Orth, H., Nash, A. and U. Weidmann

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29

A LEVEL-BASED APPROACH TO PUBLIC TRANSPORT NETWORK PLANNING Hermann Orth (corresponding author) ETH Zurich, Institute for Transport Planning and Systems (IVT) Wolfgang-Pauli-Strasse 15 8093 Zurich, Switzerland Telephone: +41 44 633 3109 FAX: +41 44 633 1057 [email protected] http://www.ivt.ethz.ch/people/orthh/index_EN Andrew Nash GreenCityStreets.com Bandgasse 21/15 1070 Vienna, Austria Telephone: +43 676 933 0483

[email protected] http://andynash.com/ Prof. Dr. Ulrich Weidmann Telephone: +41 44 633 33 50 ETH Zurich, Institute for Transport Planning and Systems (IVT) [email protected] http://www.ivt.ethz.ch/people/ulrichw/index_EN

Orth, H., Nash, A. and U. Weidmann

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

2

ABSTRACT This paper defines the concept of public transport network level and uses it to describe the performance and future improvement of Zurich’s public transport system. A public transport level is a specific type of service designed to serve a particular market. Service is defined as a combination of vehicles, infrastructure and operating characteristics. A “pure” level is when the service is targeted specifically to one particular market. A “hybrid” level is when a service is targeted to serve several markets. Urban travel is generally described as three markets: short, intermediate and long distance trips. Consequently many cities have developed three-level public transport networks: surface buses and trams to serve short trips, rapid rail to serve intermediate trips, and regional rail to serve longer distance trips. Often, by design or for historic reasons, cities have additional levels, fewer levels and/or the levels that are not precisely matched to their markets. For example, two-level networks are often found in medium size cities. Their advantage is lower costs while their main disadvantage is a mismatch between transport mode and market that manifests itself in capacity limitations. Using the level concept to help analyze and plan public transport service is useful because it focuses attention on matching service qualities to markets. This paper uses this approach to analyze the success of Zurich’s public transport system and to provide a structure for planning improvements that will be needed to meet rapidly increasing public transport demand. This approach could help other cities (re)design their public transport systems to be more attractive and efficient. More specifically, the approach could show how two-level public transport networks could be a viable option for medium sized cities and large cities with dispersed settlement patterns.

Orth, H., Nash, A. and U. Weidmann

1

A LEVEL-BASED APPROACH TO PUBLIC TRANSPORT NETWORK PLANNING

2

1.

3

INTRODUCTION

3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

Large cities and metropolitan areas rely on public transport to provide efficient mobility and reduce emissions. Many large cities provide a three level public transport network consisting of (1) surface transit (buses and trams), (2) rapid transit, and, (3) regional rail services. In contrast, many medium-size cities provide two level public transport networks. For example, the two level public transport network of Zurich, Switzerland, offers a very dense and comprehensive service and is oftentimes seen as model for successful transit (e.g. [1]). In Zurich, public transport has a mode share of 59% and fare revenues cover 64% of costs. [2]. The two level public transport network consists of surface public transport (buses and trams) and regional railway (S-Bahn), there is no rapid transit level. Zurich considered building a third-level, however voters defeated the proposal in 1973 and instead passed two initiatives that enabled the surface transport and S-Bahn to effectively serve the intermediate travel market. The city’s comprehensive public transport priority program made surface transport faster and more reliable so it could serve the shorter intermediate distance trips, and the S-Bahn initiative created an integrated regional rail system that could serve the longer, intermediate distance trips [3]. Zurich’s two-level structure could be an interesting approach for other medium-sized cities and regions with dispersed settlement patterns. The major advantages of two level networks are much lower costs, quicker implementation, and their ability to “grow” ridership. Thus they can be ideal for helping demonstrate that good public transport can support sustainable urban development. The disadvantage of two level networks is that, all other things being equal, by removing one of he levels, they have lower capacity. Therefore they must be very carefully planned to ensure that they can meet existing and projected travel demand. Interestingly, this is exactly the situation Zurich finds itself in today. The region’s two-level network has been so successful that it is quickly running out of capacity and planners are studying ways to meet projected demand. The purpose of this paper is two-fold: (1) to introduce the concept of using levels to help analyze and plan public transport systems, and (2) to illustrate the concept using the example of Zurich. Section 2 of this paper introduces the concept of system levels in public transport networks. Section 3 briefly outlines Zurich’s two-level network structure and its performance. Section 4 describes approaches for addressing growing transport demand. Section 5 presents conclusions and discussion.

31

2.

