URBAN PLANNING AND WELL-BEING. THE CASE OF

BARCELONAMontserrat Pallares-Barbera, Jordi Duch, Anna Badia

ABCD Series, January 18, 2012

Room S050 of the CGIS South Building at 1730 Cambridge St at 12 noon

Harvard University Institute for Quantitative Social Sciences Center for Geographic Analysis

Geography Dept. Universitat Autònoma de Barcelona

Center for Geographic Analysis. IQSS. Harvard University

2

structure of the talk

• Problem and motivations of the problem• Objectives and structural question• Case study• Methodology• Results• Concluding unanswered questions

Wellbeing and urban planning

• Increasing urban services• Past and present• Cities• Agglomeration• Shocks• Background

Density, malnutrition, agglomeration, overworking, infected water, illiteracy Causes

Mortality, ignorance, illness, unhappiness, uncultured, unsociability Consequences

Cities experienced a high incoming population during the industrial revolution

Population density increased in all of them

Shocks to the system:

Mortality increased and life span decreased

Determinants of the cities in the nineteenth to beginning twentieth centuries

5

OBJECTIVES AND Structural question. How urban planning and, in particular, the location of urban services, affects wellbeing

. Data from the Plan of Expansion of Barcelona of 1860, proposed by IldefonsCerdà is used for various analysis and simulations

Why-We want to answer the question of ‘who gets welfare and where do they get it’-We want to get further elements to incorporate in future planning practices

How-To study Cerdà’s proposal using current methodology from location theory, such as location-allocation models and ArcGis 10 Network Analyst for analysis

Working hypotheses- Cerdà’s proposal for the expansion of Barcelona had the aim to improve the

population living standards- He used urbanism as a redistribution tool-He included services to population as a necessary condition to get his objectives

of improving social well being

Sources: Barcelona Institut d'Estadística de Catalunya (Idescat) and Centre d'Estudis Demogràfics; Boston http://www.bpl.org/research/govdocs/boststats.htm; Chicago http://tigger.uic.edu/depts/ahaa/imagebase/chimaps/mcclendon.html; London http://www.demographia.com/dm-lon31.htm; New York http://www.demographia.com/dm-nyc.htm; Paris ; Philadelphia http://physics.bu.edu/~redner/projects/population/cities/philadelphia.html.

Population and population density in cities, per Sq Km

* 1071

* 32465

* 26978

* 6,515 * 897

*23,552-1,378

*4,923

*13,365*1,105

85,600

* 15,926

*4,266

*11,535

*30,529

*20,240

http://www.nyc.gov/html/dcp/pdf/census/1910_pop_density.pdf

Creator: Tenement InspectorSource: Chicago Historical Society (ICHi-37341)

http://encyclopedia.chicagohistory.org/pages/10727.html

Sour

ce: N

ewbe

rry L

ibra

ryht

tp://

ency

clop

edia

.chi

cago

hist

ory.

org/

page

s/36

08.h

tml

Cre

ator

: U.S

. Pub

lic H

ealth

Ser

vice

Sour

ce: N

ewbe

rry L

ibra

ry

http

://en

cycl

oped

ia.c

hica

gohi

stor

y.or

g/pa

ges/

3892

.htm

l

Barcelona. Mortality in the first floor level 1846-1865

Source. Canedo Arnedo, M. Geohistòria ambiental de la Barcelona del segle XIX. Master Research Project. Universitat Autònoma de Barcelona. Geography Department, 2010.(1) García Fària, 1894, p. 26-27.

Average life expectancy between richer and poorer classes was 38.83 and 19.68 years of age, respectively (men, average between years 1837 and 1847; Cerdà, 1867)

Words of Cerdà’s• Cerdà attributed the causes of mortality and poor urban

condition to greed and ignorance: “The errors in the lack of hygienic means in founding a city are a consequence of ignorance and greed and prevent the development of robust, wise and industrious generations. These errors increase mortality, decrease the average life expectancy, and mostly contribute to epidemic attacks every twenty years” (Cerdà, 1860: 55).

• Under these beliefs, his plan had to be built using new standards of the (new) culture of the (new) century: “[t]oattain the good conditions that the culture of our time demands for the entire population” (Cerdà, 1860: 56).

