Scholarly article on topic 'Structural comparison and enlightenment for regional rails in metropolitan areas'

Structural comparison and enlightenment for regional rails in metropolitan areas Academic research paper on "Social and economic geography"

CC BY-NC-ND
0
0
Share paper
Academic journal
Transportation Research Procedia
OECD Field of science
Keywords
{"Metropolitan area" / "Regional rail" / "Resident-employment distribution" / "Comparison study"}

Abstract of research paper on Social and economic geography, author of scientific article — Jiang Jie, Song Jiahua, Shao Yuan

Abstract Based on the division of zones in metropolitan areas, the differences in the scale, functions, passage and hubs between Tokyo, New York and Paris are compared and it is indicated that three cities have different kinds of railway network structure. Considering spatial structure of cities and the historical process, this paper analyzes the influence of resident-employment distribution and the development background on the planning of regional rails. We also explain how these three kinds of structure were formed and summarize technical essentials of the planning of regional rails in international metropolitan areas, which would provide a reference for the planning and the construction of regional rails in China.

Academic research paper on topic "Structural comparison and enlightenment for regional rails in metropolitan areas"

Available online at www.sciencedirect.com

ScienceDirect

Transportation Research

Procedía

ELSEVIER

Transportation Research Procedía 25C (2017) 2984-2993

www.elsevier.com/locate/procedia

World Conference on Transport Research - WCTR 2016 Shanghai. 10-15 July 2016

Structural comparison and enlightenment for regional rails in

metropolitan areas

Jiang Jie song Jianua , snao Yuan

"Shenzhen Urban Transport Planning Center,518021, China

Abstract

Based on the division of zones in metropolitan areas, the differences in the scale, functions, passage and hubs between Tokyo, New York and Paris are compared and it is indicated that three cities have different kinds of railway network structure. Considering spatial structure of cities and the historical process, this paper analyzes the influence of resident-employment distribution and the development background on the planning of regional rails. We also explain how these three kinds of structure were formed and summarize technical essentials of the planning of regional rails in international metropolitan areas, which would provide a reference for the planning and the construction of regional rails in China.

© 2017 The Authors Published by Elsevier B.V.

Peer-review under re^fjiKibmty °f WORLD CONFERENCE ON TRANSPORT RESEARCH SOCIETY

Keywords: MetropoHtan area; Regional raH; Resident-employment dktril^ution; Comparison study

1. Infroduction

As Chmese major cities are gradually tecom^ metlopolitabs, cities' commuti^ area are extend^ to Ithe oultside of 1raditiobal heb1ral hitiel, which plolobgl Ithe hommutibg time. The legiobal rati is hOblideled to be Ithe effihiebt

transport mode to improve the travel condition of a region and could shorten commuter time and improve

tlablpoltatiob ltluhtule (Gu ^onan and Guo Chabggobg,2010; Zhang Xmochun et al.,2012). The plabbibg of legiobal

rails has been emphasized in the new wave of railway network planning in nearly all major cities, but many key

tehhbihal plOhedulel of the plabbibg are ltill hObtlovellial. In this papel, three mtematiobal me^'o^titon areas are

* Corresponding auth°r. Tel.: +0-000-000-0000 ; fax: +0-000-000-0000 .

E-mail address: jiangjie722@126.com

2352-1465 © 2017 The Authors. Published by Elsevier B.V.

Peer-review under responsibility of WORLD CONFERENCE ON TRANSPORT RESEARCH SOCIETY. 10.1016/j.trpro.2017.05.208

taken as examples to analyze and summarize technical essentials of regional rail planning, which could provide a reference for the planning and the construction of regional rails in China.

2. Zones division in metropolitan areas

It is the goal of transportation system to move persons or goods spatially at the expense of time and money (Merlin, P, 1992). The birth and the growth of an urban rail system is the product of the combined effect of horizontal and vertical development of a city, so it is the precondition of planning the railway network to analyze the spatial structure of a city.