32 33 34 35 36 37 38 39 40 41 42 43 44 45 46

2.1 Public transport planning for urban areas Public transport service needs to be both comprehensive (i.e., serve all areas with quick access) and provide short trip times to be competitive with automobiles, therefore a combination of different technologies is necessary to build a well-working system [4]. The selection of where to operate what kind of service is driven by transport demand and urban development patterns. Specific criteria to consider include the number and type of trips, resident and working population, settlement density and geographic features. [5]. Finally since access is a key factor defining an area’s attractiveness, good transport planning must not follow development but attempt to steer it by providing better access to locations sensible to develop [6]. A high quality public transport system provides access to a large area within a short time period. The amount of area that can be accessed in a given amount of time (the isoquant) depends on vehicle speed and number of stops. There is a well known tradeoff between stops and speeds: more stops means lower speeds, but more stops also minimize station access time. The general rule is that higher speeds are needed for longer trips, so stations are located farther apart, and vice-versa. This approach has been used to develop different modes of public transport to serve different length trips. Reference [7] presents a systematic approach for selecting the most appropriate public transport modes for different situations used in

DEFINING LEVELS IN PUBLIC TRANSPORT NETWORK STRUCTURE

Orth, H., Nash, A. and U. Weidmann

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47

4

Switzerland. The following section introduces the concept of public transport level and describes how it can be used to help analyze and plan public transport. 2.2 Public Transport Network Levels The urban transportation market can be defined in terms of trip length: short, intermediate and long. A pure approach to serving these markets would be to build a specific service (or “level”) consisting of vehicles, infrastructure and operating strategies tailored to best serve each particular market. The standard approach is to build a three level network consisting of:  Level 1 – surface transit with many stops for short trips (often feeding rapid transit or regional rail stations);  Level 2 – grade separated high capacity rapid transit with closely spaced stops and higher speeds for intermediate trips; and,  Level 3 – moderately high capacity regional rail service operated with widely separated stops for longer distance trips. However, trip length is relative and public transport systems evolve over time. Therefore many regions have developed hybrid networks (i.e., networks that serve more than one market) either by design or historic growth. It is difficult to create a high quality hybrid network because vehicle type, infrastructure and service patterns cannot be specifically targeted to meet the needs of a particular market. A two-level network is, by definition, a hybrid network because the surface transit level and regional rail level must make-up for the missing rapid transit level. For example, the San Francisco region could be characterized as a two-level network since it lacks true rapid rail service. Muni Metro provides underground LRT service in the downtown core, but service can be unreliable (due to surface operation in outer areas) and capacity is low due to the vehicle size. BART functions as rapid transit within the city (closely located stations operating in all stop service), but without the ability to operate express service. It serves long distance trips less well although it reaches far into the suburbs (and functions as regional rail in these areas). In contrast, Munich’s S-Bahn also operates as rapid transit in the city center, but has the ability to operate express trains in outer segments of the lines which makes it a more effective hybrid-level service. Another example are tram-train systems in European cities. These operate as regional rail in the suburbs and as trams in city centers [8]. Their key problem is limited capacity since they must use short vehicles (to fit on downtown streets) and are subject to limited train paths on the regional railway network. Zurich carefully designed its two level network by adjusting the surface-based and regional rail networks to better serve intermediate distance trips. More specifically, Zurich improved the speed and reliability of buses and trams and increased the number of S-Bahn stations serving the city center [3, 9]. Together, these adjustments expanded the area that could be reached within a constant time budget. Figure 1 compares a traditional three-level public transport network to Zurich’s two-level network. Table 1 provides a comparison of public transport systems in several US and European cities including an assessment of the level which each mode of transport serves. The “H” is used to designate modes serving “hybrid” markets (i.e., two different markets). It can be seen that Zurich´s regional rail system is relatively slow in comparison with other cities, but the bus and light rail speeds tend to be higher. This is due to the regional rail system taking on functions of a traditional rapid transit system. It is also important to note that effective frequency is higher than shown since several lines operate on the same tracks in the central area. Together, the surface-based and S-Bahn services are capable of serving rapidtransit type trips to some extent. Not shown in the table are reliability levels, however studies in Zurich have shown that despite dense traffic, the existing prioritization has secured a relatively high level of reliability [10] – this is another key element needed to make a system without grade-separated mid-tier work.

Orth, H., Nash, A. and U. Weidmann

5

Three-Level Public Transport Network

regional rail line

regional rail station rapid transit line rapid transit station area accessible within time budget by walking from regional rail area accessible within time budget by walking from rapid transit area accessible within time budget by conventional surface public transport

Two-Level Public Transport Network

regional rail line

regional rail station rapid transit line rapid transit station area accessible within time budget by walking from regional rail area accessible within time budget by walking from rapid transit area accessible within time budget by conventional surface public transport

area accessible within time budget by upgraded surface public transport

limited additional benefit of a three-level network

1 2 3

FIGURE 1: Comparison of 2- and 3-level public transport networks

Orth, H., Nash, A. and U. Weidmann

1

6

TABLE 1: Comparison of Public Transportation Systems in selected US and European Cities City

City Metro. Area Pop.(mil.) Pop. (mil.) 3

Karlsruhe

0.3

1.3

Munich

1.4

5.6

Vienna

1.8

2.6

Zurich

0.4

1.9

Chicago

2.7

9.7

San Francisco

0.8

4.5

Philadelphia

1.5

6.0

Boston

0.6

4.6

Mode/Level Bus 1 Light Rail 1 Tram-Train H4 S-Bahn 3 Bus 1 Light Rail 1 U-Bahn 2 S-Bahn H4 Regional Rail 3 Bus 1 Light Rail 1 Rapid Transit 2 Regional Rail 3 Bus H4 Light Rail H4 Regional Rail H4 Bus 1 Rapid Transit 2 Regional Rail 3 Bus 1 Light Rail (Muni Metro) H4 Rapid Transit (BART) H4 Regional Rail (Caltrain) 3 Bus 1 Light Rail 1 Rapid Transit 2 Regional Rail 3 Bus 1 Light Rail 1 Rapid Transit 2 Regional Rail 3

