Conference PaperPDF Available

The development of metro variants Lessons from Mexico and other countries

Authors:

Abstract and Figures

Between 1967 and 1970 a first set of three lines representing 42 kilometers was built, the most part of which was underground and served the city on a so called ring layout. The process was based on relationships established between Ste, Ica and the firms associated with RATP. Right from the start, those lines have been very crowded and by the end of that step, they represented about 12 % of the total transport market shares in the City. This demonstrated the appropriateness of the layout as well as the extent of the delays accumulated in public transit. At that period, all of the equipment and rolling stock were imported (nine hundred cars) with a top option of pneumatic rolling, which then proved somehow limited from the capacity viewpoint (Coindet, 1989). The Mexican share was limited at 65 % of the work essentially representing civil engineering and some basic equipments. During a second step which can be qualified as "substitutive step" transport investments concerned road infrastructures (ring road and 20 major crossing highways). During that period the construction of the metro remained at a standstill because the industrial and technological dependancy as regards rolling stock purchasing was not so well accepted knowing that the rolling stock had to be increased considerably. From 1977 to 1985, the three existing lines were extended and four other ones were opened, running towards the East, West and North. A total of 73 km were built with large surface or overhead sections and equipped with about 1000 vehicles, half of which were produced under licence by local firms. The Federal District Department was dealing with a powerful delegate contracting authority, called Covitur, who had direct relations with Banobrass the National Bank for Development and who was linked with the main foreign providers, most of whom were French, according to sustainable institutional enacting terms. The third step, "a more critical one", led to reconsider planification despite the renewal of French subsidies, and to adjust the construction pace of the metro to the real capacities of the country. Moreover the aim of a so called "mexicanization policy" was to reinforce the local appropriation of the rolling equipments technology ; during that period, the proportion of national integration was 85 %. If during the previous step the lines were not as crowded as the three initial ones, they were to be extended until the limits of the Federal District and line 9 was to absorb the overflow on the central network ; twenty-six kilometers were put in service with more than 400 cars, produced by Concarril, the national manufacturer, who however had to face so many difficulties that it was necessary to import trains from Canada (later this State Company was sold to Bombardier company). Then, an "alternative" step occurred. In the middle of the eighties, the Mexican debt crisis made the relationships deteriorate between the Federal State and Mexico agglomeration. As a result, the fourth step was characterized by the introduction of significant variations in the system. Attempts were made to develop lighter alternative solutions and to rehabilitate the existing acquisitions, to extend the network towards the outskirts and move towards metallic rolling and electric supply by catenaries. If line 9 was achieved in conformity with the initial plans and contracts with the French suppliers, two alternative achievements were undertaken and called "light railway" : line A called Iron Metro and the Ste renewed tramway called light train. During that period, a total of fifty kilometers of metro and tramway lines were built or renewed, with a dotation of hardly 300 cars. If light metros cars were produced locally, pneumatic trains were imported from Spain.
Content may be subject to copyright.
1
The development of metro variants
Lessons from Mexico and other countries
Etienne Henry Francis Kuhn
Research Director Research engineer
Inrets/Dest Inrets/Estas
The French National Research Institute for Transport and Safety Research
2
1. A RETROSPECTIVE ANALYSIS OF THE VARIANTS
The debate about mass transit rail systems is changing namely concerning the
recommandations given to the developing cities through the international organizations
(World Bank, 1986, 1995). While metros are acknowledged as a significant factor of
urban, economic and social development (ice 1990), this concept will be adapted to local
needs and capabilities. There are many variants from one city to the other and what is
called a metro in a city may be considered as an urban light railway elsewhere (in Recife
and inversely in Medellin) or as an express regional network ; a light railway system
(such as in Guadalajara) can reach almost the same capacity as a traditional railway
system, or it may come up against its limits (in Manilla) or it can be very similar to a
modern tramway (in Tunis). Beyond taxinomies, there remain some substantial
differences between those equipments in terms of infrastructure (surface, overhead, or
underground railway), rolling stock (coupled vehicles, made up trains, or underground
trains with pneumatic or metallic rolling, with more or less automatic guidance), sites
(ordinary or proper site) and other dimensions, which all have a final influence on the
three sets of variables : functions, capacities and costs of the system.
It can be stated that functions are varying depending on the services provided in inner
cities or in the outskirts and on the size of the town. Capacities depend on the present
and future demand. As a result, the costs will range from one to ten as for investment
and from one to five as for operation (Fouracre & al. 1990). However, from many
viewpoints, costs are very difficult to compare whether technology, urban planning,
building, financing or operation are considered. There have usually been three big
categories of systems including the urban underground railway which provided services
in crowded areas with a capacity ranging from 10 000 to 40 000 seats a direction at peak
hours, the regional railway system, running in the outskirts (from 50 000 to 75 000
seats/hour/direction) and the tramway (between 5 000 and 10 000 seats/hour/direction)
for lower demand areas (Vuchic, 1988). Now, hybrid systems are emerging and derive
from one of those categories while they may include characteristics from another one.
This change will lead us to look at the dictionary in order to obtain the definition of
what is still called metro or metro variant :
"Variant : 1. Something which is slightly different from something else inside one same
species ; 2. Differences or set of differences characterizing a copy or a new project as compared
with the original work or project". (Le petit Larousse 1993).
The first meaning of the definition corresponds to the expectancies of most of the
developing cities as regards the low cost implementation of a system they would like to
be as similar as possible as a metro. The second meaning puts forwards the results,
merits and limits of what could be built and the differences between the result and the
expectancies. It also suggests to observe the variation process during the implementation
of a metro project, a long lasting process indeed.
This double meaning exists in the a posteriori assessment approach. It will be applied to
Mexico city, which, despite all kinds of difficulties, has always been willing to be
equipped with a metro in the strict sense of the concept. Mexico has progressively been
obliged to introduce several kinds of variants or even to let achievements emerge which
are notably diverging from the initial project. The rail system implemented is still
keeping the basic characteristics of an urban metro. However the spatial and
demographic dimensions of the megalopolis have made the specificities evolve either
towards a regional railway system or towards a light railway system, while none of the
achievements corresponds to one of those designations. If lessons may be drawn from
3
that experience, it's not so much because it has been considered as a reference (Parpillon
& Laraufie, 1992 ; Balat & Vince, 1990) than because it allows to compare achievements
the external variables of which are controlled, which scarcely is the case in international
comparisons (Henry 1987a). We don't intend to evaluate those achievements but to
contribute to the debate on the basis of some evaluation elements extracted from an
amount of data collected in view of specific studies about the developing cities
experience, a large scale but not very well known experience.