Metropolitans are always divided into three urban layers in related research in China: the central area, the urban area and the metropolitan area (Wu Xueming, 2003; Wang Wenjing and Lu Huapu, 2013). It is indicated in numerous research of commuting travel in metropolitans that the scope of commuting area is always beyond the boundary of city's administrative area but within the scope of the metropolitan area. Such as the maximum distance of the metropolitan area in Tokyo from the city center is approximate 110 kilometers, while the scope of commuting area is approximate 40-50 kilometers which is much bigger than the center area. The metropolitan area of New York and Paris also share similar characteristics with Tokyo. For the convenience of further comparison, the commuting area was added between the urban area and the metropolitan area, thus four zones were divided in Tokyo, New York and Paris as shown in Fig. 1 ~ Fig. 3. Table 1(a,b,c) gives statistics on surface area, population and employment of each area in three cities.

• Central Area (Zone 1): It is the central business district of the city which covers only between 20 km2 and 40 km2 and contains between 180,000 and 600,000 inhabitants. The number of people in paid employment is as high as 1~2 million, which shows the characteristics of high concentration of posts.

• Urban Area (Zone 1&2): It is the city center and outside highly built-up area. The surface area is approximate 600~800 km2 (it is approximate 15~20 km away from the central area). The urban area contains between 6 and 8 million inhabitants (its population density is about 10 000 persons/km2) and it is generally a specific administrative area, for example Tokyo Metropolis, New York City and Little Crown. The Zone 2 is one of the main residence for people work in the central area. Providing somejobs, but the density of posts in Zone 2 is less than that in Zone 1.

• Commuting Area (Zone 1,2&3): It contains the urban area and the suburban area which cover approximate 2000~6000 km2 (it is approximate 30~50 km away from the central area). The population is between 10 million and 20 million. It is the origin of most commuting travels to urban areas and their behaviors are not limited by the administrative boundary. Compared to the urban area, the post density in the Zone 3 (suburban area) decreases significantly. Meanwhile, its population density is also lower than the Zone2.

• Metropolitan Area (Zone 1,2,3&4): The Zone 4 (the exurb) is the outermost area of the metropolitan. Its commuting relationship with the urban area is significantly weakened and it keeps a general balance between the residence and the employment.

The above-mentioned division of zones in metropolitans reference the book "The Four World Cities Transport Study" which is the research of the London Research Centre (LRC). But in this paper, a parameter "the ratio of the number of posts to the number of population" was added to explain that the essence of the Zone 3 is commuting area (the word "commuter hinterland" was used to interpret Zone 3 in the original text, but its actual division had not defined the zone strictly according to the scope of the commuting area). We hereby made some adjustment to the division in Paris and Tokyo: 1) there was no Zone 4 in the original text and the Zone 4 (Little Crown) was defined as Zone 3 and Zone 3 as Zone 2. Actually, the Zone 4 (Little Crown) is made up of the eight provinces in Paris Metropolitan, whose scope is beyond the actual commuting scope of Paris. The Zone 4 in Paris Metropolitan share the same meaning with the Zone 4 in Tokyo Metropolitan and New York Metropolitan. The Zone 3 in this paper is the built-up area of Paris Metropolitan in the original text and it's the actual commuting area in Paris. We defined the Little Crown (Petite Couronne in France) which made up of Paris and its adjacent three provinces as Zone 2, as it is more in line with the definition of the urban area in Zone 2. 2) Tokyo was defined as the Zone 3 in the original text. Tokyo is made up of Tokyo Metropolis, Tama area and the island area. Although the Tama area is one of the main

residence for Tokyo employment, large numbers of area employment live in the areas within 40~50 km away from the Tokyo Station, including Kanagawa, Saitama and Chiba. Referred to the latest research on Tokyo Metropolis and combined with the detailed analysis of town level on population and employment data, this paper identifies the approximate scope of commuting area in Tokyo and defines it as Zone 3 for further comparison.

Fig.1. Zones division in Tokyo metropolitan

Fig.3. Zones division in Paris metropolitan

Table 1.a. Statistics on surface area, population and employment in Tokyo metropolitan

Area (km2) Jobs Population (person) Number of jobs/number of population

Zone 1 42 238.1 26.6 895%

Zone 1 & 2 617 724.9 816.4 89%

Zone 1, 2 & 3 3993 1126.6 2262.0 50%

Zone 1, 2, 3 & 4 13143 1644.1 3179.7 52%

Table 1.b. Statistics on surface area, population and employment in New York metropolitan