System Speed Line headways2 Grade separation / [km/h] [min] dedicated infrastructure 25 10-30 No/18 10-20 Partial/35 10-60 Partial/Partial 55 20-30 Full/Partial 18 6-10 No/19 7-20 Partial/36 2.5 Full/Yes 45 10-20 Full/Partial 75 30-120 Full/No 17 6-15 None/15 5-7 Partial/32 3-10 Full/Yes 45 10-60 Full/Partial 18 7.5 No/15-20 7.5 Partial/50 30 Full/No 14 5-20 No/35 3-15 Full/Yes 70 20-1201 Full/Partial 14 5-15 No/15-27 5-12 Partial/56 5-15 Full/Yes 1 65 30-60 Full/Partial 15 4-60 None/15 6-15 Partial/30 2-20 Full/Yes 45 30-60 Full/Partial 18 None/20 5-15 Partial/25 5-10 Full/Yes 50 20-1201 Full/Partial

1

Not regular throughout day, some lines with only few runs per day or gaps during midday During core service period 3 Service area residents 4 Hybrid service with two level´s service characteristics 2

1 km/h = 0.62 mph Based on timetable data from operators, service maps and government statistical data: Online timetable of the Zürcher Verkehrsverbund, www.zvv.ch, accessed 22.07.2014 Kanton Zürich (2014) Statistisches Jahrbuch des Kantons Zürich 2014. Online timetable of the Chicago Transit Authority, www.transitchicago.com, accessed 23.07.2014 Online timetable of the Commuter Rail Division of the Regional Transportation Authority (Metra), www.metrarail.com, accessed 23.07.2014 Annual Estimates of the Resident Population 2013, U.S. Census Bureau, http://factfinder2.census.gov/faces/nav/jsf/pages/index.xhtml, accessed 24.07.2014 Online timetable of the Münchner Verkehrsgesellschaft mbH, http://www.mvg-mobil.de/fahrplaene/index.html, accessed 23.07.2014 Bavarian State Office for Statistics and Data Processing, population data, https://www.statistikdaten.bayern.de, accessed 23.07.2014 Online timetable of the San Francisco Municipal Transportation Agency http://www.sfmta.com/, accessed 24.07.2014 Online timetable of Caltrain http://www.caltrain.com, accessed 24.07.2014 Online timetable of the Southeastern Pennsylvania Transportation Authority (SEPTA), http://www.septa.org/around/, accessed 24.07.2014 Delaware Valley Regional Planning Commission, Speeding Up SEPTA, Philadelphia, USA, 2008. [11] Online timetable of the Massachusetts Bay Transportation Authority (MBTA), http://www.mbta.com/schedules_and_maps/, accessed 24.07.2014 Online timetable of the Wiener Linien, http://www.wienerlinien.at/eportal2/, accessed 24.07.2014 Online timetable of the Karlsruher Verkehrsverbund, www.kvv.ch, accessed 27.07.2014

2

Orth, H., Nash, A. and U. Weidmann

7

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

2.4 Capacity: The Achilles Heel of Two-Level Public Transport Networks The key problem faced by two-level public transport networks is that they do not match transport demand as well as a three-level network. Rapid transit is specifically designed to serve large passenger volumes quickly and efficiently. Serving these trips with trams or modified regional rail can be inefficient and can have negative impacts on system-wide capacity at high levels of demand. For example, Muni Metro is limited by low capacity vehicles and unreliable service. Karlsruhe’s tram-train approach uses train paths that could be used by higher capacity trains not subject to the length restrictions of inner city networks. In Zurich, increasing congestion is reducing the speed of surface-based public transport and increasing ridership is reducing the S-Bahn’s reliability and attractiveness. In all three cases capacity problems are created for passengers making long- and intermediate-distance trips. Zurich’s two-level network has performed well for the last 25 years, however projections for increased transport demand mean the region must consider new approaches for improving its system. The next section describes Zurich’s two-level network in more detail. It is followed by an analysis of potential improvements based on the public transport network level framework.

17

3.