2. TWENTY FIVE YEARS TO BUILD THE METRO IN MEXICO CITY
Since 1967, Mexico City has been equipped with a modern urban transit rail system
which soon will reach 200 kms : this is considered as an achievement from the technical
viewpoint, insofar as for a long time, the idea had been that the poor underground
conditions of the lacustrian basin in Mexico city were an obstacle to the implementation
of a metro ; the 1985 earthquakes have not affected the floating underground
infrastructure which had been designed to face those problems (Navarro, 1994). But it
also represented an achievement from the economic viewpoint since no other country
has been able to build such a system within 25 years. Indeed today, this metro is the
fifth in the world and is still in extending (more than a century had been necessary to
achieve the 200 kilometers of the metro in Paris). This work is the result of the efforts
assumed by the whole Mexican nation, due to the Federal District status and the size of
the Mexico agglomeration (estimated in 1965 at 6,3 million inhabitants on a surface of
372 km2 and in 1990 at 15,6 million inhabitants on a surface of 1 250 km2, according to
the figures which are underestimated as compared with the real ones). It also represents
the work of a powerful Mexican civil engineering group (Ica Consortium - the main
contractor of the Mexican public works and even operating in the United States),
together with about 20 directly or undirectly associated companies of public works and
equipment. At last, it results from an international cooperation project which namely
associated about sixty French companies dealing with electromechanics, installation of
signals, rolling stock, such as Sofretu, Alsthom and BNP with considerable development
subsidies from the Government (representing 1/20 of the total amount of the loans with
bonus and public donation granted by France to foreign States for 30 years). Nine lines
are now operated by a public company : Stc metro. The tenth line, called light railway, a
modernized tramway line replaced by the metro is operated by Ste which also manages a
trolleybus network. Starting from an objective of ten kilometers a year, the Mexicans
succeeded in building an average of 7,4 kilometers a year, and play a role of precursor in
Latino-America (Figueroa & Henry, 1987a, 1991a, 1991b). The irregular curve on figure
1 illustrates an impressive progress in the lines operated during the three initial years,
followed by a standstill step, and then by an accelerated growth during the eighties. The
development has been more irregular for 1990. It takes about two or three years to build
a line and therefore those bends on the curve are reflecting the previous choices and the
considerable variations in the the metro policy during the building process.
4
Évolution du réseau ferroviaire urbain de Mexico 1969-1994
0
50
100
150
200
250
1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994
Kilomètres de lignes en service
Stc-Métro
Stc + Ste
Figure 1
Figure 1 : Urban rail network of Mexico evolution 1969 - 1994
Métro de Mexico, phases d'implantation
Phases Lignes en chantier Linéaire Voitures
* dénomination construit acquises
I (67-76) 1, 2 & 3 42 km 882
II (77-84) (1, 2, 3) 73 km 932
4, 5, 6 & 7
III (85-88) 6, 7 & 9 25,6 km 428
IV (87-94) 8, A
Train léger 49,5 km 267
1967-94 10 (dont LA & TL) 190,1 km 2 509
* Périodisation Inrets
Sources : Figueroa & Henry : 1987a,
Henry :1988 &1994)
Table 1 : Construction steps of Mexico’s metro
5
Table 1 sums up the implementation steps of the metro in Mexico City, based on an
homogeneous and sustainable project but with significant variations as regards the
layout of the network, the infrastructure and equipment, the production of rolling stock,
and the extent of metropolitan services. That cannot really be called periods and all the
variations don't follow a chronological order. They can be summed up in four steps
according to a logic which corresponds to three sets of factors : the economical situation
and debt capacity of Mexico, the succession of five presidential six years terms, and the
urban growth as well as the applied transport policies.
Between 1967 and 1970 a first set of three lines representing 42 kilometers was built,
the most part of which was underground and served the city on a so called ring layout.
The process was based on relationships established between Ste, Ica and the firms
associated with RATP. Right from the start, those lines have been very crowded and by
the end of that step, they represented about 12 % of the total transport market shares in
the City. This demonstrated the appropriateness of the layout as well as the extent of
the delays accumulated in public transit. At that period, all of the equipment and rolling
stock were imported (nine hundred cars) with a top option of pneumatic rolling, which
then proved somehow limited from the capacity viewpoint (Coindet, 1989). The
Mexican share was limited at 65 % of the work essentially representing civil engineering
and some basic equipments.
During a second step which can be qualified as "substitutive step" transport
investments concerned road infrastructures (ring road and 20 major crossing highways).
During that period the construction of the metro remained at a standstill because the
industrial and technological dependancy as regards rolling stock purchasing was not so
well accepted knowing that the rolling stock had to be increased considerably. From
1977 to 1985, the three existing lines were extended and four other ones were opened,
running towards the East, West and North. A total of 73 km were built with large
surface or overhead sections and equipped with about 1000 vehicles, half of which were
produced under licence by local firms. The Federal District Department was dealing
with a powerful delegate contracting authority, called Covitur, who had direct relations
with Banobrass the National Bank for Development and who was linked with the main
foreign providers, most of whom were French, according to sustainable institutional
enacting terms.
The third step, "a more critical one", led to reconsider planification despite the renewal
of French subsidies, and to adjust the construction pace of the metro to the real
capacities of the country. Moreover the aim of a so called "mexicanization policy" was
to reinforce the local appropriation of the rolling equipments technology ; during that
period, the proportion of national integration was 85 %. If during the previous step the
lines were not as crowded as the three initial ones, they were to be extended until the
limits of the Federal District and line 9 was to absorb the overflow on the central
network ; twenty-six kilometers were put in service with more than 400 cars, produced
by Concarril, the national manufacturer, who however had to face so many difficulties
that it was necessary to import trains from Canada (later this State Company was sold
to Bombardier company).
6
Then, an "alternative" step occurred. In the middle of the eighties, the Mexican debt
crisis made the relationships deteriorate between the Federal State and Mexico
agglomeration. As a result, the fourth step was characterized by the introduction of
significant variations in the system. Attempts were made to develop lighter alternative
solutions and to rehabilitate the existing acquisitions, to extend the network towards the
outskirts and move towards metallic rolling and electric supply by catenaries. If line 9
was achieved in conformity with the initial plans and contracts with the French
suppliers, two alternative achievements were undertaken and called "light railway" : line
A called Iron Metro and the Ste renewed tramway called light train. During that period,
a total of fifty kilometers of metro and tramway lines were built or renewed, with a
dotation of hardly 300 cars. If light metros cars were produced locally, pneumatic trains
were imported from Spain.
The Mexican were not really concerned by technical research on intermediary transport
systems, but financial constraints obliged them to introduce alternative solutions. During
the lucky years of oil exploitation, the country was borrowing from foreing countries to
finance public investments and in particular urban infrastructure. The most significant
project was the Mexico City metro program (besides the road system construction and
the in-depth draining program).
Dette extérieure publique mexicaine 1970-1992
0
10 000
20 000
30 000
40 000
50 000
60 000
70 000
80 000
90 000
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
Millions de US$
Figure 2
Figure 2 : Foreign public mexican debt 1970 1992
7
Impact du métro sur la dette du DF 1970-88
0
100
200
300
400
500
600
700
800
900
1970 1971 1972 1973 1974 1975 19761977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988
Dette propre DDF millions US$
DDF STC-METRO DDF+STC
maître d'ouvrage du métro
y compris COVITUR,
Figure 3
Figure 3 : Metro effect on the DF debt 1970 - 1988
During the eighties, the situation deteriorated and in 1987, the balance of account of the
external public debt reached the record deficit of 80 billions Dollars (figure 2). The
renegociation of this debt with the Club of Paris allowed to stop it and induced strict
measures of structural adjustment and a substantial reduction of large works programs.
Since the metro was funded by the Federal District, the debt of this department, whose
direction was under the supervision of the Central State, was as heavy as the nation's
debt (figure 3) : a limit was to be reached by the end of the seventies. Therefore, the
mexicanization process accelerated and the Stc operating Company was deprived of the
metro construction control. Since 1981 this control had been assumed by Covitur, a
company directly connected with Ddf. 80 % of its transport investment budget had
been devoted to the metro. The Federal District debt increased again due to the new
metro investment program. In 1985 this debt was so deep that the Central State had to
intervene again to stabilize the Ddf financial situation. Since 1987 alternatives have had
to be found to maintain the metro construction pace and to limit the external debt. This
alternative solution was the "light metro" construction period with new project
definitions. Further to that stabilization period, the improved economic situation
allowed other projects to be undertaken, such as line 8 at the beginning of the present
decade.
