BookPDF Available

Rethinking Urban Practices: Designing for Jurong Vision 2050

Authors:
R e t h i n k i n g u R b a n p R a c t i c e s
Rudi StouffS PatRick JanSSen
Designing for Jurong Vision 2050
2
Published by the
Centre of Advanced Studies in Architecture (CASA)
Department of Architecture
School of Design and Environment
National University of Singapore
4 Architecture Drive
Singapore 117566
Tel: +65 65163452
Fax: +65 67793078
Rethinking Urban Practices: Designing for Jurong Vision 2050
ISBN: 978-981-09-9579-9
Date of Publication: September 2016
Printed by
First Printers Pte Ltd
© CASA, Centre of Advanced Studies in Architecture
Department of Architecture
School of Design and Environment
National University of Singapore
© Rudi Stous/Patrick Janssen/Individual Contributors
Sponsor
NUS-JTC i3 Centre
All rights reserved; no part of this publication may be reproduced,
stored in a retrieval system, or transmitted in any form or by
any means, electronic, mechanical, photocopying, recording, or
otherwise without prior written permission of the publisher.
The publisher does not warrant or assume any legal responsibility
for the publication’s contents. All opinions expressed in the book
are of the authors and do not necessarily reect those of the
National University of Singapore.
08 Introduction
10 EleVate
22 Performative Imagination
26 The Ascent
38 IFoU
42 Eden of Jurong
54 The Industrial City
58 Small Hill Town
72 Generative
76 Envisioning Ecotopia
90 Evaluative
94 Urban Metabolism
108 Conclusion
110 Credits
7
IntroductIon
rethInkIng urban
PractIces
This book rethinks urban practices, both its processes and
its products, within the context of JTC Corporation’s 2050
vision for the Jurong Industrial Estate (JIE). This vision entails
the transformation of JIE from an almost mono-functional,
segregated and fragmented, polluted industrial area into a major
catchment area for future population growth that integrates
clean(ed) industrial plants with green lungs, attractive housing
and vibrant urbanity for one million people.
The book presents selected student projects at
varying scales that propose innovative ideas on how JIE can
be re-imagined for a future Singapore. A rst set of student
projects is selected from the Final Design Project studio of the
Master of Architecture specialisation in Design Technology
and Sustainability at NUS. These projects present innovative,
prototypical designs addressing collocation and mixed
developments, with a focus on the Jurong Hill area and Pulau
Samulun within JIE.
A second set of student projects is selected from the
Master of Arts/Architecture (Urban Design) and Master of Urban
Planning studios at NUS. These projects serve as demonstration
for an urban planning and design workow that leverage new
computational tools and techniques in order to enable planning
and design options to be systematically developed, evaluated,
and analysed. The workow addresses complex challenges that
require new thinking about our city and its planning and design.
It supports the exploration and development of alternative
visions of the future city, and enables the evaluation and analysis
of dierent urban planning and design options.
All selected projects have their genesis in the International
Forum on Urbanism (IFoU) Winter School, held at the Department
of Architecture, NUS, in January 2015, in collaboration with JTC
Corporation.
eLeVate
Future transPort
sYsteMs For the cItY
koh teck WeI
FInaL desIgn ProJect 2014-15
tutor :
ProF. rudI stouFFs
ABSTRACT
The “EleVate” project proposes an elevated tramway system
as connectivity infrastructure for the transformation of Jurong
Industrial Estate from its current focus on industry into a major
catchment area for future population growth. As such, it can
be developed in advance of or alongside the transformational
developments integrating clean(ed) industrial plants with green
lungs, attractive housing and vibrant urbanity for up to one
million people.
1110
Tram station at bridge level
1312
The tram seamlessly integrates with other modes of transportation, especially,
means of active connectivity, such as pedestrian movement, bicycles, and other
current or future, personal, green transportation. Trams, pedestrians, bicycles, etc.,
all share the same physical space. The elevated deck also enables developments
along the infrastructure to reinforce the deck’s public connectivity function by
linking directly to it. The deck becomes the new ground level for pedestrians and
bicyclists.
3
Tram station plan at bridge level
1:100
Tram and MRT interchange plan at bridge level
1:350
Buildings
Green network
Elevated infrastructure network
Road network
Layers of the proposal
1514
Elevation of tram station crossing a park connector
Tram and MRT interchange at bridge level
1716
Tram and MRT exchange above Ayer Rajah expressway
1918
Tramway integrated with mall development
21
PerForMatIVe
IMagInatIon
desIgnIng For the
Jurong hILL area
A rst selection of student projects is taken from the results
of an M.Arch. Design Technology and Sustainability thesis
studio led by Rudi Stous, addressing JTC’s Jurong Vision
2050 with an emphasis on the Jurong Hill area and delivering
innovative, prototypical designs targeting collocation and mixed
developments and construction over air-space.
Due to a shortage of space, Singapore is experimenting
with providing developers air rights, e.g., allowing them to build
over a road or highway. However, building over a highway is not
just a legal issue, nor a structural issue of spanning the width
of the road. There are many other performative issues that may
play a role, such as noise pollution, air pollution, accessibility,
connectivity, safety, etc. Some of these issues might be avoided
by burying the road under the development. Other issues might
be solved adequately during the design materialisation and
detailing stages, considering material choices, technological
innovations, structural dimensions, etc. However, considering
selected issues as design drivers in the conceptual design stage
might yield much more imaginative and eective solutions,
merging form, function and materialisation in innovative and
synergistic ways.
In this guided M.Arch. thesis studio, students considered
one or more performative qualities as design drivers in the
exploratory design of a multi-functional building or building
cluster serving as an activity hub for a mixed-use industrial area.