PUBLIC TRANSPORT IN ZURICH

3.1

Introduction

18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46

Zurich is Switzerland’s largest city with population of 400,000 and a metropolitan area of 1.4 million inhabitants. There are a total of 380,000 jobs in the city proper. The settlement density is 4,337 persons/km² in the city and no more than a fifth of that in the surrounding region [12,13]. In- and outbound commuting amounts to 213,000 and 59,000 persons every day. The daily ridership on Zurich’s bus/tram network is 815,000 trips with an additional 446,000 trips by train into and out of the city. Public transport has a mode share of 59% within the city and 36% in the metropolitan area [12,13]. The mode shares are especially notable since average income in the Zurich area is high at an annual 92,200 USD [14]. The high quality of public transport is also reflected in the relatively low automobile ownership rate of 498 vehicles per 1000 persons [15]. 3.2 History of Public Transport Policy in Zurich As outlined above, the Zurich region has a two-level public transport network. The system was designed to provide fast, frequent and reliable surface-based buses and trams for short and intermediate distance trips and a dense regional railway network for intermediate and long distance trips. Transportation policy in Zurich has been significantly influenced by Switzerland’s direct democracy process in which citizens vote on all major capital expenditures and policy questions. Public transportation has been on the ballot often: In the late 1950s and early 1960s Zurich citizens rejected plans to place the tram system underground and in 1973 rejected plans for U-Bahn (rapid transit) and S-Bahn (regional rail) systems. However, in 1986 voters approved a citywide public transport priority program and construction of the S-Bahn. In the city, the comprehensive application of public transport priority improvements including traffic signal pre-emption at almost all intersections and extensive use of separate right of ways led to the development of a very high quality bus and tram network [3]. The S-Bahn initiative´s key infrastructure was a tunnel under the city and a through station at the main railway station designed to link previously unconnected regional services into an integrated network. The improved system opened in 1990 providing significantly increased capacity, reliability and speed,

Orth, H., Nash, A. and U. Weidmann

1 2 3 4 5 6 7 8 9 10 11

8

making it possible to provide fast and comfortable services into the surrounding suburbs and for key destinations within the city. As shown in Figure 2, the S-Bahn has been a great success: Between 1990 and 2013, ridership grew by an average of 4.4% per year, with a total growth of 169% [2] compared to population growth of only 12% in the city and 26% in the metropolitan area. The ridership of the urban bus and tram network declined in the years after the S-Bahn introduction but have recovered and, compared to 1990, grew about as much as the city population.. This indicates that many commuter used the new S-Bahn service instead of surface public transport for intra-city trips as it offered a more attractive service. While surface network accommodated the growth of the city population, it does not offer travel times that are comparable to those of the S-Bahn network which thus had to handle the bulk of the additional trips.

Regional rail passenger numbers and population growth

Growth, indexed against 1990

280% 260% 240% 220% 200% 180% 160% 140%

120% 100% 80% 1990

1995 Regional Rail Passengers

2000

2005

Year

City Population

2010

Metropolitan Area Population Bus/Tram Passengers

Metropolitan area definition of 2014: Die Raumgliederungen der Schweiz 2014, Swiss Federal Statistics Office, Neuchâtel, Switzerland, 2014 Community Population Data: Bilanz der ständigen Wohnbevölkerung nach Bezirken und Gemeinden, Swiss Federal Statistics Office, Neuchâtel, Switzerland, 2014 Canton of Zurich Population Data: Statistisches Jahrbuch des Kantons Zürich 2014, Kanton Zürich, 2014

12 13 14 15 16 17 18 19 20 21 22 23 24

FIGURE 2: – Zurich S-Bahn Ridership Growth: 1990 – 2012 The S-Bahn initiative also called for creation of a regional public transport agency, the Zürcher Verkehrsverbund (ZVV). It was made responsible for planning and coordinating all regional public transport services, administering a common fare system and highly integrated scheduling to minimize transfer wait times. Since the ZVV collects fares and contracts with transport companies to operate services, it has considerable influence on the service provision. In addition, Zurich also implemented several programs to curb the growth of automobile traffic in the city center since the city’s historic layout is not able to provide sufficient space to handle that traffic and still provide space for dedicated tram lines and a livable urban environment. Specific measures include a cap on the number of parking spaces and a road capacity and traffic control strategy to let no more vehicles into the city center than can move about without causing a traffic flow breakdown.

Orth, H., Nash, A. and U. Weidmann

1 2 3 4 5 6 7

3.3 Performance of Zurich’s Two-Level Public Transport Network Zurich’s public transport system is easy to use, very accessible, and generally still provides high quality service (see e.g. [10]). However, increasing demand is reducing service quality on both the surface-based and S-Bahn network levels. This section analyzes the current performance of Zurich’s two-level network. Table 2 presents the operating characteristics and Table 3 presents the vehicles operated. TABLE 2: Zurich Public Transport System Characteristics Avg. Service Speed [km/h]

Level 1

1

Freq. per Median stop Grade line/per spacing [m] Sep. segment [min]

Line capacity [P/h]

15

7.5/1.2

365 Partial 1600-6500 105-210

Bus

18

7.5/2.5

390 None

500-2000

50-130

LRT

20

7.5-15/6.7

640 Full

1600-3200

200

50

30/4

2600 Full 1000-18000 490-1160

Capacity is set as all seats plus standees at 3 persons per m

656.67

36.54

2106

22.1

2

TABLE 3: Zurich Rolling Stock Level

1

Type and Characteristics

Capacity1 Length [m]

Low floor, 3-door standard bus

50-60

12

Low floor, 4-door articulated Bus

75-100

18

130

25

Low floor, 5-door double articulated bus

10 11

Vehicle Pass.- km Veh.- km capacity1 traveled traveled [P] [mil. km] [mil. km]

Tram, inner city

Level 3 S-Bahn

8 9

9

Orth, H., Nash, A. and U. Weidmann

1

10

“Tram 2000” High floor, 4-door trams , normally run as double-units

105

21

“Cobra” Low floor, 7-door trams

200

36

“DPZ” Partially low floor, 2-door per railcar/side push-pull units with locomotive and control cab on at 490 ends, can be run in 2-3 unit compositions