Turnkey imports, technological substitution, critical debt, and lighter alternative research
are phases which are perfectly illustrating the various aspects of the metro
implementation in developing cities. Despite the 1972-77 and 1987 suspensions of
8
Mexico city works, tensions did not stop the construction as they often do : the French
financial assistance, which has now been closed because of the entrance of Mexico into
OECD, had a stabilizing effect. But significant variations have marked the
implementation process. During the whole process, the prevailing idea was to provide a
backbone to the transport system in Mexico City. It was the octogonal network, which
criss-crossed the Federal District. It also was to provide huge connection stations with
buses at the nodes between residential and industrial areas in Mexico State.
This option, which, according to the last version of the Covitur program should result in
2010 in 460 kilometers of lines, is a large capacity urban metro option. With about 4,5
million passengers carried a day, the Mexico City Metro is the fourth in the world from
the attendance viewpoint even if, according to the recent origin-destination survey it
only represents 13,4 % in the modal distribution of the trips in the whole agglomeration.
Indeed, this result reveals the paradox of mass transport systems in developing cities. If
transit growth pace is very rapid in Mexico city, it is slower than the urbanization pace.
It's from that viewpoint that the metro variants have to be evaluated.
3. THE PROGRESSIVE IMPLEMENTATION OF THE BASIC NETWORK IN
MEXICO CITY
The basic design which guided the three first phases was a classical urban metro design
with the technical specifications proper to the site and to the option choosen in terms of
rolling stock (with tyres similar to the recent innovations in Paris). It also corresponded
to the will to adopt high technologies for the electro-mechanic equipment. The goal was
to reach the performance of about 60 000 places at peak hours ; indeed on some lines,
the technical limit of 95 seconds between two trains can be reached (according to the
driving system in use).
As regards civil engineering, the main technical options derived from the choice of the
guided transport system in integral proper site in a dense built up area with pneumatic
rolling stock, able to stand more significant slopes and curves than metal rolling does.
Solutions were looked in order to obtain a difference of level with general traffic or with
the ground and to lower the achievement delays and costs. During the three first phases
four options of civil engineering were adopted : underground, covered cutting in central
areas, for most of the first lines ; surface platforms large enough central platform was
available on large avenues ; elevated viaduct crossing less sensitive areas in order to avoid
interferences between networks or on hillsides as for the whole line 4 ; and deep tunnel
at about 30 meters under the surface when the underground required it as for a section of
line 7. In order to lower the costs of civil engineering, the most part of which was due to
stations construction, the farther the center the longer the inter-stations (from 800
meters to above 1 500 meters). Among the 20 connection stations, two are triple and
one quadruple, a real congestion node in the system at the eastern periphery. The meal
total investments thus turned around sixty million US $ a kilometer, situating the
Mexico City Metro in the lower third of the global range of building costs, ranging from
1 to 10 depending to the sites and estimations (Armstrong Wright, 1986 ; Fouracre & al.
1987 ; Hugonnard, 1990). The surface option was a relatively expensive work because
of the networks deviations, of the uneven crossings at junctions with general traffic, and
of the numerous pedestrian footbridges built to compensate the cutting effects of the
areas crossed. Concerning the only civil engineering the cost of surface works was
inferior by 5 % than the covered cutting. On the opposite, the elevated option was 25 %
more expensive than the ground option, and the tunnel option was 30 % more
9
expensive. Considering that civil engineering represents one half of the investments,
those cost differences between the type of works are blurred.
Caractéristiques du réseau de Mexico
Lignes Linéaire Stations Rames Interval. Site
km nombre nb. HP sec. dominant
1192037115 Tunnel
2242438130 Tun./Surf.
3242140125 Surf./Tun.
411107350 Viaduc
5161313260 Surf./Tun.
615118350 Tunnel
7141414260 Tunnel
8201928165 Surf./Tun.
9161218170 Viad./Tun.
A171018170 Surface
Métro 176 154 221
Tr. lég. 13 14 17 240 Surface
Ens. 189 168 238 Sources: Stc, Ste
Table 2 : Mexico’s network characteristics
During the three first phases, the rolling stock was composed of trains with six to nine
pneumatic cars. The trains were directly imported from France. They were then locally
produced under licence using import or integrated production of components. Since the
production pace did not follow the construction pace, it was necessary to import 200
additional trains from Canada. The return towards the national monopoly was made in
parallel with some diversification of outside purchasing, in particular towards Japan and
Germany. By the end of the third phase, the 76 % national integration degree allowed
Mexico to produce a total of almost 1 000 cars.
If the track and electric power supply were produced locally, all of the electromechanic
equipment had to be imported : centralized command (two command stations),
automatic piloting, signals, telephone and communication, and other equipments for
pumping, lighting, ventilation, sound and automatic toll gates and escalators.
Electromechanics as a whole did not have any technological handicap and in a field of
low local knowledge, no real technological tranfer was achieved. Between 1967 and 1994
the French firms participation can be estimated at 17 %.
During the three first construction phases of the Mexico City metro the main change
was linked with the replacement of a turnkey system by a growing national integration
system, from engineering to bogies. As regards the network (figure 4), the first set of
lines have been extended towards the limits of the central agglomeration, with more or
less success depending on the lines.
10
Source : http://www.webspawner.com/users/mopliq/
Figure 4 : Urban rail network of Mexico in 2000
11
4. INTRODUCTION OF VARIANTS INTO THE MEXICO CITY BACKBONE
During the fourth implementation phase of the metro in Mexico since middle eighties,
significant variants of financial, technological, institutional and urban planning type have
been introduced into the basic system. A required objective was to produce a cheaper
metro during the financial crisis. Moreover it was necessary to invest in mass transport
rail systems implementation in the province agglomerations the needs of which had been
sacrificed for the benefit of the capital. At the technical level, alternative solutions to
traditional urban metro were looked for through the exploration of what is globally called
"light metro". At the institutional level, the national and international components of
urban rail systems production were diversified and the investment distribution between
the various transport public firms was improved. At last, the question of the areas
thrown off the center, maintained apart according to the prevailing design of a backbone
reduced to the central region bounded the limits of the federal district which did not
include more than half of the megalopolis population.
That phase (during which two additional systems were implemented in Guadalajara and
Monterrey) was characterized by three achievements : line 8, a traditional pneumatic
metro, with two third of underground section and one third on surface, equipped with a
new piloting system (the Sacem, produced by Matra for the RER in Paris, was
introduced in Mexico from line A) and two surface lines with less capacity called "light
metro", with metallic rolling. They provided a connection service with the basic system.
The design and implementation of the light railway have been the subject of significant
local debate and international negociations. The result was the construction of two lines
which were redefined through an eventful process with work interruptions during the
fiscal crisis of ddf.