The selection of design drivers is argued from a design idea
and supported by an investigation into (analytical and other)
means for assessing the respective performative qualities. These
investigations formed the research component at the start of the
studio, where each student investigated a performative quality
and shared the results of these investigations among the studio
participants.The site proposed by the studio is the surroundings
of the Jurong Hill Flyover on the Ayer Rajah Expressway (AYE).
The site is part of Jurong West, which is currently under
consideration by JTC Corporation to be transformed from an
2322
almost mono-functional, fragmented and polluted industrial
area into an attractive and healthy mixed-use area for clean-tech
industry and housing for one million people, amongst greenery
and water. The attractiveness of the site for the studio lies in
the proximity of Jurong Hill and Jurong Bird Park, the complex
interaction of the elevated AYE with its parallel road, Jalan Ahmad
Ibrahim, and a traversing industrial main road, Jurong Pier Road,
and the projected development of an MRT station on Jurong Pier
Road.
In order to attract living in the Jurong West area, new
centralities would need to be developed that transform existing
transportation nodes into vibrant urban centres integrating
living, commerce, and culture in a high-density development. The
Jurong Hill area is a prime target for such developments, creating
a new centrality that builds upon its current qualities and further
enhances the surrounding area to attract living and other mixed
uses. Land use planning and major improvements in this area
should create a network of vibrant, mixed-use, higher-density
pedestrian and transit-oriented corridors that link to major,
existing community centres such as the Jurong lake district.
The studio’s design research was funded by JTC. All
students also participated in the International Forum on
Urbanism (IFoU) Winter School, held at the Department of
Architecture, NUS, in collaboration with JTC, in January 2015, and
the projects have their genesis in this winter school. Students
interpreted the site context from the results of the winter school,
with the proposed building accentuating a future transition
towards a more mixed-use development.
Four students successfully participated in the studio. The
rst two here described have their design included in the book.
Koh Teck Wei designed an elevated tramway system as
connectivity infrastructure for the transformation of Jurong
Industrial Estate from its current focus on industry into a major
catchment area for future population growth. It is elevated above
the ground so as not to hinder, nor to be hindered, by the existing
road network. The structural system is designed as a V-shape that
can be adapted and reused in dierent forms to support dierent
versions of the elevated deck (see EleVate).
Chiang Wei Han proposed a reclamation of the ground
under the Jurong Hill Flyover to pedestrians. His design explores
how re-routing vehicular roads underground could create a
green oasis for a densely populated neighbourhood, even in
the presence of a major expressway. The space is developed as a
conduit to bring people to Jurong Hill (see The Ascent).
Hao Yi Chun designed a mixed-use building with both
residential (including SOHO (small oce/home oce)) and oce
clusters within a same organisational structure and oering full
adaptability over time.
Jasmin Mok designed a stacked industrial building that
acts as a sound barrier and billboard to shelter the Jurong Hill
from the noise generated on the expressway.
the ascent
Jurong hILL 2050
chIang WeI han
FInaL desIgn ProJect 2014-15
tutor :
ProF. rudI stouFFs
ABSTRACT
With a projected population of 6.9 million people residing in
Singapore by the year 2030, we can expect that much of the
land-use will need to be reconsidered. In particular, the intense
industrial zoning of Jurong Industrial Estate will likely need
to make way for the large inux of residents, but still retain its
identity as the industrial heart of Singapore. This project explores
what a possible scenario in the future urban landscape of
Singapore could be, based on estimations and postulations made
according to statistics of the current situation in this country.
2726
3
Site plan First oor plan
2928
The project aims to develop a prototype solution for the site of Jurong Hill Flyover along
the Ayer Rajah Expressway. Serving as an intersection of major lateral and longitudinal
axes, the site is presently utilised by vehicular trac. This commits a large surface area
on the ground to hard, asphalt roads while turning away pedestrians. Proposing a
reclamation of the ground for pedestrians, the project explores how re-routing vehicular
roads underground could create a green oasis for a densely populated neighbourhood.
3
Current site plan based on 2014 Urban Redevelopment
Authority (URA) Singapore masterplan
Proposed site plan for transformation of the area by 2050
AYER RAJAH EXPRESSWAY
JALAN BOON LAY
JURONG PIER ROAD
INDUSTRIAL ZONES
TRANSPORT
FACILITY RECREATION
JURONG HILL
JURONG BIRD PARK
OBSERVATION TOWER
JALAN BOON LAY
JURONG PIER ROAD
RESIDENTIAL ZONES
JURONG HILL
JURONG BIRD PARK
OBSERVATION TOWER
Residential zones
Jurong hill green
AYER RAJAH EXPRESSWAY
JALAN BOON LAY
JURONG PIER ROAD
INDUSTRIAL ZONES
TRANSPORT
FACILITY RECREATION
JURONG HILL
JURONG BIRD PARK
OBSERVATION TOWER
AYER RAJAH EXPRESSWAY
JALAN BOON LAY
JURONG PIER ROAD
INDUSTRIAL ZONES
TRANSPORT
FACILITY RECREATION
JURONG HILL
JURONG BIRD PARK
OBSERVATION TOWER
Industrial zones
Jurong hill green
TAMAN
JURONG
TEBAN
GARDENS
JURONG
HILL
SITE
Current land-use in Jurong Industrial Estate
HILL
BUILDING
SEAT RAMP
PERFORMANCE SPACE
GENERAL PUBLIC
SPACE
CARNIVAL SPACE
Scale of elements used in
project
Proposal to reclaim the
ground plane for pedestrians
Layout of programs
In particular, the currently under-utilised Jurong Hill can be exploited as a natural
resource to enhance the quality of the living environment when the area is fully built up.