Zone 1 Zone 1 & 2 Zone 1, 2 & 3 Zone 1, 2, 3 & 4 33165

Area (km2) Jobs Population (person) Number of jobs/number of population

23 196.7 54.3 362%

757 413.2 749.7 55%

5793 744.4 1352.6 55%

33165 1067.4 1984.3 54%

Table 1 .c. Statistics on surface area, population and employment in Paris metropolitan

Area (km2) Jobs Population (person) Number of jobs/number of population

Zone 1 29 102.5 62.2 165%

Zone 1 & 2 762 351.4 614.1 57%

Zone 1, 2 & 3 2060 449.4 879.1 51%

Zone 1, 2, 3 & 4 12011 507.5 1066.1 48%

Note: Statistics reference to "The Four World Cities Transport Study". This paper updated the statistics for adjusted areas including Zone 3 in Tokyo, Zone 2 and Zone 3 in Paris

3. Structural comparison research on regional rails in metropolitan areas

3.1. The Scale and distribution of the network

The statistics for regional rails in three metropolitan areas and each zones is shown in Figure2. Some features are found:

1) The scale: The total route length for regional rails has reached 1500~3000 km which is approximate 4-8 times as long as metros.

2) The distribution: Compared to metros, regional rails are mainly distributed in Zone 3 and Zone 4, while the route length in Zone 1 and Zone 2 is much shorter. Tokyo and the Zone 2 in Paris have remarkably longer regional rails than New York.

Table 2 Statistics on route length for regional rails and metros in three metropolitan areas

Tokyo New York Paris

Regional rail Metro Regional rail Metro Regional rail Metro

Zone 1 36 99 7 73 23 86

Zone 2 368 191 160 317 383 115

Zone 3 1283 66 558 30 598 0

Zone 4 1321 0 877 0 496 0

Total 3009 356 1602 420 1501 201

Note: 1) Regional rails in Tokyo are maintained by the Japan Railways and other private operators, in New York are maintained by the Long Island Railroad, the Metro-North and the New Jersey Transit, while in Paris are the Regional Express Railway (RER for short) and the Transilien system.2) The statistics of New York was cited from "The four world cities transport study" and the statistics of Tokyo and Paris was updated to 2005 according to the government statistics reports.

3.2. Functional division for regional rails

The substantial growth of route length for regional rails were mainly occurred in the stage of the suburban development. In this stage, the continuous expanding of metropolitan areas and commuting areas had exceeded the efficient service scope of metros or light rails. The city needed a rail transit system which is not only faster but also provides certain transport capacity to meet the travel demand which has longer travel distances and certain transportation intensity in metropolitan areas.

At the beginning of the construction of regional rails, most of them mainly served for the travel demand in commuting areas. Their service areas were within commuting areas thus defined them as "commuting rails", such as the RER system in Paris, private rails in Tokyo (they mainly provide the service of commuting rails) and three commuting transit system in New York. There are some regional rails whose service scope extend to the Zone 4. They provide all weather relative even services and its departure interval does not show significant valley features to achieve the main purpose for connecting exurb areas and urban areas. These rails are defined as connecting rails. Tokyo's JR rails, some commuting rails in New York and the Transilien system in Paris all show same typical features as connecting rails. The service scope and operating features of two kinds of rails are shown in Table 3.

Table 3 Functional division of two regional rails

Rail type Spatial service scope Functions Operating features

No clear peak-hour features. Departure internals are relatively even and the departure frequency is low.

It shows clear peak-hour features and its departure frequency during peak hours is close to metros.

Connecting Rail

Commuting Rail

Connects the Zone 4 and Zone 3 with Zone 2

Mainly connects the Zone 3 with the Zone 2

Provides fast communication between the urban area with the outskirts and improves the accessibility of the outskirts Mainly serves for commuting travels entering the Zone 3, but also for some interior travels in the urban area.

3.3. Division of radial rails in the inner area

With long regional rail routes, the ratio of radial regional rails to radial rails ("ratio of radial regional rail" for short) has sharp distinction between metropolitan areas, which means existing huge differences in resources allocation of radial corridors in the Zone 2. Table 4 indicates that the ratio of radial regional rail in Tokyo has reached 71%, Paris with 47% while New York with only 21%. Actually, the number of radial corridors in the inner area could determine the corridor capacity entering the urban area from outskirts, and then to determine the number of commuters entering the urban area from outskirts in peak hours. In Tokyo, allocating more radial corridors in the inner area to regional rails suggests that there are numerous people who work in Zone 2 but live in outskirts.