3

100

“DTZ” Low floor, 2-door per railcar/side 560 electrical multiple unit, can be run in 2-3 unit compositions

100

“KISS” Low floor, 2-door per railcar/side 1160 electrical multiple units, can be run in 2 unit compositions

150

1

Capacity is understood as all seats plus standees at 3 persons per m2 Vehicle Information: Swiss Federal Railways SBB CFF FFS, Verkehrsbetriebe Zürich VBZ Images: Hermann Orth

1 2 3 4 5 6

Level 1 Public Transport Network – Trams and Buses The bus and tram network in the city of Zurich is largely a radial network. In the city center, the routes effectively form a circle around the small, but dense, core area. There are also a few tangential lines that mainly connect outlying S-Bahn stations. In addition to the trams and buses operated within the city limits of Zurich, a new light rail line was built in the Glattal region in the late 1990s. This line operates like a tram

Orth, H., Nash, A. and U. Weidmann

1 2 3 4 5 6 7 8 9 10 11

12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

11

in the city center and with wider station spacing and exclusive right-of-way in the suburban region [16]. This line can be considered a further refinement of the two-level network. Zurich’s public transport prioritization program has helped ensure that speeds and reliability are relatively high. However, Table 2 shows clearly that level 1 network operating speeds are far below those on the level 3 network. The tram network’s relatively low speeds result from frequent stops and heavy traffic in the center city. These problems will worsen as transport demand increases ridership and competition for scarce street space. Recent data supports the conclusion that both speed and reliability (ontime performance) of the public transport system have been falling in recent years [17,18]. Figure 3 shows recorded utilization on bus and tram services. The rush hour periods show very high loads, considering that the 100% capacity includes standees at 4 P/m2. This indicates insufficient service levels when referenced with the Transit Capacity and Quality of Service Manual [19].

FIGURE 3: Utilization of bus and tram services in the city of Zurich In summary, Zurich’s level 1 network is approaching its capacity limits and service quality is falling, even though it may still be attractive by international standards. Unfortunately, it will be difficult to significantly increase capacity on the inner-city level 1 network since almost all routes are already served by the largest possible vehicles and frequencies are at the limit of what can be operated reliably (since several lines operate over the same segments). Level 3 Public Transport Network: S-Bahn As shown in Table 2, the level 3 network operates at much higher speeds and larger stop spacing than the level 1 network. Almost all S-Bahn lines converge on Zurich´s main railway station and the median stop spacing declines to 2.3 km from 3.0 km in the central area. All S-Bahn stations are very well connected to the level 1 network. This design provides multiple transfer points to achieve better travel times and not to overload the very highly used main station. The S-Bahn operates all day service from early in the morning until late at night with trains calling on stations at clockface headways (the same times every hour: e.g., 6:33, 7:33, etc.). Most of the lines operate with a frequency of every 30-minutes. Since multiple lines operate on the central sections, effective frequency at these stations is much higher.

Orth, H., Nash, A. and U. Weidmann

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26

As shown in Figure 2, ridership on the S-Bahn has increased by approximately 169% since 1990. The S-Bahn has implemented a continuous program of capacity improvements designed to serve this demand including:  Dedicated double deck rolling stock – in 1990 the region purchased double deck rolling stock with higher capacity and wider doors (“DPZ” on Table 3) than the previously used standard longdistance railcars.  Bahn 2000 – The Swiss Bahn 2000 program was a set of capacity improvements designed to transform the national railway system in to a highly regular service following timed-transfer principles at clockface headways (i.e., fixed interval timetable, or Taktfahrplan) [21]. This significantly increased overall rail use.  Low-floor rolling stock – low-floor multiple unit “DTZ” and “KISS” rolling stock which started being introduced in 2007 feature multiple-unit propulsion (eliminating the dead zone created by a locomotive in the earlier trains) [9], level-boarding and higher capacity than the older cars.  Durchmesserlinie – Löwenstrasse station – in June 2014, the SBB opened a second tunnel under the city and underground through-station under the main station. This project was approved through a public vote in 2006 and cost approximately 2 billion CHF (approximately 2.2 billion USD). It will provide significantly increased capacity for S-Bahn and national trains. [2] However, in spite of these significant improvements, growing demand is pushing the S-Bahn system to its limits (Figure 4). Passenger loads are very high in the city center but drop rapidly with increasing distance. This means that the use of the S-Bahn for intermediate distance requires a high capacity that is only utilized for a small section of a line while the costly rolling stock runs at very low utilizations for long stretches. This can be observed for the whole network (Figure 4a) and is very prominent for some individual lines (Figure 4b). This means that some lines need to operate 2-3 DPZ/DTZ units or a 1-2 KISS-units to meet inner segment demand, however run half full or less for most of the route. a)

27 28 29 30

12

b)