Since it is anticipated that Jurong Industrial Estate will house one million people by 2050,
the hill as a natural green space will become invaluable in the future. Specically, the
space reclaimed by re-routing the roads on the surface underground is developed as a
conduit to bring people to the hill.
3130
Section through carnival space
Long elevation
Section through public square space
3332
Public square space
3534
Carnival space for community events and festivals
37
IFou
Jurong VIsIon 2050
The International Forum on Urbanism (IFoU) organised a Winter
School in January 2015 at the School of Design and Environment,
NUS, in collaboration with JTC Corporation. The Winter School
had the title Jurong Vision 2050’ and the aim to generate a long-
term vision for the re-development of the Jurong area in the
west of Singapore, more specically, the Jurong Industrial Estate
(JIE), a 5000 ha industrial area. The brief was to develop proposals
for the transformation of JIE from an almost mono-functional,
segregated and fragmented, polluted industrial area into a major
catchment area for future population growth that integrates
clean(ed) industrial plants with green lungs, attractive housing
and vibrant urbanity for one million people.
More than 170 students and 30 design tutors, including
the authors, participated from 18 dierent universities of the IFoU
network, from all over the world. The Winter School lasted for
12 days. Students were divided into teams of 8 to 10 students,
each focusing on one of ten topics: industrialisation, connectivity,
energy, food, liveability, transit networks, city in the garden,
waterfront development, and living together. The students within
each team then worked intensively together to develop visions
and proposals. The grouping of the students was organised in
such a way so as to ensure that each group included a mix of
students from dierent institutions and regions in the world.
While this rich mix of students created great opportunities for
cross-cultural interaction and design, it also presented some
signicant challenges.
One of these challenges was the dynamic and at times
even chaotic nature of the design process. The design process
that emerged from within these groups of students was highly
non-linear, and combined sketching and drawing with digital
methods and tools. Typically, groups would meet intermittently
to discuss overall strategies and goals, and would then break
into smaller sub-clusters to work in parallel on specic topics.
Often, these sub-clusters would then start exploring completely
new ideas, which would later have to be reconciled within the
3938
larger group. In many cases, conicts emerged which then had
to be resolved through heated discussion. In general, this reects
the fact that urban planning and design is fundamentally an
unstructured or ‘wicked’ process characterised by (1) multiple
actors with diering, legitimate values and opinions; (2) high
uncertainty; (3) aspects of irreversibility; (4) no clear solutions;
(5) being fraught with contradictions; (6) being persistent and
unsolvable (Rutledge et al., 2008).
At the start of the winter school, students were provided
with a USB stick that contained extensive data on the existing
conditions of the JIE. The students were then encouraged to use
this data as a starting point for their design process, and to use
digital methods and tools to develop proposals. However, while
almost all students brought laptops, the software tools that
they tended to use were very diverse. For this kind of context,
suggesting the idea of using a single integrated computational
environment would have been too inexible. Instead, students
had to be able to use the tools they were familiar with and to
exchange data in a variety of formats. Although some students
had experience in using GIS software, most were only able to use
simpler, modelling types of tools such as AutoCAD and SketchUp.
These issues may seem to be secondary, but in reality they have a
signicant impact on the types of proposals that were developed
and on the arguments made to support those proposals.
Prior to the start of the winter school, a semi-automated
data synthesis method was developed to generate building
models based on a set of simple rules. As the main purpose of
these building models was to quantify the overall massing and
oor areas that can be achieved, these models only needed to
consist of simple massing with oor plates and did not need to be
highly detailed in their visual appearance. In order to support fast
iterative generation of large-scale urban models, a parametric
modelling method was conceived that generates building
models based on parameter elds encoded as images (see
Generative). These building models include oor plates for each
oor, which makes it straightforward to calculate oor areas, such
as the total oor area for each function.
However, having developed these workows, it quickly
became apparent during the winter school that there was not
enough time for groups to apply these types of workows.
This was mainly due to the fact that the students had limited
knowledge of both GIS and parametric modelling, and learning
these methods in such a short time-frame was not feasible.
Some of the results of the IFoU Winter School were further
developed within an M.A. Urban Design studio and a Master of
Urban Planning studio, both at the Department of Architecture,
NUS. The workow was adapted and used to further develop
a number of proposals from these studios to demonstrate the
applicability of the method in the context of a design studio.
Two such proposals are included in the book, both as designs
resulting from the studio and as cityscapes resulting from the
application of the workow (see Envisioning Ecotopia and Urban
Metabolism).
D. T. Rutledge, M. Cameron, S. Elliott, T. Fenton, B. Huser, G. McBride,
G. McDonald, M. O’Connor, D. Phyn, J. Poot, R. Price, F. Scrimgeour, B.
Small, A. Tait, H. Van Delden, M. E. Wedderburn, R. A. Woods (2008).
Choosing Regional Futures: Challenges and choices in building
integrated models to support long-term regional planning in New
Zealand, Regional Science Policy & Practice 1(1), 85-108.
eden oF Jurong
IntegratIng urban
ecoLogY WIth hIgh
densItY LIVIng
LIn XIong
FInaL desIgn ProJect 2014-15
tutor :
ProF. PatrIck Janssen
ABSTRACT
“Eden of Jurong” focuses on the design of a high-density
development for a population of 25,000 on Pulau Samulun,
targeting the issue of liveability for high-density mixed-use
design that combines housing and residential amenities with
other commercial and industrial programmes. It investigates
the use of performance-based and parametric modelling in
design, focusing on the development of a new self-sucient
and environmentally stable urban model. The main focus of the
project revolves around the loss of biodiversity within the urban
context. It explores the issues that contribute to the phenomena
and proposes a possible design implementation to reintroduce
biodiversity into the urban context, striving for a balance
between nature and man.