The "light railway" line was produced for Ste, a public company which had been
abandoned during the previous implementation phases of the metro. On a 13 km section
it is connected with the metro network at the eastern terminus of Xochimilco, a middle
class residential area now integrated in Mexico City agglomeration. It's a former proper
site tramway line with about ten junctions and a mean interval of one kilometer between
stations. Two steps were necessary to renew it ; the first one consisted in pouring a
concrete track on protected site and to put renewed rolling stock in service ; the second
one consisted in a reconstruction of the road system on a combined site and to produce
new rolling stock. Today, 17 double vehicles are in service on that site. They are running
with catenaries power supply and tramway type operation system. The lines offers a
capacity of 5 000 places a direction at peak hours. This offer could be double when the
site will be rehabilitated and if road modes competition is better controled.
The "iron metro" while diverging from its master plan, has become the line A of the Stc
Metro. On a 17 km section it is connecting a big exchange terminus with La Paz, a town
outside of the Federal District. A section of 4,5 km is running inside the neighbouring
jurisdiction, the State of Mexico. Diverging from the light railway initial idea, the iron
metro was implemented on surface on exclusive proper site along the central lane of the
way out from Mexico City towards Puebla. It induced a subsequent planning of the road
system and the development of crossing works. With a mean interval of 1 900 meters
between stations, it is running towards the South along the most significant popular area
of Mexico City. However, it does not serve that area (Nezahualcoyotl, 2,5 million
12
inhabitants) and reaches the end of a new growing area (Chalco Valley). At present 20
trains are operated on line A. They are composed of 6 metallic rolling cars with catenary
power supply. It is equipped with an autonomous centralized driving station and a very
performing operating system. Il could reach intervals of 105 seconds. Its present
capacity is about 15 000 places a direction at peak hours. The offer could be four times
more. The condition would be to double the number of trains. Moreover, the demand
should justify it if road transport offer did not absorb it due to the development of side
lanes on the renewed Zaragoza Avenue.
Do light railway and iron metro correspond to the design criteria of light metro ? Do
they introduce any intermediary capacity alternative solution into the Mexico City
urban railway system ? The answer is positive from the site viewpoint, since the two
lines are running on surface and replace existing tracks. Moreover, they are running
outside from the crowded central area. In the first case, the line partly runs at the same
level as general traffic with intersections, while in the second case, the line is on integral
protected site. As regards rolling stock, the answer is less clear while both lines are
equipped with metallic rolling. The equipment comes from local production with
German, Japanese, French and American components. The light railway typically is a
tramway. Iron railway cars are equipped with minimal equipment corresponding to
tramway cars with intervals of three minutes. The iron metro is equipped with an
operating system calibrated for regional services. Consequently, there are enormous
variations in theoretical capacities. While it was expected to offer 35 000 places a
direction at peak hours, the light railway overflows the admitted limits of the modern
tramway. The iron metro can offer up to 60000 places a direction at peak hours. That is
what situates it in the metro range. However, in their present operation situation, with
an offer of 5 000 places a direction at peak hours for the first one and 15 000 places for
the second, they would correspond to intermediary systems for developing cities.
In order to make comparisons, two other Mexican achievements can be taken as
references. They also are inspired from the light metro concept. The first line of
Guadalajara (2,5 million residents in the State of Jalisco), ie. 15,5 km with 6,5 km in
tunnel, involves the operation of 16 trains. Those trains are similar to the Mexico light
railway trains. They are designed for a theoretical capacity of 18 600 places a direction
at peak hours. However today, the operating conditions correspond to 4 100 places.
The first line of Monterrey (the capital of the northern border, 3,5 million inhabitants) is
running on 17,5 kilometers on viaduct with 25 trains coming from the local production,
according to the same prototype as the Mexico and Guadalajara iron metro. The offer
corresponds to a theoretical capacity of 40 000 places a direction at peak hours. The
present operation offers 10 000 places. In those four examples, the variants introduced
as compared with the traditional metro are more affecting the progress in equipment and
rolling stock allowance than the theoretical capacities, calibrated according to higher
standards than those of the European light metro. If the constraints of the fourth phase
induced a significant reduction of the investments costs, the various metro variants
should be compared as regards the unit investment costs.
5. EVOLUTION OF THE MEXICAN CONSTRUCTION COSTS
The debate about metros in developing cities often is limited to financial considerations :
how high can the implementation cost be estimated for this type of construction ?
Which are the variants depending on the experiences and the type of systems
implemented ? Metro costs cannot be estimated with accuracy because its
13
implementation may last several decades and involve multiple financing sources together
with a lot of suppliers and service companies. International comparisons are very
difficult to make because each system has its own characteristics, and the construction
modalities and paces may be very different. Moreover estimations are depending on the
local and international currencies variations. This makes a risky exercise to estimate and
display investment costs as regards the various alternative solutions, even if the Mexico
City railway system is a relatively homogeneous.
Within less than three decades, the Mexican Government has invested about twelve
billion US $ in the construction of the urban railway network in Mexico City. The
figures were collected and controled on the basis of several direct French and Mexican
sources (60 billion francs with a reliability degree of 5 %. In 1994 this amount almost
represents 3,5 % of the interior gross product, about 9 % of the total exterior debt of the
country and 15 % of the public debt). This figure does not include financial expenditure.
A quarter of the total invoice of this transport infrastructure was financed by exterior
credits. At the economical level, the validity of this investment is not put in question : it
was ratified by the calculation of the investment return rate according to a model devised
for developing countries metros (Allport & al. 1990). At the social level, this investment
can be related with the 85 million Mexicans who support this national priority work
including the 16 million inhabitants of the capital and the seven million living in the
limits of the Federal District where the system is almost totally included (while a third
of the users are coming from the state of Mexico). But the total amount of capital
engaged is significant enough to make the a posteriori estimation concerning the various
achievements.
Investissement total et unitaire par
p
hase du métro de Mexico en m° US $
Phases Coûts totaux Investissemen
t
construction trains total au km
I1 924 972 2 896 69
II 4 131 984 5 115 70
III 1 032 431 1 463 57
IV 1 754 358 2 112 42
Ensemble 8 840 2 745 11 586 61
Source : Inrets (Covitur & contrats) Élab° F
(Henry : 1988 & 1994)
1O/94 : 1 US$ = 5,21 FF
Table 3 : Total Investment and by step of metro’s Mexico en M US $
Table 3 shows that the effort consented to build a network at sustained pace (figure 1)
was significant during the first phase and even more dureing the second one. The costs
tied up in the construction doubled and later on suddenly decreased in the context of the
debt crisis. The three quarters of this global investment were attributed to the
construction according to priority option in favour of the network extension over the
rolling stock. If we refer to the number of construction kilometers, the two first phases
were the most significant at 70 million $ a kilometer. Control gained as regards civil
engineering made the kilometer cost lower during the third phase and the introduction of
variants lowered the cost even more during the fourth phase.
14
Investissement du métro de Mexico par ligne
1967-1994 en millions de US $ (10/94)
Exp./lignes Km Const° Trains Total T. $/km
Stc/ligne 1 19 1 028 545 1 573 83
Stc/ligne 2 24 819 551 1 370 57
Stc/ligne 3 24 1 346 528 1 874 78
Stc/ligne 4 11 614 107 721 66
Stc/ligne 5 16 979 161 1 140 71
Stc/ligne 6 15 651 94 745 50
Stc/ligne 7 14 813 177 989 71
Stc/ligne 8 21 1 199 194 1 393 66
Stc/ligne 9 16 836 225 1 062 66
Stc/M° fer A 17 516 145 660 39
Ste/Tr. léger 13 40 19 59 5
Ensemble 190 8 840 2 745 11 586 61
Table 4 : Investment of Mexico’s metro per line
Table 4 shows some variation according to the lines and works, reflecting their
characteristics together with their implementation period. The most expensive are the
original central lines, except line 2 which was running on the former tramway line. Lines
4 to 7 were built during the second and third phases at lower cost, and it may be
explained by mexicanization and by the rolling stock low cost. Line 6 is the cheapest in
the traditional system but it also has the lowest performance because of its voluntarist
layout running from East to West along the northern side of the Federal District. With
lines 8 and 9 the mean cost is 66 million a kilometer.