Creator: I. T. Palmatary and Christian IngerSource: Chicago Historical Society (ICHi-05656)

http://encyclopedia.chicagohistory.org/pages/10603.html

NYC-GRID-1811

http

://en

.wik

iped

ia.o

rg/w

iki/F

ile:N

YC

-GR

ID-1

811.

png

Com

mis

sion

er's

Grid

pla

n fo

r Man

hatta

n

http

://en

.wik

iped

ia.o

rg/w

iki/C

omm

issi

oner

s%27

_Pla

n_of

_181

1

http://www.oldlondonmaps.com/stanfordpages/cityMAIN.html

Edward StanfordThe School-Board Map of London, c. 1872Scale 6 inches to one mile.

Streets of the New Barcelona

Area:  1,975 Ha

228,3Perimeter

118,8Streets Outside Enlargement

117,4Streets with train 

18350 m

77,530 m

237,720 m

Longitude (km)Street type/ wide

4003-6 m

200<3 m

Number of streets

Street type/ wide

Streets of the Old Barcelona

Area:  193,97 Ha

2512Government Institutions

3333Schools

Outside the city3Hospitals

1010Markets

388Parks

Number of blocksOccupancy

NumberType of services

Planning equipments

- 10 midwifes and 69 surgeon doctors (Cerdà, 1867)

Service provision in the Old Barcelona

- 3 markets, 2 of them fisheries (Pescadería del Mercado del Borne, 425 m2, and Pescadería del Mercado de Isabel II, 900 m2), and 1 of them of general groceries (Mercado de la Plaza de Isabel II, 3,525 m2;

Service location made by Cerdà

Table 2. Housing category and number of floors projected, area and inhabitantsHousing category Floor number Number of people Area sq meters sq meters/Inhab Lenghtxdepth

living in each floor

Housing for the wealthyProjects A-F PB (Ground) 0 225-400 20x20

1rs & 2nd floor 9 to 13 225-400Mean 16 787 48.92

Housing for the midle classProject G PB (Ground) 2 shops 400

1rst 22 4002nd 12 400Mean 34 1200 35.29 15x15

Housing for the working classProject A-H PB (Ground) 1 shop 69-103

1rs 4 to 10 69-136.52nd 4 to 5 69-136.53rd 4 to 5 69-136.5Mean 4 to 5 69 to >103 23.62 10.15x10.15

Source. Cerdà p. 75, (MAP. Aj. Barcelona 1991). [Table] "97a. Clasificacióbn de las casas, número de personas que pueden albergarse en ellas, superficie que ocuparán, precio e importe de la construcción, importe tottal y renta que producirán"

MethodologyTypes of housing projectedHousing category and population density per category Cerdà’s Barcelona

168314100,00TOTAL

27971,662Error

16551798,338TOTAL

4490126,677No contribuyentesNon-paying taxes

7320,435Pobres de solemnidadPoor

7566744,956JornalerosDay laborers

25471,513Propietarios de todas clasesOwners

122897,301IndustrialesIndustrials

15650,93FabricantesManufacturers

18951,126ComerciantesMerchants

4880,29LabradoresFarmers

44602,65PropietariosLand owners

6820,405Militares retiradosRetired army

161979,623Militares activosArmy

3840,228Empleados cesantesUnemployed workers

25681,526Empleados activosEmployed workers

11410,678Eclesiásticos de todas clasesClergies

Number% ProfesionesEmployment Categories

T a b l e 2 B a r c e l o n a P o p u l a t i o n C e n s u s 1 8 5 5 . P o p u l a t i o n b y e m p l o y e m e n t ( " C l a s i f i c a c i ó n d e l o s h a b i t a n t e s p o r p r o f e s i o n e s , o f i c i o s , o c u p a c i o n e s r e f e r i d o a 1 0 0 i n d iv i d u o s d e l t o t a l d e p o b l a c ió n ( … ) )

Sou

rce:

Cer

dà, I

. (18

59 p

. 185

). T

eorí

a de

la c

onst

rucc

ión

de la

s ci

udad

es.

Apl

icad

a al

pro

yect

o de

refo

rma

y en

sanc

he d

e B

arce

lona

. In

Teo

ría

de la

con

stru

cció

n de

las

ciud

ades

. Cer

dà

y B

arce

lona

(19

91).

Min

iste

rio p

ara

las

Adm

inis

trac

ione

s P

úblic

as. A

junt

amen

t de

Bar

celo

na.