4342
Site plan for Pulau Samulun
1:1500
150m
The island is divided into three smaller
neighbourhood zones, with one mega block
development per zone. Each zone supports
dierent types of programmes: the north zone
focuses on biodiversity, the middle zone focuses
on commercial and industrial, and the south
zone focuses on education and sport recreation.
4544
The patio mega block is used as the main building typology for the development.
This typology has good potential for the creation of connected green roofs and also
oers higher potential for site green coverage. The roofs are covered with a deep
layer of soil supporting a diversity of ora and a rich bird habitat. At certain points,
the green roof connects with the water’s edge, thereby helping in attracting birds.
Parts of these roofs also serve as recreational parks for the residents. Each mega
block includes a set of residential towers that rise above the green roof. In order to
avoid interrupting the green connectivity, the ats in these towers are lifted above
the level of the planting.
Meadow
Roof habitants types
Waterfront
(nature)
Green corridor Waterfront
(recreational)
Refuge
Site green coverage
Layers of development
Commercial + industrial
Bridge residential
typology
Connected green roof
Residential tower
typology
4746
150m
Rear elevation
Sectional perspective across site
4948
Street view
Courtyard view
Water’s edge view
5150
Aerial view
53
the IndustrIaL
cItY
ProPosaLs For
25,000 PeoPLe
A next selection of student projects is taken from the results of
an M.Arch. Design Technology and Sustainability thesis studio
led by Patrick Janssen. The studio participated in the IFoU Winter
School, and the students then went on to develop their own
individual projects. The overall vision was the transformation of
Jurong Industrial Estate, with one of the fundamental ambitions
being the creation of attractive neighbourhoods that combined
residential programmes with clean-tech industries. However,
even for industries that are relatively ‘clean’, the placing of
residential and industrial programmes in close proximity raised
numerous potential conicts. Key issues were heavy goods
transport, industrial noise and pollution, and health and safety
risks. What was required was a new positive vision of an industrial
city that overcomes these inherent conicts.
The site for the studio was Pulau Samulun, a 30 ha island
separated from the mainland by a narrow 50 metre waterway.
Currently, the island has a range of low quality industrial building
and sheds on it. As part of the larger planning proposal, the
island was to be completely redeveloped as a high density mix of
residential and industrial programmes.
The studio started with an investigation into the 1960’s
Metabolist movement. In particular, two competing ides were
investigated: megastructures versus group form (Lin 2010).
Megastructures were seen as gigantic long-lasting frameworks
that provided an infrastructure within which the diverse
functions of a city could grow and evolve. Fumihiko Maki (1964)
described such megastructure as being like a “great hill on which
Italian towns were built”.
As one of the earliest examples, the studio studied the
1959 Boston Harbour project, developed by Kenzo Tange and
four of his 5th year students at MIT. It consisted of a pair of huge
curving A-frame structures planned for a population of 25,000
people, built as an articial island in the middle of the harbour.
The project was similar in size to the site for the Singapore studio,
and as a challenge for highdensity development, the studio also
5554
adopted the target of 25,000.
Maki (1964) argued that the megastructure approach was
too rigid, and as an alternative, he proposed a more bottom-
up approach, where large-scale order would arise from the
grouping of smaller elements. He referred to this as the ‘group
form’ approach. “The ideal is a kind of master form which can
move into ever new states of equilibrium and yet maintain visual
consistency and a sense of containing order in the long run. This
suggests that the megastructure which is composed of several
independent systems that can expand and contract with the least
disturbance to the others would be more preferable to the one of
a rigid hierarchical structure.
The studio explored how these two ideas of megastructure
and group form could be combined. In particular, strategies
of vertical stacking were explore. This allowed both the
required high densities to be achieved while at the same time
circumventing some of the inherent conicts in the bringing
together of residential and industrial programmes. In general,
the industrial programmes were place on the ground with access
to the water for the transport of heavy goods. The residential
programmes were elevated above, and integrated with nature
and greenery. Three students took part in the studio. The rst two
projects are included in the book.
Lin Xiong divided the island into three smaller
neighbourhood zones, with one mega block development per
zone. Each zone supported dierent types of programmes: the
north zone focuses on biodiversity, the middle zone focuses
on commercial and industrial, and the south zone focuses on
education and sport recreation. Each mega block included a set
of residential towers that rose above the green roof. In order
to avoid interrupting the green connectivity, the ats in these
towers were lifted above the level of the planting (see Eden of
Jurong).
Elvira Tan proposed the creation of an articial landscape
consisting of a set of undulating hills. The landscape was
constructed as a concrete space-frame truss, incorporating all
services and supporting a layer of green planting. Industrial
spaces were placed underneath the hills while residential
buildings were constructed on top. The topography of the hill
was low in the middle and high at the edges, thereby giving
the industrial spaces direct access to the water. The residential
buildings on top of the hills were designed as neighbourhoods of
low-rise but tightly-packed blocks. An articial beach was created
along the southern edge of the island (see Small Hill Town).
Zhaoying Chew also proposed a hill with industrial
programmes below and residential programmes above. However,
in his case, the topography of the hill was high in the middle
and low at the waterfront, thereby allowing the residential
communities to have a more direct interaction with the water’s
edge. In order to allow heavy good to be delivered via the water,
a harbour was created along the southern edge linked to the
industrial space via a goods road that created a canyon through
the hill.
Z. Lin (2010). Kenzo Tange and the Metabolist Movement: Urban
Utopias of Modern Japan. London; New York: Routledge.