The two last lines (table 4) were built at a much lower cost than the other works. That
could justify their name "light metro". However it can be noticed that the non metro cost
is 22 % lower than the cheaper pneumatic metro on line 6. But it is eight times more
expensive than the light train, the investment of which is particularly low (but half of
the work consisted in renewal).
Coût comparé en place offerte
du métro de Mexico et variantes
(US$94/Passager/km/heure) ∆O/T
Train léger Théorique 843
Ste Observ
é
1 17
0
28%
Métro fer Théorique 870
Ligne A Observ
é
2 59
0
66%
Métro Théorique 1 400
Ligne 8 Observ
é
2 13
0
34%
Source : Inrets (Henry : 1988 & 1994)Élab° FK
Table 5 : Offered seat compared cost of Mexico’s metro and variants
15
In order to make an accurate comparison, the three works of the fourth phase have been
chosen. They respectively correspond to traditional metro, iron metro, (both having the
same operating system) and light railway (table 5). We have established the theoretical
cost corresponding to the capacity defined on project - the comparative costs of the
theoretical number of places offered at peak hours would be the same for both systems
of "light metro", ie. one third cheaper than traditional metro costs. We then have made
the same comparison according to the present train supply and the operation system. If
the light railway is cheaper, the cost of the places offered in the iron metro is quite
higher than the cost of the traditional metro. If we compare the values observed with the
theoretical values, the cost difference of 66 % is confirmed against the iron metro, while
the two other systems are operated at the two thirds of their performance. The
differences of performance between the metro and its variants are not very significant
except as regards the tramway, the capacity of which is much lower. It also can be
observed that system calibration beyond its present operation conditions is expensive in
the short term. That is the paradox of the light metro in developing cities. In order to be
progressive they have to keep significant reserves. It leads to increase the immediate
investment in offered places without being ensured that those reserves will be justified
in the future.
Structure comparée des coûts de construction du métro de Mexico et variantes
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Train léger Ligne A Ligne 8
Matériel roulant
Foncier et taxes
Maîtrise d'ouvrage
Électromécanique
Ouvrages annexes
Génie civil direct
Figure 5 : Cost compared structure of Mexico’s metro construction and variants
However, structures differences remain in the costs of the three systems (figure 5).
Direct civil engineering (without taking into account accompaniement works) represents
30 % of the costs in the two light metros while it represents 45 % in the traditional
16
metro. Inversely, the relative cost of the rolling stock only represents 30 % in the
traditional metro while it is above 40 % in the renewed tramway. The tramway estate
cost is near zero while it represents 10 % in the other systems. At last, the part of
electro mechanics which represents 16 % in the light railway progressively decreases
and just represents 8 % in the traditional metros costs. The relative difference are
essentially made by civil engineering and estate and they depend more on the site than
on technology. It is possible to build a metro at low cost when a surface is available and
if the work consists in a rehabilitation ; but data on table 4 recall that benefits from the
operation results are not guaranteed.
6. IMPACTS ON THE MEXICO CITY TRANSPORT SYSTEM
The impacts of a metro in developing cities are more visible at the sectorial level than
urban planning impacts (Henry 1986a) ; it is the case in Mexico where the metro
development program has been carried out quite independantly from the megalopolis
urban planning. It cannot be denied that the implementation of a system which provides
every day a rapid service to 4 million citizens has an impact on the urban environment
and the constructed framework. However at a similar scale, the Sao Paulo megalopolis
only represents one quarter of the Mexican network. Besides its urban impact, it is
reasonable to wonder which impacts the metro can have on transport.
Fréquentation du métro de Mexico par ligne en 1994
1
23%
2
24%
3
20%
4
2%
5
6%
6
3%
7
6%
8
2%
9
8%
A
6%
Figure 6 : Mexico’s metro frequentation by line in 1994
First it can be noticed (figure 6) that the success of the Mexico City metro always has
been based on the three first lines which still were very crowded in 1994 with 67 % of
the travellers. This concentration slightly increases with time but line 9 was the only one
to produce a real emballasting. It cannot be said that the construction of the other lines
substantially has modified the network equilibrium, all the more because the internal
17
connection only concerned 40 % of the metro trips. The uneven distribution of the
metro patronage is asking a double question about the relevance of the choices in terms
of lines number, extension and layout. Compared with the lines in indutrialized
countries, the patronage on the less performing lines in Mexico is honourable ; but is it
justified in the present case ? The a posteriori evaluation just can be made with
detachment and the time elapsed in Mexico is not sufficient to make an accurate
judgment.
Mexico 1969-1993 : fréquentation journalière moyenne des
modes ferroviaires
0
500 000
1 000 000
1 500 000
2 000 000
2 500 000
3 000 000
3 500 000
4 000 000
4 500 000
5 000 000
1969 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993
Métro pneu
Métro fer
Tramway
Figure 7 : Frequentation of Mexico’s rail means 1969 - 1993
It can be observed that line A, an outskirt service line has developed more quickly than
the traditional lines. After three years of operation it represented 6 % of the demand.
There is a significant potential demand which clearly appears in the dramatic crowd at
the bus/metro integrated stations. Moreover, the fact that 35 % of the terminus stations
of the network (vs. 55 % at internal stations and 10 % at connection stations) reveals a
gap between two functions of the metro : internal services in the city and satisfaction of
the outskirts demand. In the Mexican case, can one system, the metro system perform
both urban and regional transport function (which are respectively performed by the
metro and the RER in the case of Paris) ?
Then it can be observed that if transport demand regularly increased during the three
first phases, it reached a peak in 1989 and has totally decreased since then. This change
led to questions. The demand behaviour does not seem to be linked with tariff either.
Tariffs have remained very low until 1984 (the covering rate of operation expenses by
tickets represented 5 %). They have known for long substantial increases without direct
effect on patronage (in the frame of a social policy compensating the effects of structural
adjustment on workers, the covering rate still is 38,3 % in 1994). The drop in patronage
observed during the four last years poses questions on the sustained pace of
construction when the demand is developing without depending on the successive
openings of the new lines (figure 8). What is striking is the significance of the gap
18
between the number of kilometers put in service during the five last years and the
decrease of patronage on the network.
Accroissements annuels du réseau et de la demande du métro de Mexico
1969-1994
-10
-5
0
5
10
15
20
1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994
Kilométres mis en service Passagers transportés par jour
Figure 8 : Annual increase of Mexico network, and of metro’s demand 1969 - 1994
According to very approximative figures which could be gathered about the development
of trips modal split over the 25 last years (if metro and public firms data are reliable, the
estimations concerning mini-buses and other private modes are hazardous) the prevailing
goal of the metro concerning transport offer in the Mexico agglomeration is not totally
reached. An increased competition with road modes even can be observed, namely with
mini-buses. Mini-buses have expanded considerably over the last years (while they may
be over-evaluated in figure 9). The connection between metro and buses is not a real
success since Ruta 100 public firm has faced a crisis (dropping from 6 000 to 3 500
buses). And the number of individual vehicles regularly increases despite the recent
measures adopted in order to restrict car use and to protect the very polluted air in
Mexico City.