Table 4 Route length of radial rails and their constitution in three metropolitan areas

Tokyo New York Paris

Number of radial rails 34 24 36

Number of radial regional rails 24 5 17

Ratio of radial regional rail 71% 21% 47%

Further considering the functional division of regional rails, two-thirds of radial regional rails in Tokyo are commuting rails, while the ratio in Paris is approximate one-half. It also indicates that approximate one half of corridor resources in the inner area are allocated to commuting rails while Paris allots nearly one-forth.

3.4. Selection of regional rail corridors

1) Tokyo: Connected operation of regional rails and metros

At the early stage, most of regional rails in Tokyo ended in the Yamanote Line, so travelers should transfer metros or trams to enter the central area. However, huge amount of transfer caused the paralysis of transferring hubs (Liu Longsheng et.al, 2013). Tokyo government began to realize the necessary of introducing regional rails into the central city and planned to allot limited corridor resources by connected operation of regional rails and metros. According to the operating characteristics of metros and trains of regional rails in the running section of metros, 13 lines of metros in Tokyo is divided to 4 kinds (as shown in Table 5). Among that, type I only provides metro service; type II provides some connected operation of regional rails during peak hours and off-peak hours; type HI mainly provides connected operation during peak hours; and type IV acts as regional rails to enter corridors in the central city. So, except early-built metro lines and circle lines, all lines could provide convenient connected operation during peak hours. The same as metros, regional rails would stop at each station after entering the running section of metros. The corridor capacity of metros would not be affected.

Table 5 Operation types of Tokyo metros

Type Characteristics Typical line Comment

With special system, Ginza Line and Marunouchi Line could not provide connected operation. The Oedo Line is a circle line.

No metro service on Asakusa Line (from Sengakuji to Oshiage) and Hanzomon Line. The Namboku Line has only three metro trains on every _weekday._

2) Paris: Parallel laying of regional rails

Different from Tokyo, there is big difference in function between the RER system and the Transilien system in Paris. The RER system show typical features of commuting rails in service scope and operation characteristics, such as the service scope is mainly in Zone 3, the departure interval has significant peak and valley characteristics and the departure frequency is close to metros during peak hours. The Transilien system show typical features of connecting rails. One RER line being parallel with the other Transilien line to entering the urban area is the most common seen laying method. Because the RER system has more stations after entering Zone 2 ( to disperse passenger pressure), and its lines enter the city central (RER lines are parallel with Transilien lines after entering the city central), while the Transilien system ends at seven train stations at the edge of city central after stopping at several major stations (Feng Li and Gu Baonan, 2008).

3) New York: Independent channel for regional rails and metros

The network of regional rails in New York is very developed, but its structure is typical tree-like structure: large number of branches are used in outskirts to improve the coverage rate and are merged into several major rails in the urban area to enter the city central. In New York, regional rails are independent from metro corridors, so there is almost no shared corridors or rails. In the urban area, regional rails are laid along minor and fringe corridors while major corridors are left to metros.

Type I

No connected operation, only metro service

Oedo Line, Ginza Line, Marunouchi Line

Type II

Type III

Type IV

During peak hours and off-peak hour, metros service has stable operating frequency and provids some connected operation. Mainly provids connected operation service during peak hours. Operating frequency of metros is low during peak hours while high during off-peak hours.

With no or few metro service, so metro trains actually provids connected operation.

Mita Line, Hibiya Line, Yurakucho Line, Fukutoshin Line

Asakusa Line (from Nishi-

magome to Sengakuji), Shinjuku Line, Tozai Line, Chiyoda Line

Asakusa Line (from Sengakuji to Oshiage), Hanzomon Line, Namboku Line

3.5. Rail hub system

The significant difference in resources allocation and corridors selection further leads to the difference in the layout system of the rail hub. Comparative analysis suggests that the biggest difference in rail hub system between three metropolitan areas is the hierarchical structure of hubs for regional rails (as shown in Table 6). The transfer hub for regional rails-metros in Tokyo is the biggest (because of the connected operation of regional rails and metros, most of metro hubs are also the hubs for regional rails-metros), thus they could disperse tremendous transfer pressure during peak hours of commuting travels. New York has the fewest regional rails entering the urban area and their channels are independent. Its regional rails concentrated on few large-scale hubs, such as the Central Station and the Penn Station. The hubs for regional rail-metro in Paris operate at scale, but they are independent from metro hubs. Moreover, there are some large hubs for regional rails in Tokyo and Paris which locate outside of the city. Their locations are always coupled with some important nodes of the city to promote the development of key areas on outskirts.