F. Maki (1964). Investigations in Collective Form. School of
Architecture, Washington University.
sMaLL hILL toWn
LoW–rIse housIng
For the Future
oF sIngaPore
eLVIra tan
FInaL desIgn ProJect 2014-15
tutor :
ProF. PatrIck Janssen
ABSTRACT
“Small Hill Town” envisions Pulau Samulun in 2050 as a low-
rise, high density town for 25,000 people, integrating industrial
and residential functions. One of the key requirements of the
Jurong Industrial Estate master plan is to maintain a high level
of clean-tech industry. This project explores how residential
neighbourhoods can be placed on top of large factories and
warehouse spaces for industry.
5958
The main concept of this project is the creation of an articial
landscape consisting of a set of undulating hills. The landscape
is constructed as a concrete space-frame truss, incorporating
all services and supporting a layer of green planting. Industrial
spaces are placed underneath the hills while residential buildings
are constructed on top. The industrial spaces have access to
a road along the perimeter of the island, while the residential
buildings have access to two main streets that cross the island.
Since the perimeter road is at the waterfront, the industrial
spaces have the advantage of using water-based transport
systems to transport heavy goods.
The residential buildings on top of the hills are designed as
neighbourhoods of low-rise but tightly-packed blocks. For the
design of these buildings, one of the main drivers was the desire
to achieve the perception of a low-rise human-scale town from
the two main streets that cross the island. The heights of the
buildings were therefore limited to be between 3 to 6 oors
high. Walking along the main streets, pedestrians will see no
tall buildings protruding into the view of the sky. The residential
buildings are grouped into small neighbourhoods with an
open space at the centre. The ground slabs of these spaces are
constructed using translucent building blocks, thereby allowing
daylight to penetrate deep into the industrial spaces below.
Master plan - physical model
6160
Residential
Articial hill
Industry
Layers of the island
The introduction of an articial hill generates space for industry and residence.
Transport system
Land zoning
Tram line
6362
Parametric arrangement of housing blocks drawn to attractors along main and cross streets
Examples of housing plans
1:200
Middle oor 1Ground oor Top oorMiddle oor 2
6564
Industry facing the waterfront
6766
Main north-south commercial street
Residential plaza
View of central plaza
Articial beach along southern shore
6968
Aerial view
71
generatIVe
WorkFLoWs For
sYnthesIsIng data
Prior to the start of the IFoU Winter School, a semi-automated
data synthesis method was developed to generate building
models based on a set of simple rules. As the main purpose of
these building models was to quantify the overall massing and
oor areas that can be achieved, these models only needed to
consist of simple massing with oor plates and did not need to be
highly detailed in their visual appearance. In order to support fast
iterative generation of large-scale urban models, a parametric
modelling method was conceived that generated building
models based on parameter elds encoded as images.
This method consisted of three main stages (Stous
and Janssen, 2016). First, the proposed urban typologies were
encoded as parametric models with a small number of general
parameters, such as maximum building height, site coverage,
plot ratio, and ratio of functions (e.g., residential, commercial, and
industrial). Second, for each parameter, a grey scale image was
overlaid over the site and used to create a parameter eld. These
images could either be created by hand in an image-editing tool,
or in a GIS system based on certain proximity rules. Third, large
parcels on the site were subdivided into smaller plots according
to various subdivision rules, and the parametric models were
then used to generate varying building models for each plot,
reducing the need to manually detail the road network and plot
distribution. These building models included oor plates for each
oor, which made it straightforward to calculate oor areas, such
as the total oor area for each function.
SideFX Houdini was used to implement the data synthesis
method. The parametric models that serve to generate the
building models were actually encoded as procedural rules
taking as input the respective values from the dierent parameter
elds. Additionally, a number of workows were developed to
support both generating and evaluating urban models. Exporting
the Houdini model to Unity Technologies’ game engine allowed
for advanced visualisation. In order to be able to import the data
back into a GIS system the 3D model of each plot was converted
7372
back to the base polygon for that plot, with attribute data
attached.
After the winter school was completed, the workow was
adapted and used to further develop a number of proposals from
an M.A. Urban Design studio and a Master of Urban Planning
studio to demonstrate the applicability of the method in the
context of a design studio (see Envisioning Ecotopia and Urban
Metabolism). Specically, rather than relying on parameter eld
images, the rules were adapted to consider proximity to roads of
dierent categories, MRT lines, parks, waterfront and, possibly,
other boundaries, all elements that would have been previously
identied in the 2D model. Additionally, the workows were
further relaxed to allow the import of the 2D model as a drawing
(e.g., from AutoCAD).
More recently, a parametric GIS workow was developed
combining QGIS and Möbius, a newly developed parametric
modeller that provides a novel approach for visual programming,
uses a rich topological data structure that supports objects
consisting of multiple geometric entities, and is entirely web-
based (Janssen et al., 2016). In the proposed parametric GIS
workow, QGIS is linked to Möbius in order to support fast
iterative generation and evaluation of large-scale urban models.
The workow uses parameter elds and relies heavily on the
ability to attach attribute data to geometric entities in the
model in both QGIS and Möbius. The workow consists of ve
main stages that alternate between QGIS and Möbius, with data
exchange using the GeoJSON le format.
In stage 1, QGIS is used to create a map of the area,
including existing buildings and infrastructure. Geographic data
is collected and manually integrated into a single geospatial
dataset. A series of large parcels are dened in QGIS, indicating
the areas of the site where future development is proposed. The
parcels are exported from QGIS and imported into Möbius.