19
Modes de déplacement Mexico 1971-92
0
5
10
15
20
25
30
1 971
1 974
1 977
1 983
1 986
1 989
1 992
Millions/jour : TC
0
5
10
15
20
25
30
35
40
45
50
Millions/jour: VP & Total
Métro & Tram. Bus & Trolley Minibus & Taxi V.P. & Autres Total
Figure 8
Sources : Inrets, C
g
t, Navarro Élab° EH
Figure 9 : Riding means in Mexico 1971 - 1992
If the structuring effects of the metro have to be assessed on the long term, however it is
interesting to measure the short term consequences a high investment. The saturation of
the three first lines, calibrated under the present levels of demand induces a decrease of
patronage among users who do not use other lines. Considering the present demand
situation the choices made during the first phase may be discussed in terms of capacity
more than in terms of layout. The following phases are in the opposite situation. Is the
network being abandoned or do travellers use other modes ? The example of line A
illustrates the difficulties in the metro. At one end it arrives at one of the most saturated
stations of the metro network (Pantitlan over 350 000 passengers a day). At the other
end it only serves in a tangential popular residential areas. The construction of the metro
has led to a significant renewal of road works providing direct links between the areas
served by mini-buses and employment areas without connection or time losses. Thus, a
large capacity is faced with small size vehicles. That situation also occurs on the other
lines of the network. However, the increase of mini-buses number is not a natural trend
if we consider the sudden stress of the 1989 mini-buses increase curb. 1989 was marked
by a change in transport policy applied by Ddf, further to a significant strike of buses
which crippled the city and to the Ruta 100 crisis. Since then in view of liberalization
the option has consisted in an institutional and financial support in favour of the mini-
buses sector at the expense of public firms as a whole and of the metro in particular. It
reaffirms the need of consistency of the three dimensions of multimodal integration
with transport policy (Henry ice 1990).
20
7. METROS COSTS AND VARIANTS IN THE WORLD
Mexico is in favour of the metro variants which are called "light metro" (Barry, 1991 ;
Booz-Allen, 1991 ; Gardner & al. 1994). In industrialized countries, the functions and
capacities of the system are arousing interest for this alternative technology. That is true
in particular as regards mass transport systems equipments in cities the size of which
would not justify the implementation of a metro or as regards exceptional low demand
services linked with bigger size systems. The developments made are close to the
tramway concept, a modernized equipment regarding the rolling stock and operation
modes (Kuhn & al. 1992). In the towns under 1 million inhabitants, systems are
implemented on proper site with a capacity of about 5 000 places a direction at peak
hours, at costs between 10 and 35 million $ a kilometer. In the context of developing
countries interest mostly focuses on cost problems. It means that a light metro would
mean lower investment even beyond the need threshold of a heavy metro (Barbieux &
Kuhn, 1990). There is a tendency to expect similar functions and capacities
performances from the light metro as from the metro. Cities up to 7 million inhabitants
don't invest more than 20 million $ a kilometer in systems ranging from 8 000 to 20 000
places a direction at peak hours but could reach 50 000 places if the traffic and number
of trains were substantially increased.
If the experience in Mexico significantly differs from the definition of "light metro" the
Guadalajara and Monterrey experiences are more similar to Mexico. In the first city, an
existing tunnel was exploited to build a 15 kilometers North-South line, on the former
tramway layout which was changed in proper site with protected level interactions : in
those conditions the cost remained relatively low and can be estimated at seven million
1994 $ a kilometer. But later on the construction of the East-West line (8,5 kilometers
underground) made the reserve capacity requirements increase (three cars trains, over-
dimensioned platforms, operation able to reach 28 000 passengers an hour per direction)
and the construction cost double. The Monterrey experience also is a light metro
experience on viaduct, the capacities of which will increase on the long term (45 000
places a direction at peak hours vs. 10 000 presently). The unit cost would be half of
the iron metro cost in Mexico. In the Mexican case local variants are looked for (rolling
stock is produced in Mexico). The aim is to be as close as possible to the metro as
regards the expected capacities, and keep the cost level of light metros.
It also is the case in Medellin, Manilla, Istanbul, or Tuen Mun, where a tension can be
observed between an already high demand which is expected to grow significantly during
the next decades and where saturation is rapidly reached in the operation conditions
allowed by a minimum investment. The situation is different in tunis where a network of
four modern tramway lines has been produced and is similar to the Belgian and French
achievements (Godard & al. 1994). It faces problems of interference with road traffic
which remains a major problem in developing cities.
Is it possible to compare those achievements with oneanother and with the
achievements in industrialized countries ? The exercise is difficult because of the local
specificities, the lack of consistent data and the monetary actualization and conversion
(Henry 1988a). Comparisons may be made if exceptional information sources showing
too many variations are not taken into account, if financial costs are excluded, and if the
objective is to obtain comparative values and not nominal values which always are
21
questionable. The most commonly used criteria is the kilometer cost including the rolling
stock (Hugonnard, 1990).
Coûts kilométriques de 34 métros
0
50
100
150
200
250
MRT São
Paulo
MRT Hong
Kong
MRT
Caracas
MRT
Marseille
MRT
Baltimore
MRT Lyon MRT Lille MRT
Calcutta
RRT Le
Caire
MRT
Singapour
MRT Pékin LRT
Grenoble
LRT
Bobigny
LRT
Guadalajara
LRT
Monterrey
LRT
Manille
LRT Tunis
Km Réseaux et I00 000 habitants
0
20
40
60
80
100
120
140
160
180
200
Millions US$94/Km
Réseau km
100 000 hab.
Coût/km
Figure 10 : Cost per kilometer of 34 Metros’networks.
To make a posteriori comparisons sets methodological problems. On the basis of
information concerning some lines or sections of each system, a table can be made,
showing the kind of results which can be observed in that kind of analysis (figure 10).
Out of 34 selected situations (which are not including such important systems as Tokyo
Moscow or Montreal) it should be recalled that the size of the cities ranges from 400
000 to 16 million inhabitants and their metro sizes from 15 to 201 kilometers (an average
of 50 km). The histogram on figure 10 comparing the extension of the networks with
population in agglomerations, illustrates the advantages of the situation of industrialized
countries, while developing cities are overcrowded as compared with the existing metro
offer. Having made those remarks, the cities can be classified according to their kilometer
cost. Except in Washington, the five systems with the highest unit costs (between 140
and 180 million $) are not the systems with the largest networks (while they include
megalopolis such as Sao Paulo and Rio de Janeiro). One third of the reference cities have
kilometer costs ranging from 50 to 100 million $ a kilometer : they may correspond to
very large networks (while London and Moscow with almost 400 km are not
represented here) and to metropolis such as Peking, Le Caire or Calcutta, where the
metro seems to be very small as compared with the population. Thirteen cities which
chose light metro are in the lower side of the curve : the middle sized cities in
industrialized countries, and the metropolis in developing countries. At last, the metro in
22
Asia seems relatively cheaper than the others, except in Hong Kong. If the decision to
build a metro was made on the basis of those considerations, it would be easy... or very
efficient if the cheapest system were a road system rather than a rail system (Armstrong
Wright, 1986). But if another comparison criteria is adopted such as the cost of the
place offered at peak hours, which better corresponds to the functions of a metro, the
results are quite different.