Table 6 Hierarchical Structure and the Scale of Rail Hubs in Three Metropolitan Areas

Type Tokyo New York Paris

Number of hubs for high speed rail-urban 12 6 7

Number of hubs for

Number of hubs regional rails

for regional rail Number of hubs for

3 , , 25-30 Almost no 10-15

regional rail-metro

Number of the metro hub 35-40 35-40 35-40

Note: 1) This paper does not take coincident stations on shared rails of regional rails and metros in Tokyo into account; 2) This paper does not take coincident stations on shared rails of the RER system and the Transilien system into account.

30-35 5-10 20-25

3.6. Structure of railway network

Abovementioned analysis indicates that the structure of railway network are totally different between three metropolitan areas. The further analysis of resident-employment distribution suggests high correlation between the railway network and the resident-employment distribution (as shown in Table 7).

(1) Railway Network in Tokyo

There are approximate 3.04 million workers living more than 20 km outside the urban area (beyond the efficient service scope of metros) of Tokyo, which covers 42% of people work in the urban area. Based on commuting rails, Tokyo government constructed very complex railway network to introduce large number of commuting rails into the urban area and the city central: On the one hand, the railway network had to increase the proportion in corridor resources allocation by the way of connected operation; on the other hand, tremendous regional rails and metros intersect at the city central, which formed numerous multi-station transfer hubs. (According to statistics, among 76 stations on the Yamanote Line, there are 34 transfer stations, of which ten four-line transfer stations and three five-line transfer stations).

(2) Railway Network in New York

With similar commuting scope with Tokyo, there are only 0.4 million workers living 20 km outside the urban area of New York. Metro system could afford 90% of commuting travels, which suggest that it does not require large number of commuting rails. On the one hand, the tree-like structure of commuting rails are appropriate for widespread and scattered features of workers living in outskirts, on the other hand, as the passenger volume is relative small, the tree-like structure could focus on few corridors to enter the city central and utilize several major transfer stations. The railway network in New York is more simple and more efficient than the network in Tokyo.

(3) Railway Network in Paris

In 1990s, there are 0.4 million workers living outside the urban area of Paris. Their distribution are concentrated on river valleys or arterials. At this stage, the RER system was constructed at large scale and served for these commuting travels (Zeng Gang and Wang Chen, 2004). Compared to Tokyo, Paris has less commuting rails and hubs. Its lines are operating independently and the complexity of the network is between Tokyo and New York.

Table 7 Resident-Employment Distribution Features and Network Structure in Three Metropolitan Areas

Metropolitan area The number of workers in Zone 1&2 The ratio of radial The number of

Network structure living 20 km outside the city regional rails to radial hubs for regional

(persons)/ratio (%) rails (%) rails

Tokyo Concentrates on commuting rails 3.04 million/42% 71% 55-65

New York Concentrates on metros 0.4 million/10% 21% 5-10

Paris Metros and regional rails 0.5 million/14% 47% 30-40

4. Enlightenment to the planning of regional rails in China

Comparative study of three metropolitan areas indicates that they have completely different structure of railway network, among that regional rails have biggest difference. This section attempts to analyze their causes from the perspective of historical process.