In stage 2, Möbius is used to recursively subdivide each
parcel into similar size plots, with tertiary roads also inserted
between the plots. The subdivision is performed by a parametric
procedure that attempts to create plots that are as evenly sized as
possible. The parameters allow the designer to specify the target
size of the plot. The plots and tertiary roads are exported as a
GeoJSON le and imported back into QGIS.
In stage 3, QGIS is used to create parameter elds through
a combination of proximity functions and custom formulas. A
series of additional attributes are dened in the attribute table for
the plots for this purpose. For example, rst, proximity attributes
can be created that calculate proximity to roads, parks, and the
waterfront. Second, parameter attributes can be created that
calculate the building height and plot coverage based on the
proximity attributes. This results in each plot being assigned
multiple building heights and plot coverages. Third, a nal pair of
attributes can be created that calculates the nal value for both
parameters using formulas that give priority to certain rules or
conditions. The plots with the attributes are exported from QGIS
and imported into Möbius.
In stage 4, Möbius is used to generate urban models
using a library of parametric urban typologies. The parameters
for each plot are extracted from the attributes attached to the
plot polygon, e.g., the building height and the plot coverage. The
urban model is then generated by selecting and instantiating
a parametric urban typology on each plot. All polygons in the
model will have a ‘type’ attribute assigned that denes the
types of building elements that they represent, such as ‘plot’,
‘footprint’, ‘oor’, ‘wall’, ‘roof’, and ‘window’. Floors can be further
categorised as ‘residential’, ‘commercial’, and ‘industrial’. Möbius
can then be used to calculate the oor areas for each of the
dierent functions, and these values can be added as attributes
to the footprint polygons. Finally, in order to be able to export the
model back to QGIS, the model is attened back to 2D. The critical
information that is transferred is the building footprints together
with the oor area attributes. These attributes are important as
they will be used for the QGIS analysis in the nal stage. Möbius
exports the building footprints as a 2D model which can then be
imported back into QGIS.
In stage 5, QGIS is used to analyse the attened map of
the urban model. Calculating the number of people in each
building for each of the functions using a formula that divides
the oor area for each function by average ‘area-per-person’
values, population densities can also be determined. Using a
simple buer analysis, the percentage of people within a certain
walking distance of transport nodes can also be determined (see
Evaluative).
P. Janssen, R. Stous, A. Mohanty, E. Tan, R. Li (2016). Parametric
modelling with GIS, Complexity & Simplicity: Proceedings of the 34th
International Conference on Education and research in Computer
Aided Architectural Design in Europe, 22-26 August 2016, Oulu,
Finland.
Stous, R., P. Janssen (2016). A rule-based generative analysis
approach for urban planning. In: J.-H. Lee (ed.), Morphological
Analysis of Cultural DNA, Springer.
enVIsIonIng
ecotoPIa
a carbon neutraL
cItY aMIdst MId-rIse
LIVeabILItY
urban PLannIng studIo 2014-15 :
andrea MeInartI rachMat,
teY huI PIng serene,
deLon Leonard,
Wu XIng PIng,
Loh sZe sIan
tutor :
ProF. oscar carracedo
research teaM :
ProF. PatrIck Janssen,
ProF. rudI stouFFs,
LIn XIong,
andrea MeInartI rachMat
ABSTRACT
ECOTOPIA is a vision of a new urban metabolism model. Bearing
in mind future needs of resources according to a projected
2050 population of 7.4 million, ECOTOPIA strives to become the
future sustainability model of self-suciency by considering ve
metabolism elements : water, energy, food, waste, and greenery.
7776
Residential 100%
Residential 60% : commercial 40%
Residential 60% : industrial 40%
Residential 60% : facilities 40%
300
900 3000m
1500
Land Use Master Plan
1:30000
N
Industrial 100 %
Industrial 60% : commercial 40%
Industrial 60% : residential 40%
Industrial 60% : facilities 40%
Commercial 100%
Commercial 60% : industrial 40%
Commercial 60% : residential 40%
Commercial 60% : facilities 40%
Special use
Land use Master plan
1:30000
Eco-strips boundary
Category 1 green: parks
Category 2 green: district parks
Category 3 green: community parks
Green buer at Jurong island
7978
ECOTOPIA has four goals:
- a new industrial model with an emphasis on R&D
- a hybrid urban metabolic system
- a mid-rise and sustainable city with increased energy eciency
- a unique and enhanced liveability
Using the concept of Urban Metabolism as the basis for
planning, Jurong Industrial Estate is divided into ECO-grid cells.
Horizontally, the site is divided into three dierent mixed-use
belts : residential, industrial, and commercial. Three dierent
mixed-use cells make up a self-sustainable ECO-strip. Using
projected numbers of future demands on resources, a ‘per person
area requirement’ is calculated, which is taken into consideration
in calculating the maximum possible population.
ECOTOPIA also proposes a new model of industrial clustering
according to the metabolism elements of waste, energy, food,
and water. To ensure self-suciency, each dierent industry is
represented in each cell to then join together in a closed loop
within each ECO-strip. The new industries in Jurong Industrial
Estate will be steered towards high-end manufacturing, clean
energy, and R&D. The metabolism clusters are allocated based
on the existing context such as the presence of Pandan Reservoir
and clean-tech Park near NTU.