An alternative method consists in estimating the offer in present conditions on the basis
of the maximum headway accepted by operating systems, by the road system
characteristics and the number of trains (figure 11). In order to make comparisons, the
comfort levels are reported to the levels admitted in developing cities (8 passengers a m2)
and the train composition together with the differences between peumatic and metallic
rolling stock are considered. The capacity which is calculated in number of places
offered at peak hours varies between two poles of 4 000 and 100 000 passengers an
hour and a direction at peak hours. It marks the difference between the brand new
tramway in Rouen and the metro in Hong Kong. If this capacity is reported to the total
construction investment, a "capacitive cost" is obtained ranging from 1 to 10 (725 $ for
the light metro in Manilla vs. 8 500 in Rouen).
It appears (on the upper curve on figure 11) that the range of the cities according to this
ratio is quite different from the range according to the kilometer cost (curve on figure
10). Metros such as in Lyon, Marseille or Santiago, and also light railway trains such as
Nantes, Strasbourg or Atlanta, with a low kilometer investment show quite the highest
capacitive costs. Inversely, the expensive character of the metro kilometer in Sao Paulo
(2 700 US $) or in Caracas (1 650 US $) is strongly compensated when reported to the
capacity. Baltimore and Rio de Janeiro respectively are situated at 80 and 180 million $
a kilometer but their capacitive cost is about 3 900 US $.
23
Coûts capacitifs de 34 métros
0
20
40
60
80
100
120
140
LRT Rouen LRT
Strasbourg
LRT Nantes MRT São
Paulo
LRT
Istanbul
MRT Pékin LRT
Monterrey
RRT Le
Caire
MRT Séoul
Capacité actuelle et théorique (1000 phs en pointe)
0
1
2
3
4
5
6
7
8
9
Coût place offerte (1000 US$94 p/h/s)
Capacité actuelle
Capacité théorique
US$/Place Actuel
US$Place Théorique
Sources: World Bank, Booz-Allen, So
f
retue & Melte Élab° FK
Figure 11
Figure 11 : Capacity costs of 34 Metro’s networks
Considering the capacities histogram (in black on figure 11), a significant offer makes the
offered place cost lower, all the more if the investment is low (Paris and Seoul). Finally,
the capacitive costs of the light systems are not so clearly lower ; for instance the light
metro in Istanbul has the same cost as the automated metro in Lille (about 2 300 US $)
or in Guadalajara (1 750 US $) which can be compared with the Hong Kong light
railway. In this list, after Paris (1 900 US $) and Hong Kong (1 800 US $) the systems
are developing cities heavy or light systems. Among them Mexico confirms its good
capacity/cost relation.
Railway systems also are developing in developing cities, most of which are recent and
where needs are keeping on growing. The present capacity can be improved by the
through headways, number and setting of trains (and don't requires change in the
stations civil engineering). The theoretical capacity obtained is presented when projects
have to be chosen. On the lower curve on figure 11, the capacitive cost is significantly
reduced (in Istanbul, Guadalajara, Monterrey or Tunis) in particular as regards light
systems intended to develop (it could drop to 290 US $ in the case of Manilla).
However, nothing ensures that the demand will justify such additional investments. On
the opposite, in big metropolis, in industrialized countries, and in cities where operation
is close to saturation, less gains are to be expected in the future.
Cities of several million inhabitants even with poor GIP not only need metros and
intermediary capacity systems : their extension, spatial segregation and centre-outskirts
tensions justify regional rail systems to be implemented. Some of them adapt the
24
services created before urbanization (in Rio de Janeiro or Buenos Aires). Very few of
them build systems which go the the outskirts like in Cairo (at a kilometer cost of about
55 million US $, similar the RER in Paris extra muros despite a lower capacitive cost).
They don't build either trains such as in Porto Allegre, Recife and Belo Horizonte, the
layout choice of which proved to be questionable. The case of Mexico City line A is
designed as "light metro" and runs according to the same criteria as the metro. But its
implementation and site are close to RER ones. The results show how difficult it is to
combine the advantages of those three types of technologies or urban rail transport even
if some may think that in such a megalopolis it's a TGV that is to be considered to
ensure the services of satellite cities (Musset 1995). But it also may be considered that
the variations of the basic metro design have regrettable consequences. They produce
hybrid products which cannot satisfy the multiple objectives assigned to them and they
don't serve any. However investment in light systems is two or three times lower than
the metro in terms of constructed kilometer. Paradoxically its capacitive cost is 40 %
higher than the metro capacitive cost in industrialized countries while it is 30 % lower in
developing cities. Is a single list useful to compare such contrasted realities ?
8. INITIALS AND ACRONYMS
Cgt, Coordinación general del transporte (autorité organisatrice du Ddf de Mexico)
Covitur, Comisión de vialidad y transporte urbano (maître d'œuvre délégué du Ddf)
Ddf, Departamento del Distrito federal (directement rattaché à la présidence mexicaine)
Ica, Ingenieros consultores asociados (principal maître d'ouvrage du métro de Mexico)
Lrt, Light rail transit (ou métro léger, tramway moderne, U-Bahn, etc.)
Mrt, Mass rapid transit (ou métro, underground, subway)
Ratp, Régie autonome des transports parisiens (exploitant du métro, liée à la Sofretu)
Rrt, Rapid rail transit, ou Rer, Réseau express régional, Train suburbain
Sacem, Système d'aide à l'exploitation, à la conduite et à la maintenance
Sofretu, Société française d'études et de réalisations de transports urbains (Systra)
Stc, Sistema de transporte colectivo (exploitant du métro de Mexico), Ddf
Ste, Sistema de transportes eléctricos (exploitant du tramway et des trolleybus), Ddf
9. BIBLIOGRAPHIE
Allport R., Thomson J., Halcrow Fox & Ass., 1990, Study of mass transit in Developing countries,
Contracted report 188, Trrl, London (non publié)
Armstrong-Wright, 1986, Urban Transit Systems: Guidelines for Examining Options, World Bank
Technical Paper n° 52, Washington
Balat J-L. & Vince P., 1990, "Planification des transports de l'aire métropolitaine de Mexico city :
un exemple de la forte présence mondiale de l'ingénierie française", in Tec n° 103, 11-12/90
Barbieux C. & Kühn F., 1990, Les atouts des métros légers dans une politique de développement
des transports collectifs : critères de choix techniques et financiers, in Actes de la Codatu V São
Paulo
Barry M, 1991, Through the cities: the revolution in light rail, Frankfurt Press
Booz-Allen & Hamilton Inc., 1991, Light Rail Transit Capital Cost Study, prepared for Umta,
Washington, 70 p. (non publié)
Coindet J-P., 1989, "Les métros de Mexico et du Caire : enseignement comparé de deux réalisations
dans les Ped", in Recherche transports sécurité n° 24, p. 13-22
Connolly P. & García S., 1994, Evaluación de los "metros ligeros" de la ciudad de Mexico : estudio
financiero y de impacto local, Informe para el Inrets, Mexico (non publié)
25
Figueroa O. & Henry E., 1987a, (ss. dir. de, avec Clichevsky N., Galilea S., González O., Maurel D.,
Morales S., Müller A., Navarro B., Nassi C., Orrico R., Pinheiro Machado D.), Évaluation des
métros latino-américains, Rapport de recherche pour le Mres, Inrets/Redes, 14 vol. (non publié)
Figueroa O. & Henry E., 1991a, “Éléments de débat sur les métros en Amérique latine”, in
Recherche transports sécurité n° 21, 3/89, p. 5-10
Figueroa O. & Henry E., 1991b, Analysis of the Underground Systems in Latin America, in Heraty
M. (ed.), Urban Transport in Developing Countries. Lessons in Innovation, London, Ptrc,
Perspectives 1, p. 232-240
Fouracre P., Allport R. & Thomson J., 1990, The performance and impact of rail mass transit in
developing countries, Trrl, Research Report 278, 28 p.