4.1. High correlation between network structure and resident-employment distribution features

Tokyo, New York and Paris are regarded as representative cities for those have strong concentration on the city central [14-15]. But these metropolitan areas are completely different in resident-employment distribution and the railway network structure. From the perspective of historical process, there exists interactive developing relationship and route-dependent effect between the spatial structure of the city and the structure of rai lway network. The beginning of the route is the special resources of each city[16-17]. Tokyo has tremendous early-built wooden buildings with low plot ratio and limited supply of residence (Liu Longsheng et.al, 2013). At that cultural and legal environment, the blocked renewal of the city pushed tremendous workers run to outskirts of the city, which caused large-scale commuting travel demand thus the structure of railway network generally changed to focus on commuting rails. Large amount of residence live in middle-rise and high-rise buildings in the Zone 2 of New York, which keeps 90% of commuters within the urban area (the efficient service scope of metros). Hence, the metro system in New York has the largest scale and the most complex layout method and operation method between three metropolitan areas. Before 1990s, Paris was regarded as a typical city with strong concentration on the city central. But limited by the restrictions on land use and development intensity, newly increasedjobs were distributed to several sub-centers in the Zone 2 and Zone 3. In recent years, the population of the urban area has increased and the resident-employment distribution becomes more balanced. It even appears a certain proportion of reverse direction commuting travel (Aguilera, A, 2009). During this process, the function of commuting rails has changed from typical radial lines of city central to fast connecting rails between city central and sub-centers.

Compared to Tokyo and Paris, New York effectively controlled the disorder spread of commuting area by the urban land use planning, which not only significantly reduced the complexity of railway network, but also reduced infrastructure investment and improve the efficiency of the transportation system and its service. Tokyo also made some reflection in recent years on its structure of the resident-employment distribution and put forward series of approaches to deal with the problems of commuting areas (referred to the website of Tokyo Metropolitan Transportation Planning Association). Commuting areas in some Chinese cities has reached 15~20 km, which means it would be the emphasis of this wave of big-city planning to control the boundary of commuting areas. Few megacities with 30-50 km commuting areas need to consider reservation control of corridors and hubs to create conditions for introducing regional rails into the city central.

4.2. Subivision of city structure and functions

The scale of regional rails has high correlation with the structure and the layout of the city, so it is of great importance to identify and subdivide the structure of the city, which is also the confusion of the regional rails planning in China. Referred to related research and based on the truth that the definition scope and empirical scope of metropolitan areas are bigger than commuting areas, this paper define the commuting area and more generalized

metropolitan area. Regional rails are also divided into commuting rails and connecting rails which have huge differences in service scope, function, operation characteristics and planning principles.

The following points need more attention when planning connecting rails and commuting rails: 1) In principle, although all regional rails need to be introduced into the city central, as higher sensitivity to time and transfer for commuting travelers (Li Yiqing and Wu Binghua, 2004), commuting rails require more needs than connecting rails. 2) Commuting rails need more important corridors resources. 3) With large passenger capacity during peak hours, it is better to plan more stations for commuting rails after entering the urban area to diffuse pressure from boarding and alighting passenger and transferring.

4.3. Analysis of the development context for regional rail planning

Given different development background, three metropolitan areas make different choices in the planning of regional corridors. So we must take into account of different development context when using overseas cases for reference to avoid blind copy. Firstly, the difference in resident-employment distribution must be considered. It would be better conditions for independent layout when the proportion of radial regional rails was low. In other words, the way of shared rail or shared corridors is not always the first choice. Secondly, different background conditions of railway network must be considered. In 1960s when most of metros in Tokyo had not yet constructed, they encountered various problems without introducing commuting rails into the city central. They timely realized the need for connected operation and kept the open of metros and the connected operation at the same pace. At the time of the construction of the RER system in Paris, metros and suburban rails had already been constructed. Paris government connected part of suburban railways with newly-built underground corridors to form the RER system, which is the reason that many RER lines are parallel with the Transilien lines. Thirdly, the limitation of each method must be given careful analysis. Such as the connected operation in Tokyo, it solved the problem of corridors allocation, but regional rails have to stop at each station after entering the running section of metros, which affected their running speed. So trains have to accelerate by overtaking rails on the outskirts of the city. The overtaking organization of railways would result in increased investment and decreased passenger capacity. But private railways of Tokyo and JR lines were used for freight transportation at the early stage, so they have longer station ( using more marshalling stations to improve the capacity of each train ) and more overtaking rails. The RER system integrated scattered suburban railways, but given different supply connection, it was hard for trains to overtake rails until the release of the Dual Volt Train.

4.4. Rational planning and reserving transfer hubs

The location of railway hubs needs coupling with city centrals. Introducing regional rails into the city centrals must increase the number of hubs, especially transfer hubs for multi-line (more than three-line). When planning the railway network, in order to guarantee the best accessibility and transfer conditions for highly concentrated areas, it is better to rationally reserve transfer hubs according to the demand. Constructing railway hubs in sub-centers on the outskirts or important development nodes could improve their position advantage and drive the development of the major areas on the outskirts.