FIGURE 4: Passenger loads on S-Bahn services over distance from Zurich HB

Orth, H., Nash, A. and U. Weidmann

13

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Two-Level Network Performance Analysis As the surface-based public transport network (level 1) becomes slower and less reliable, it is not attractive enough to capture a large share growing number of trips and passengers shift to the S-Bahn (level 3) for short trips within the city or to the inner suburbs. This results in high peak loads in central Zurich (Figure 4). An analysis shows that the average occupancy per train across the whole day is 180 on line segments within 10 km of the Zurich HB compared to only 85 for segments beyond 10 km. Similarly, the 95% quantiles for occupancy are 515 passengers and 395 passengers, respectively. In addition to high passenger loads, center area S-Bahn stations have very high passenger exchange volumes and therefore higher dwell times. Together with reduced stop spacings, this results in a lower average service speed of 49 km/h in the center areas vs. 54 km/h outside. Furthermore, the current double-deck rolling stock with two doors per side is not efficient for the heavy transferring in inner segments. In summary, the shift of passengers to the S-Bahn for intermediate distance trips has led to inefficiencies in the service, reducing trip quality for travellers from further away areas. While Zurich’s public transport system is operating relatively well, continuing growth may increase problems and the region’s public transport system may become a victim of its own success.

19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48

This section outlines several conceptual solutions for meeting Zurich’s growing transport demand in the context of the level-based public transport network structure.

4.

MOVING AHEAD: A THIRD LEVEL OR GRADUAL ADAPTATION?

4.1 Increase active transport mode share This measure consists of increasing the share of people biking or walking (modes that might be characterized as “level 0”) to reduce demand on both surface-based public transport and short S-Bahn trips. A recent study shows that users significantly reduced their short bus and Metro trips after joining the Washington D.C. bike share program. [22]. This approach is being pursued directly in several European cities (e.g., Amsterdam and Copenhagen) where it is not economically feasible to expand Metro systems to meet projected transport demand. Instead they are encouraging active transport by reallocating street space from automobiles to pedestrians and bicyclists, helping create safer and more pleasant walkways and bike lanes [22]. Reallocating street space is a significant political issue in all cities although there is growing evidence that providing more space for pedestrians, bikes and public transport has positive economic, transport and environmental benefits [23]. In Zurich there is a special aspect of this problem: A large share of local streets is currently dedicated for public transport. This creates a conflict between public transport and those seeking more space for biking and walking. In short, reallocation of street space, even where it makes sense rationally is a highly sensitive and political decision. 4.2 Improve surface-based public transport This measure consists of increasing the capacity and speed of surface-based public transport to reduce demand on the S-Bahn for intermediate distance trips. However, as outlined above, Zurich has almost exhausted the potential to increase vehicle size and frequencies. The other option, increasing public transport speed and reliability, would require implementing additional measures to reduce automobile traffic in the center city. Again this would be a complex political process. Given the political complexity of reallocating street space in the short term and the limited potential of marginal upgrades of the current surface-based public transport system, planners are considering several major alternatives. The next two sections outline these in the context of the gap between markets served between the level 1 and level 3 systems. This gap can be reduced from above by changing the S-Bahn or below by enhancing the tram network.

Orth, H., Nash, A. and U. Weidmann

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48

14

4.3 Reducing the gap from above: S-Bahn Service and Capital Improvements One way of addressing the mismatch between S-Bahn service and demand is to better differentiate between intermediate-length center city and longer-distance regional trips by developing a so-called inner and an outer S-Bahn [25]. The approach would create two different service types operating on the same tracks: a frequent high capacity service on the central network, and a fast lower capacity service that operates on the outer sections of the network. The outer S-Bahn would serve communities in the Zurich agglomeration and would operate similar to today from lower density areas to the fringe of the urban areas. Once they reach these stations they would operate as express service to the city center. This would reduce travel times from the outer areas and the service could use the current double-deck rolling stock. Since this pattern would not serve intermediate length center city trips, passenger loads would be more even, providing improved vehicle utilization. It would also free platform capacity at city center stations. In contrast, the inner S-Bahn would be precisely tailored to serve intermediate center city trips. Trains would stop at all stations in the central area and terminate at stations on the urban fringe. They would operate at higher frequencies and use new rolling stock with a single level, many doors and fast acceleration. This approach splits the level 2 network into an outer section operating similar to a traditional regional rail system, and an inner section operating rapid rail type service within the city and immediately adjacent suburbs. Since the inner S-Bahn would largely run along existing tracks, it could be relatively economical to implement, requiring only localized infrastructure extensions at junctions and stations. The challenge herein is how to preserve the regular intervals because different service types need to be accommodated on the same infrastructure. 4.4 Reducing the gap from below: Upgrading the tram network Another concept being studied is the so-called “Metrotram”. This consists of building a set of tunnels in the central city to serve enhanced tram lines and has been the subject of intense discussion in Zurich [26]. The tunnels would lift the tram vehicle size constraints posed by the narrow city center and enable the use of higher capacity trams. They would use surface routes on the outskirts of the city and then enter tunnels to serve the main corridors in the central city. The benefits would affect the whole tram network since it would address the segments that are currently the slowest and most limiting ones. Some of Zurich’s existing tram lines (operating with performance characteristics similar to streetcars or buses) would be replaced with higher performance LRT type service. In many cases, trams operating on the outer right-of-way sections of Zurich’s tram network already operate at LRT levels, although vehicle size limitations in the center city reduce capacity. This type of high quality service would be attractive for intermediate length trips within the central city. The drawbacks of this alternative are that its main effects are limited to trips within the central city, and it would be very expensive and difficult to build due to the considerable tunneling required. It would also be a complex political decision since Zurich voted against similar ideas for placing trams underground in the 1950s, 1962 and 1973 [26]. 4.5 Public Transport Network Level Analysis of Proposed Solutions It is constructive to analyze the different solutions in terms of the public transport network level structure used in this paper. Figure 5 illustrates the main solutions in terms of the trips that are served attractively (i.e. a substitute for travel speed). The first alternative, improving the S-Bahn by creating an inner and an outer service strategy, can be more readily implemented than building the Metrotram. The improved S-Bahn will also provide improved service to the outer region where there is a greater risk that travelers will shift to automobiles, thus providing benefits where they are needed more urgently. It will also improve service within the city, although to a lesser extent than building the new Metrotram. Interestingly, since the two approaches aim at different

