Data-driven urban planning & design

Data-driven urban planning & design

ULI Leadership Convivium

Singapore12th November 2019

Tim StonorArchitect | Urban Planner

@Tim_StonorData-driven urban planning & design Space Syntax © 2019




6.5 billion

9 billion

2010 2050World

Population

3 billion

7 billion

2010 2050Urban

Population

1 billion

3 billion

2010 2050

1.5 billion

2010 2050Middle Class

4 billion

Informal Urban Population

Source: UN-Habitat, Goldmann SachsA growing trend towards cities


Suppressed movement economy. Enhanced movement economy.

Main street, mixing global & local movement.

Fast highways, separating global & local movement.

Urban challenge #1 Fast highways have replaced main streets


The fragmented "city" of disconnected developments.

The integrated city of continuously connected neighbourhoods.


1. Presence

2. Copresence

3. Communication

4. Interaction

5. Transaction

6. Introduction

7. Innovation

8. Broadcast

Transaction machine

Spatial Accessibility A fundamental property of street networks

Key discovery #1 Spatial Accessibility predicts movement

0

1

2

3

4

5

6

7

8

0 1 2 3 4 5 6 7

y = .784x + 2.147, R-squared: .589

Spatial accessibility

Pede

stria

n m

ovem

ent

Spatial accessibility-10123456789

0 1 2 3 4 5 6 7

y = 1.235x + 1.048, R-squared: .702

Vehi

cle

mov

emen

t

Spatial masterplanning Designing the geometry of the grid


Low High


Low High



Low High



Low High


Courtesy of The TLRN Central London Pedestrian Study by Atkins

80% retail located on 20% most spatially accessible streets

Key discovery #2 Spatial Accessibility influences land use viability

Nur-Sultan Masterplan, Kazakhstan

Key discovery #3 Spatial Accessibility influences crime & safety


Low High

Connected layout Disconnected layout High accessibility scoreLow accessibility score

Disconnected layout Connected layout Difference

IRR 15% 18% 3%

NPV RMB 6.65 Billion RMB 10.0 Billion RMB3.35 Billion

Key discovery #4 Spatial Accessibility influences land value

Travel to workDrive car or van

Local scale integrationActive centres

Oxford

Key discovery #5 Spatial layout influences congestion

HealthVery good health

Local scale integrationActive centres

Oxford

Key discovery #6 Spatial layout influences health

Spatial layout, car dependence & social isolation

Integrated Urban Modelling

Urban Data Platform

Drivers | ImpactsOpportunities | Technologies

Applications

Macro Meso Micro

Existing 长春 Proposed 规划深入

Spatial accessibilityhigh

low

Rapid urban growth

Darwin CBD Masterplan, Australia

2,000

3,000

750

2,500

1,5001,000

500250

2,000

750

2,500

1,5001,000

500250

0

totototototototo

Unimproved Capital Value$ per m²

above3,000 to

Existing land value

Land Value Model

predicts> $AUD 3.7 billiontotal land value

Space Syntax © 2019@Tim_StonorData-driven urban planning & design

NAC

h10k

&1.2

kLand Use Attraction Retail and Catering MultiScale (10k & 1.2k)

Map

box

MultiScale (global and local)

Primary Routes

Global (vehicle oriented)

Local (pedestrian oriented)

Super MultiScale (v.global and v.local)

RetailCatering

Ground floor land use


Existing

First floor

Visibility

Low High

Proposed

Enhancing asset performance Retail circulation


Case Study St David’s Centre

Cardiff, UK

Space Syntax ServicesVisitor footfall prediction

Project statusConstruction complete:129,562 sq.m

40 million visitors per annum

Project typeMajor extension of existing, city centre retail development

Greyfriars Road


Existing visitor flow

Cardiff Central Rail Station

Cardiff Queen St Rail Station

Cardiff Millennium Stadium

Cardiff Market

2,187

705

719

840

491648 156

472

242

249

230

524

347

122

215

260

2,680

2,269

1,434

1,099

1,996

721

1,784 2,844

2,009

2,869

3,392

1,379

519

7153,378

2,470

823

380

1,724717

164

402

792

877

1,013

400

642

1,012

559

875

812

279140

312

567362

153129

232

1,741

500

731

422

538

540

539313 970

1,934

250015001000500250

0

+250015001000500250

------

Pedestrian movement (people per hour)

*

28 million people per year

Greyfriars Road





Cardiff Market

0 100 200

metres

Footfall forecast model Step 1

Greyfriars Road

0 100 200

metres


Forecast pedestrian movementAll day average (people per hour)

3,500 - 4,0003,000 - 3,500

2,500 - 3,000

2,000 - 2,500

1,500 - 2,000

1,000 - 1,500

0 – 1,000

\

\

\




Footfall forecast model Step 2

38 million people per year


Case Study Trinity Centre

Leeds, UK

Space Syntax ServicesCirculation Design Review

Circulation Design Scenario Modelling (Strategic modelling)

Project statusConstruction complete92,903 sq.m

23 million visitors per annum

Project typeNew, major city-centre retail development

0 25 50

metres

Albi

on S

tree

t

Mill

Hill

Brig

gate

Mill Hill Chapel


Route Hierarchy Key

High Choice

Low Choice

Footfall modelling

Space Syntax IUM Transport

Mobile operator Origin-Destination matrix and mode share at LSOA level

Physical activity

Third party apps eg Strava

SweatCoinaverage steps per day per user per LSOA

Health & wellbeing

Health Dataset NHS open data eg Dr:patientratios, patient satisfaction scores

Socio-economic

Local employment and workforce datasets

OSM Census

Qualitative

Community builders area scores

Linked data visualisation platformEmu Analytics’ Location Insights Explorer

Digital Urban Platform

Data-driven urban planning & design

Documents