5. Conclusions and Prospects

The planning of regional rails is an important and difficult step in the new wave of railway network planning. Exemplified by three metropolitan areas including Tokyo, New York and Paris, this paper firstly divides four areas for metropolitan areas according to their resident-employment distribution features. These areas are: the city central, the urban area, the commuting area and the metropolitan area. Given that, this paper compares their differences in the scale of railway network, functions of regional rails, the corridors allocation of inner areas, the corridors selection and the layout of hubs. It is concluded that three metropolitan areas have different network structure of railways. Moreover, considering resident-employment distribution features and specific development conditions, this paper analyzes the origin of each network structure and summarizes some lessons and enlightenment for the regional rails planning in China.

The regional rail is the product of city development, so its analysis could not neglect specific development conditions and the system environment. Future study should give more attention to the system design and the planning of regional rails to provide more comprehensive reference for the implementation of regional rails.

References

Gu Baonan, Guo Changgong.,2010.0n Shanghai Urban Rail Transit Network Planning (Chinese) [J].Urban Mass Transit13(10), 4-9.

Zhang Xiaochun, Song Jiahua, Shao Yuan.,2012. The Strategy for Rail Transit Development of Shenzhen in the Context of Metropolitan Areas (Chinese).Urban Rapid Rail Transit25(3), 1-6.

Merlin, P.,1992. Géographie des Transports, Que sais-je?. Presses Universitaires de France

Wu Xueming.,2003. The City Structure and Population Distribution-a comparison between London, Paris, New York and Tokyo and a Simulation of Shanghai (Chinese). World Economy Study7, 22-27.

Wang Wenjing, Lu Huapu.,2013. The Research of Coupling Relationship Between Rai Transit and City Structure of Metropolitans (Chinese).Urban Development Studies20(4), 112-118.

Condit, C.,1980. The port of New York, A History of the Rail and Terminal System from the Beginnings to Pennsylvania Station. University of Chicago Press.

Caralampo, F.,1998. The Four World Cities Transport Study. London Research Centre.

Liu Longsheng, Du Jianhua, Zhang Daohai.,2013. City of Rail: Urban and Transport Research on Tokyo Metropolitan Area (Chinese) .China Communications Press.

Zhang Xiaolian.,2001. The Suburbanization and Metropolitans Development of American Cities (Chinese).Urban Problems4, 58-60.

Feng Li, Gu Baonan.,2008. Development Experiences of Suburban Railway in Major Cities Abroad (Chinese). Urban Mass Transit11(12), 49-53.

Zeng Gang, Wang Chen.,2004. The Plan and Development of Paris Area (Chinese) . Urban Planning Overseas 19(5), 44-49.

Thomson M.,1982. Great Cities and Their Traffic. China Architecture & Building Press.

Rodrigue JP.,2006. The Geography of Transport Systems. Routledge.

Liu Weiqi, Jiao Binlong.,2006. A Reinterpretation of City and Urbanization (Chinese).Urban Problems6, 7-10.

Li Daoyong, Yun Yingxia, Ren Jingjing.,2013. The Coordinated Development Between the New Towns and the Rail Transit in the Polycentric Metropolitan Area:Paris's Experience in Building and Inspiration (Chinese).Urban Development Studies20(11), 81-86.

Aguiléra, A., Wenglenski, S., Proulhac, L.,2009. Employment suburbanization, reverse commuting and travel behaviour by residents of the central city in the Paris metropolitan area. Transportation Research Part A43, 685-691.

Tokyo Metropolitan Transportation Planning Association.^)® 32 ^â®H^^Îff+[EB/0L].http://www.tokyo-ptjp/index.html.

Zhao Linna, Wang Wei, Ji Yanjie, Fan Rui.,2014. Impact of Diversified Passenger Demand on Transit Satisfaction (Chinese) [J].Urban Transport of China4, 65-71.

Li Yiqing, Wu Binghua.,2004. The Development of the Regional Expressed Railway in Paris (Chinese) [J]. Urban Mass Transit7(3), 77-81.