Orth, H., Nash, A. and U. Weidmann

1 2 3 4 5 6 7 8 9 10 11 12 13

15

market segments, they are not necessarily redundant. The Metrotram could be considered as a (major) improvement to Zurich’s existing level one network over a longer implementation timeframe. Interestingly, the result of implementing both alternatives would be a system that operates at four service levels: buses as local circulators, light rail/LRT for intermediate length trips within the city, a metrotype heavy rail system for longer trips within the city and inner suburbs, and a regional rail system serving the more outlying suburbs. Compared to a traditional three level network, this would offer four distinct products, however use only two infrastructure systems. This would reduce costs and enable development to be undertaken step by step, thus keeping up with growth and avoiding costly leaps that may be neither necessary nor affordable for smaller metropolitan areas. In summary, a two-level public transport system is unlikely to provide sufficient capacity in very large cities due to complexity of sharing infrastructure. However, Zurich’s experience to date and the possibilities to extend its two-level network to provide additional capacity, show that it is a viable option for medium-sized cities and large cities with moderately dense settlement patterns. a) initial two-level system S-Bahn

level 3

Tram level 1

Bus Cycling „level 0"

Walking trip length served attractively gap

b) two-level system reaching its limits

S-Bahn

level 3

regional rail rolling stock not suited for short trips when crowded

loss in service speed due to crowding and longer passenger exchange times

Tram level 1

loss in service speed due increased traffic and crowding as capacity cannot keep up

Bus

Cycling „level 0"

Walking

trip length served attractively

c) upgraded two-level system

inner S-Bahn: high frequency and high capacity rolling stockoptimized for short passenger exchange times

level 3

outer S-Bahn: faster service on regional rail rolling stock increases area served attractively

S-Bahn Larger stop spacings Metrotram: Faster service and higher capacity units

Tram level 1

Bus

Cycling „level 0"

Walking

attractiveness increases gap closed

trip length served attractively

14 15

Level 3: Regional rail systems Level 2 (missing): Rapid transit Level 1: Surface public transport (bus and tram) Level 0: Non-motorized and access modes (wlaking, cycling)

FIGURE 5: Concepts for transforming Zurich´s public transport system

Orth, H., Nash, A. and U. Weidmann

1

5

16

SUMMARY AND CONCLUSIONS

2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28

This paper introduces the concept of levels in public transport networks and outlines how it can be used to help plan improved service. The concept is used to show how the function of rapid transit has been fulfilled to a large part by upgraded surface transport and modified regional rail services in the case of Zurich. The paper also shows how the concept can be used in Zurich to address the region’s growing capacity problems. The paper outlines several solutions for addressing this capacity problem including one which introduces elements of rapid transit into the surface transport system and one which introduces a new rapid transit type service to the regional rail system. While both approaches are limited in capacity when compared to traditional rapid transit systems, they are interesting options for cities considering rapid transit systems because they are less expensive and can be implemented more easily. When attempting to transfer these findings, several points need to be kept in mind. First, there seems to be a strong predisposition to use public transport in Switzerland. This mindset cannot be recreated quickly. Also, the types of improvements described in this paper require that the regional rail operator have full control of the infrastructure and operations. This is not always the case, especially in American cities. Lastly, Zurich’s two level network was achieved using a highly integrated planning process. It is critical that all involved entities cooperate closely since timed transfer, service complementarity and common ticketing are crucial elements for success. In summary, two level public transport networks may be able to provide a similar quality service level to three level networks with full rapid transit networks. Since the cost of a two level system is lower, this may be way to offer a service level otherwise unattainable. Further differentiation of such a system helps the public transport network to keep up with growth through incremental changes. Improvements at the regional rail network level can be especially attractive since they may require only the use of better tailored rolling stock (at acceptable cost) to quickly serve large increases in demand. More significant improvements, such as upgrades to the surface-based trams would be necessary as demand continues to grow. While two level networks might not be appropriate for a large and dense metropolis, they may very well be a viable option for providing high quality public transport service to medium sized cities and large cities with dispersed development patterns.

29

ACKNOWLEDGEMENTS

30 31

The data used in study was provided by Verkehrsbetriebe Zürich (VBZ) and the Zürcher Verkehrsverbund (ZVV) for which the researchers are highly grateful.

