Transportation examines key elements associated with the creation of a six-step market-based strategy for improving the quality of public mode services at U.S. airports. The report also addresses the context for public transportation to major airports, explores the attributes of successful airport ground access systems, presents an airport by airport summary of air traveler ground access mode-share by public transportation services, and more.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!