Waste
Food
Energy
Water
Reservoir
New industrial metabolism model Waste ow (centralised biomass energy plant)
Energy ow (centralised clean energy plants) Food ow (new age industrial farming)
8180
Parameters for roads and MRT in residential belt Parameters for roads and MRT in industrial belt
Proposed road network
Category A
Category B
Category C
Category D
East west line
Cross island line
Jurong regional line
PRT networks (on ground)
Proposed public transportation network
Tracks above ground
Tracks underground
Water ow (new reservoir supply)
Parameters for roads and MRT in commercial belt Parameters for parks category 2 and 3
Heritage railway corridor
Leisure parks
Universities
To Nusajaya
Western Water
Catchment
Chinese
Garden
Existing and proposed landmarks
Clean energy research
Convention centre
Incineration plant
8382
Building height analysis overlaid on aerial view
06 Meters
45 Meters
8584
Street view
Harbour view
8786
Aerial view
89
eVaLuatIVe
data anaLYsIs
WorkFLoWs
One key challenge all groups in the IFoU Winter School faced,
was the diculty in backing-up claims about their proposals with
quantitative data. The overall target for the winter school was
to create housing and other amenities for one million people,
while still maintaining a signicant portion of industrial clean-
tech industry. The target resulted in some quite well dened
requirements of oor areas for dierent functions. However, most
of the urban models were purely visual and could not be used
to quantify the total oor area that would result. With all groups,
it was dicult to understand whether the proposed typologies
and urban massing were appropriate for one million people or
not. In some cases, the oor areas may have fallen far short. This
type of quantication is important since proposals that work well
at lower densities may be fundamentally awed at these higher
densities.
These types of issues can be tackled using semi-
automated data synthesis methods that generate building
models based on a set of rules, as described previously. These
building models only need to consist of simple massings with
oor plates and do not need to be highly detailed in their visual
appearance, as their main purpose is to quantify the overall
oor areas that can be achieved. Having identied oor areas,
the number of people in each building for dierent functions
such as residential, commercial, and industrial, can be calculated
using a formula that divides the oor area for each function
by average ‘area-per-person values. Subsequently, population
densities can be determined. In addition, GIS systems such as
ESRI ArcGIS and QGIS both have tools that could be used to
quantify improvements in accessibility, such as proximity analysis
and network analysis. For example, this would allow for the
calculation of indicators such as the percentage of people who
live within ve minutes of a transport node. This would then
allow for direct comparison of alternative options.
Data mining as a data analysis technique was not explored
for use in the IFoU Winter School, though it was recognised
9190
to have the potential to further support the students’ design
processes. In particular, it was recognised that data mining could
play a role in integrating performance simulation techniques
and tools into the urban planning and design process. Two
of the groups in the winter school specically focused their
investigations on sustainable energy production, consumption,
and distribution within the built environment, elaborating the
topic on two levels. While new approaches of energy exchange,
circulation, and balancing were investigated at the urban level, at
a precinct and block level design explorations aimed at reducing
energy demand of buildings, improving natural ventilation and
daylight access, and integrating photovoltaic, solar thermal
energy, and other technologies into the building design. While
the latter explorations can be supported with simulations,
applying these simulations at the urban level or extrapolating
the ndings from a block level to the urban level is not always
possible.
Exactly this issue of extrapolating ndings was
investigated after the winter school, based on the data
made available to the students. Specically, the object of
investigation was the prediction of energy consumption of
public residential (HDB) buildings at the urban scale. Performing
building simulation requires extensive knowledge on the
physical behaviour of the building and the local environment.
Furthermore, simulations have to be repeated every time the
building information is changed, such as its number of stories,
footprint, orientation, etc. At the same time, simulation results
may dier from actually measured results as a consequence of
various aspects, including lifestyle of the occupants. On the other
hand, data mining the actual electricity consumption of all HDB
buildings can give insight into the specic relationship between,
e.g., building site coverage and electricity consumption (Liu et al.,
2015).
Preliminary research involved training a neural network to
relate actual energy consumption to a number of building and
environmental variables, including the green plot ratio (dened
as the total leaf area divided by the site area), the oor area ratio
(dened as gross oor area divided by the site area), a sky view
factor, a compacity’ factor reecting on the compactness of the
building, the total number of stories of the building, the building
typology (shape of the footprint of the building), and the gross
oor area. Additional climatic parameters were also considered.
The research showed that the most important variable predicting
energy consumption for HDB buildings is the number of stories,
followed by the gross oor area and the building compacity.
The green plot ratio, the oor area ratio, and the sky view factor,
on the other hand, have little impact on the actual electricity
consumption. Certainly, the research doesn’t fully explain why
this is the case, but it does show that these relationships hold
with considerable accuracy. Furthermore, including a baseline
simulation greatly improved the eciency of the neural network
learning process, without relinquishing accuracy. Though
this reintroduces a simulation into the process, the baseline
simulation is made for an existing HDB building for which data
on actual electricity consumption is available that has no other
relation to the planning site.
While it may not be possible as of yet for urban planners
and designers to adopt such data mining methodology, the
research does show that it would be possible to assess electricity
consumption for public residential housing at the level of detail
of the generated building data described above in real-time. The
development of an application that makes this methodology
accessible to urban planners and designers is envisioned
Y. Liu, Y. Huang, R. Stous (2008). Using a data-driven approach
to support the design of energy-ecient buildings, Journal of
Information Technology in Construction 20, 80-96.
urban
MetaboLIsM
re-thInkIng urban FLoWs
urban desIgn studIo 2 2014-15 :
tuLIka agraWaL,
raVIsh kuMar,
YutIng LIu
tutors :
ProF. Jurgen roseMann,
ProF. LoW boon LIang
research teaM :
ProF. PatrIck Janssen,
ProF. rudI stouFFs,
LIn XIong,
raVIsh kuMar
ABSTRACT
Urban Metabolism is an analogy describing the city as a living
organism that requires resources to nourish its activities. The ow
of goods, people, biota, energy, fresh water, and fresh air is critical
in maintaining the Urban Metabolism. For Jurong Vision 2050,
three important aspects were considered: Flow of Goods, Flow of
People, and Flow of Water & Green.