32

REFERENCES

33 34 35 36 37 38 39 40 41 42 43 44

1. 2. 3. 4. 5. 6.

Cervero, R. The Transit Metropolis. Island Press, Washington, D.C., 1998 ZVV Zürcher Verkehrsverbund, Geschäftsbericht 2013, ZVV Zürcher Verkehrsverbund, Zurich, Switzerland, 2014. Nash, A., 2003. Implementing Zurich’s Transit Priority Program. Transportation Research Record 1835, 59–65 V.R. Vuchic, Urban Transit Systems and Technology, John Wiley & Sons, Inc., Hoboken, NJ, USA, 2007. V.R. Vuchic, Urban transit: operations, planning and economics, J. Wiley & Sons, Hoboken, N.J, 2005. W.W. Millar, Public Transportation Today and Tomorrow: 2012 Thomas B. Deen Distinguished Lecturer, Transportation Research Record: Journal of the Transportation Research Board. 2274 (2012) 5–11.

Orth, H., Nash, A. and U. Weidmann

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

7.

8.

9.

10.

11. 12. 13. 14. 15.

16.

17. 18.

19. 20.

21.

22. 23.

17

U. Weidmann, R. Dorbritz, H. Orth, M. Scherer, P. Spacek, Einsatzbereiche verschiedener Verkehrsmittel in Agglomerationen, Eidgenössisches Departement für Umwelt, Verkehr, Energie und Kommunikation UVEK, Berne, Switzerland, 2011. L. Naegeli, U. Weidmann, A. Nash, Checklist for Successful Application of Tram-Train Systems in Europe, Transportation Research Record: Journal of the Transportation Research Board. 2275 (2012) 39–48. Nash, A., Weidmann, U., Bollinger, S., Luethi, M., Buchmueller, S., 2006. Commuter Rail: Increasing Schedule Reliability on the S-Bahn in Zurich, Switzerland: Computer Analysis and Simulation. Transportation Research Record 1955, 16–25 H. Orth, N. Carrasco, M. Schwertner, U. Weidmann, Calibration of a Public Transport Performance Measurement System for Switzerland, Transportation Research Record: Journal of the Transportation Research Board. 2351 (2013) 104–114. Delaware Valley Regional Planning Commission, Speeding Up SEPTA, Philadelphia, USA, 2008. City of Zurich, Statistical Yearbook of the City of Zurich 2013 (in German), Zurich, Switzerland, 2014. Kanton Zürich, Statistisches Jahrbuch des Kantons Zürich 2014, Zurich, 2014. Swiss Federal Statistics Office, Monthly gross salary by region and economic sector, admin.ch/bfs/portal/de/index/themen/03/04/blank/data/01/06_02.html, accessed 21. July 2014 Swiss Federal Statistics Office, Motorisierungsgrad, http://www.bfs.admin.ch/bfs/portal/de/index/themen/11/03/blank/02/01/01.html, accessed 21. July 2014. H. Schneebeli, A. Nash, M. Scherer, Glattalbahn: Innovative Transport Solution for Suburban Areas, Transportation Research Record: Journal of the Transportation Research Board. 2146 (2010) 1–9. G. Schoch, Koexistenz und Kooperation im Stadtverkehr: Kampf um die knappen Ressourcen Raum und Zeit - Die Sicht der Verkehrsunternehmung, (2011). L. Naegeli, H. Orth, U. Weidmann, A. Nash, High-Quality Public Transport and Promotion of Nonmotorized Transport—Compromise or Complement?: An Analytical Approach for Assessing Conflicts, Transportation Research Record: Journal of the Transportation Research Board. 2350 (2013) 26–36. Transportation Research Board. Transit Capacity and Quality of Service Manual, Washington, D.C., 2003 U. Weidmann, H. Orth, R. Dorbritz, M. Schwertner, N. Carrasco, Verkehrsqualität und Leistungsfähigkeit des strassengebundenen ÖV, Eidgenössisches Departement für Umwelt, Verkehr, Energie und Kommunikation UVEK, Berne, Switzerland, 2013. R. Maxwell, Intercity Rail Fixed-Interval, Timed-Transfer, Multihub System: Applicability of the Integraler Taktfahrplan Strategy to North America, Transportation Research Record. 1691 (1999) 1–11. Martin, Elliot W., and Susan A. Shaheen; Evaluating public transit modal shift dynamics in response to bikesharing: a tale of two U.S. cities; Journal of Transport Geography, forthcoming. City of Copenhagen; Action Plan for Green Mobility; available at: http://kk.sites.itera.dk/apps/kk_pub2/pdf/1123_dM2NAXVaGm.pdf accessed 30 July 2014.

Orth, H., Nash, A. and U. Weidmann

1 2 3 4 5

24.

25. 26.

New York City Department of Transportation; The economic benefits of sustainable streets; 2013; available at: http://www.nyc.gov/html/dot/downloads/pdf/dot-economic-benefits-of-sustainablestreets.pdf accessed 30 July 2014. ZVV Zürcher Verkehrsverbund, Strategie 2016-2019, Zurich, Switzerland, 2013. U. Weidmann, W. Huber, Metro/tram - Eine Perspektive für Zürich, Zurich, Switzerland, 2012.

18