Gardner G., Rutter J.& Kühn F., 1994, The performance and potential of light rail transit in
developing cities, Project Report 69 R5596, Trl, Londres
Godard X., 1987 (ss. dir. de, avec Coindet J-P., Huzzain A. & Kühn F.) Évaluation a posteriori du
métro du Caire, Expertise pour le Mef-Direction du Trésor, Inrets/Drtpc, 3 vol., 140 p. (non
publié)
Godard X., 1994 (ss. dir. de, avec Kühn F., Baltagi H., Boukhris T. & Baltagi A., Évaluation a
posteriori du métro léger de Tunis, rapport Inrets-Smlt
Henry E., 1987, L'insertion des métros dans l'offre de transport en Amérique latine, in Actes de
Métropolis II, Congrès de l'Association mondiale des grandes métropoles, Mexico, 19-21/5/1987,
20 p.
Henry E., 1987a, Principales interrogations sur les métros dans les pays en développement, in
Quelles politiques de transport urbain pour les pays en développement ? Débat autour des options
de la Banque mondiale, Journée spécialisée Inrets, 2/6/87
Henry E., 1988, (ss. dir. de, avec Coindet J-P., Figueroa O., Godard X., González O., Hiernaux D.,
Kühn F., Maurel D. & Navarro B.), Évaluation rétrospective du métro de Mexico, Expertise pour
le Mef-Direction du Trésor, Inrets/Cenvi, 3 v., 350 p. (non publié)
Henry E., 1988a, Quels impacts produisent les métros sur les transports et les villes latino-
américains ?, in Actes de la Codatu IV, Jakarta, 5-10/6/88
Henry E., 1994 (ss. dir. de, avec Connolly P. & Kühn F.), Évaluation des variantes du métro de
Mexico : ligne A et train léger, Expertise pour le Mef-Direction du Trésor, Inrets/Cenvi, 184 p.
(non publié)
Hugonnard J-C., 1990, Intérêt des métros dans le monde, Sofretu, 20 p. (non publié)
Ice, Inrets/Trrl, 1990, Rail Mass Transit for Developing Countries , London, Thomas Telford, 393
p.
Jane's, 1995, Urban Transport Systems, XXX° edition, London, Chris Bushell
Kühn F., Martinet F, Marx P. & Constantin B, 1992, Études comparatives des coûts de génie civil
selon les systèmes de transport adoptés, Inrets-Metram-Cetu, Rapport Afme, 258 p.
Kühn F., 1994, Comparison and evaluation of the new transit systems, in Proceedings of the
International Conference on the New Transit Systems, The Korea Transport Institute, Pusan, p.
27-58.
Musset A., 1995, "De la pirogue au Tgv : les transports urbains à Mexico", in Transports urbains
86, p. 5-16
Navarro B., 1994, El metro y sus usuarios, Uam, Mexico
Parpillon J-C. & Larraufie G., 1992, "Sofretu et le métro de Mexico. 25 ans d’une collaboration
exemplaire", in Revue générale des chemins de fer, 10/92.p. 49-59.
Vuchic V., 1988, "Design, evaluation and selection of transit systems : theory and practice", in Les
transports collectifs urbains : un défi pour nos villes, Colloque Enpc, Paris, 3-5/5/88
World Bank, 1986, Urban Transport: A Sector Policy, Washington (et 1995: Sustainable Transport:
A Sector Policy Review -document de circulation interne- non publié)
ResearchGate has not been able to resolve any citations for this publication.
Technical Report
Full-text available
This report looks at the performance of Light Rail Transit (LRT) systems and tramways in developing cities. Case study and estimation procedures have been used to examine the maximum passenger carrying capacity of various types of LRT. This report complements other work by TRL on metros (Fouracre et al., 1990) and on high capacity busways (Gardner et al. 1991). The traditional tramways surveyed in India, Egypt and China all carry flows of less than 7,000 passengers par hour per direction (pphd). The more modern systems in Tunis. Budapest and Alexandria carry up to 13,000 pphd. Only the system in Manila, which is elevated (and hence segregated) along its entire length ca carry flows over 18,000 pphd. The main problem for LRT in developing cities appears to be the level of interference fro other road users, together with the difficulty of controlling and organising a regular headway service. The inability of vehicles to overtake when others are delayed, and long turn around times at terminals compound the problem. Despite their technical limitations under developing country conditions. LRT systems are very popular, and can have advantages in improving the quality of life and civic pride. They should not however be seen as an ideal solution for a poor developing city.
Conference Paper
Full-text available
Cette communication porte sur un choix modal spécifique, le métro léger : son évolution et ses perspectives. Elle s'intéresse à un certain nombre de critères essentiels d'ordre urbanistique, technique et financier. Elle s'articule autour de deux points : I. L'évolution et les perspectives du métro léger dans les pays en développement. II. L'approche méthodologique et les critères de choix.
Article
There has been little recorded research on the performance and impact of metro projects in Third World cities. This report presents material which: reviews the experience of planning, building and operating rail mass transit systems in developing cities; and identifies the conditions under which investment in such systems might be justified. The work divided mainly into three parts: case studies of 21 developing cities with metros in operation or in prospect; establishment of a strategic traffic and evaluation model; and application of the model to some of the case study cities. The case studies provided data with which to furnish the model, and the model provided the tool for evaluating metros, either operational or projected. The application of the model to the cities completed the comparative analysis on which some general conclusions about metros in developing cities could be drawn. -from Authors
Coordinación general del transporte
  • Cgt
Cgt, Coordinación general del transporte (autorité organisatrice du Ddf de Mexico)
Departamento del Distrito federal (directement rattaché à la présidence mexicaine)
  • Ddf
Ddf, Departamento del Distrito federal (directement rattaché à la présidence mexicaine)
Ingenieros consultores asociados (principal maître d'ouvrage du métro de Mexico) Lrt, Light rail transit (ou métro léger, tramway moderne
  • Ica
Ica, Ingenieros consultores asociados (principal maître d'ouvrage du métro de Mexico) Lrt, Light rail transit (ou métro léger, tramway moderne, U-Bahn, etc.)
Mass rapid transit (ou métro, underground, subway)
  • Mrt
Mrt, Mass rapid transit (ou métro, underground, subway)
Réseau express régional, Train suburbain Sacem, Système d'aide à l'exploitation, à la conduite et à la maintenance Sofretu, Société française d'études et de réalisations de transports urbains (Systra) Stc
  • Ratp
Ratp, Régie autonome des transports parisiens (exploitant du métro, liée à la Sofretu) Rrt, Rapid rail transit, ou Rer, Réseau express régional, Train suburbain Sacem, Système d'aide à l'exploitation, à la conduite et à la maintenance Sofretu, Société française d'études et de réalisations de transports urbains (Systra) Stc, Sistema de transporte colectivo (exploitant du métro de Mexico), Ddf
Sistema de transportes eléctricos
  • Ste
Ste, Sistema de transportes eléctricos (exploitant du tramway et des trolleybus), Ddf