9594
015
30 120
60 240m
015
30 120m
60
N
Jurong basin development
1:3000
015
30 120
60 240m
015
30 120m
60
N
Residential mix | 38 % |108 hectare
Adaptive reuse
Amenities
Basin edge landscaping
Commercial mix | 21% | 65 hectare
Industrial mix | 41 % | 103 hectare
Total site area : 87 hectare
Fsi : 3.2 (Average)
Dwelling unit density : 140
Programmatic diversity
9796
Viewing deck
Underground conveyor belt for goods movement
Section through conveyor belt system
Underground conveyor
belt for goods
movement
Industries and mid rise housing
Industrial basement connected to underground
conveyor belt system
Community greens
Facilities and amenities
Industries and mid rise housing
Industrial basement connected to
underground conveyor belt system
Community greens
Facilities and amenities
Basin
edge
Steps
Informal sitting
Deck
Landscaped plaza
Open air restaurants
Pioneer road
Industrial spine
Shopping street
Flagship stores
Oces
Soho | moho
Section through MRT and immediate centrality
One of the main interventions of the proposal is to create a spine along Pioneer Road.
The spine consists of an underground goods moving system, a road transport system,
and an elevated MRT system. The second intervention is the priority placement of
residential (mixed-use) land-use along the green and blue (water) infrastructure to
provide a quality environment for residents. These residential areas are connected by
PRT lines for the last kilometre coverage. The third intervention is to allocate industrial
mixed-use along Pioneer Road. The central distribution system is located at the central
spine, and connects via an underground conveyor belt system to the surrounding
industries as a secondary mode of transportation of the goods. Major intersections or
nodes on the central spine act as centralities with commercial mixed-use and residential
mixed-use volumes on the roof of industrial buildings.
3
Industrial + commercial + residential
Underground conveyor belt
connected to basement of
surrounding industries
Ground level | rst interaction level
Clean tech industries and retail and
commercial programmes connected to
overhead MRT station
Community green terraces
Creating intermediate centrality by
placing commercial programmes
connected to MRT and residential
at top
Placing mid-rise housing on the rest of
the deck with community green areas
Second interaction level
as a connected deck above industries
9998
Creating an industrial spine
Industrial mix land-use along the spine
Industrial mix-use
Residential mix-use
Basin
Adaptive reuse
Commercial mix-use
Residential mix land-use facing blue and green areas
Existing and proposed extensions in 2030 for MRT lines : E-W line & Jurong regional line
Proposed extension of MRT lines for 2050
Last kilometre through PRT lines : 100% of the area covered by public transport in 2050
10110 0
Aerial view
10310 2
Bird’s eye view Street view (industrial + commercial + residential)
10510 4
Aerial view
107
concLusIon
This book rethinks urban practices, both its processes and its
products, within the context of JTC Corporation’s 2050 vision
for the Jurong Industrial Estate (JIE). Selected student projects
illustrated in this book are taken from dierent planning and
design studios, and operate at varying scales, but they all propose
innovative ideas on how JIE can be re-imagined for a future
Singapore. In addition, all the planning and design studios
featured in this book, including the IFoU Winter School, were
supported nancially by the NUS-JTC I3 (Industrial Infrastructure
Innovation) Centre.
The selected projects serve as demonstration for urban
planning and design workows that leverage new computational
tools and techniques in order to enable planning and design
options to be systematically developed, evaluated, and analysed.
The workows address complex challenges that require new
thinking about our city and its planning and design. They support
the exploration and development of alternative visions of the
future city, and enable the evaluation and analysis of dierent
urban planning and design options.
The urban planning and design workows demonstrated
in this book are not conclusive. They are the intermediate results
of a continuing investigation into rethinking urban practices. The
results in this book present a stepping-stone in this investigation.
109
credIts
EleVate: Future transport systems for the city
Design: Koh Teck Wei
Tutor: Rudi Stous
Final Design Project, Master of Architecture, 2014-15
The Ascent: Jurong Hill 2050
Design: Chiang Wei Han
Tutor: Rudi Stous
Final Design Project, Master of Architecture, 2014-15
Eden of Jurong: Integrating urban ecology with high density living
Design: Lin Xiong
Tutor: Patrick Janssen
Final Design Project, Master of Architecture, 2014-15
Small Hill Town: Low–rise housing for the future of Singapore
Design: Elvira Tan
Tutor: Patrick Janssen
Final Design Project, Master of Architecture, 2014-15
Envisioning Ecotopia: A carbon neutral city amidst mid-rise
liveability
Design: Andrea Meinarti Rachmat, Tey Hui Ping Serene, Delon
Leonard, Wu Xin Peng, Loh Sze Sian
Tutor: Oscar Carracedo
Urban Planning Studio, Master of Urban Planning, 2014-15
Urban Metabolism: Re-thinking on urban ows
Design: Tulika Agrawal, Ravish Kumar, Yuting Liu
Tutor: Jürgen Rosemann, Low Boon Liang
Urban Design Studio 2, Master of Arts (Urban Design), 2014-15
Research
Chen Kian Wee, Chiang Wei Han, Hu Jiajun Kason, Ravish Kumar,
Delon Leonard, Lin Xiong, Ibrahim Nazim, Yazid Ninsalam, Ong
Shi Hui Shirlynn, Andrea Meinarti Rachmat, Elvira Tan, Wu Xin
Peng, Dr. Zhang Ji, Zhang Wei
book desIgn :
anshuMan roY,
eLVIra tan,
tuLIka agraWaL
coVer Page desIgn :
eLVIra tan
sPonsor :
nus-Jtc I3 centre
110
ResearchGate has not been able to resolve any references for this publication.