% InhabitantsHousing categories Inhabitants/House Sq m/house Sq m/Person1 High income class Land owners 2,65 4460 1 13 800 61,542 High income class Merchants 1,13 1895 2 22 1200 54,553 High income class Manufacturers 0,93 1565 3 12 640 53,334 High income class Industrials 7,30 12289 4 21 960 45,715 High income class Owners 1,51 2547 5 11 450 40,91

6 18 675 37,5TOTAL 13,52 Mean Hous High Class 16,17 787,5 48,92

6 Middle income class Clergies 0,68 1141 7 34 1200 35,297 Middle income class Employed workers 1,53 2568 7 34 1200 35,298 Middle income class Army 9,62 16197 7 34 1200 35,299 Middle income class Retired army 0,41 682 7 34 1200 35,29

TOTAL 12,23 Mean Hous Med Class 34 1200 35,2910 Low Income class Farmers 0,29 488 8 15 309 20,611 Low Income class Unemployed workers 0,23 384 9 20 412 20,612 Low Income class Day laborers 44,96 75667 10 12 207 17,2513 Low Income class Poor 0,44 732 11 16 276 17,2514 Low Income class Non-paying taxes 26,68 44901 12 12 270 22,5

13 16 360 22,514 12 409,5 34,1315 16 546 34,13

Error 1,66 2797TOTAL 72,59 Mean Hous Low Class 14,88 348,69 23,62

TOTAL 100,00 168314

Table 5. Assigning population groups to housing types

Block type Sq m Housing Scenario_ Sq m/Categ Inhabitants1 2180,2 13,52 294,76304 6

12,23 266,63846 874,25 1618,7985 69

TOTAL 100,00 2180,2 822 5000 13,52 676 14

12,23 611,5 1774,25 3712,5 157

TOTAL 100,00 5000 1883 1608,4 13,52 217,45568 4

12,23 196,70732 674,25 1194,237 51

TOTAL 100,00 1608,4 614 4450 13,52 601,64 12

12,23 544,235 1574,25 3304,125 140

TOTAL 100,00 4450 1685 5250 13,52 709,8 15

12,23 642,075 1874,25 3898,125 165

TOTAL 100,00 5250 1986 4116 13,52 556,4832 11

12,23 503,3868 1474,25 3056,13 129

TOTAL 100,00 4116 1557 9700 13,52 1311,44 27

12,23 1186,31 3474,25 7202,25 305

TOTAL 100,00 9700 3658 9636,87 13,52 1302,904824 27

12,23 1178,589201 33

(…)

Table 6. Population per block type

Scenario_1: 13.52% of houses to High Income Class; 12.23% of housing to Middle Class; 74.25% of housing per Low Income class.

35.6012,500Intersection of buildings

50.004,021Block with alley

40.0012,500Two sides built

Blocks with two sides built (the most common type found in the Expansion)

20.0010,901Outside of Expansion

One side built

% of built occupancy

Block area m2

LocationBuilding form

Block Typology

(…)

• Location-Allocation

Minimize impedance. Minimize weighted impedance (P-Median). It chooses facilities such as the total sum of weighted impedances (demand allocated to a facility multiplied by the impedance to the facility) is minimized.

• ArcMap_10 Network Analyst

The optimization model (1)

25

The optimization model (2)General model:

Given

Choose

Where

In order to minimize Z equal to

Subject to

• Where,• ai = quantity of population in node i,• i = origin of population,• j = possible service location,• p = number of services,• dij = the shortest distance between node i and node j,• xij = 1 if population of node i is assigned to j, 0 otherwise, • xjj = 1 if a service is located in node j, 0 otherwise.

26

Public versus private goods

• The use of a service do not affect provision of this to other people

• Public good is not subject to congestion• Allocation of people to schools, for

instance, do not decrease the utility for other people.

• Choosing a service depends on individual’s distance to a service

School service areas

•Table 4. Population within each school time interval

Interval time in minutes Population % Cummulative< 5 41,297 27 275 ‐ 10 75,259 49 7610,1 ‐ 15 27,263 18 94> 15 8,656 6 100

Total 152,475 100

School location-allocation general model; Impedance cutoff 5 minutes

Park service areas, 5, 10 and 20 minutes

Population and parks

Population served by parksInterval in time minutes Population % Cumulative %< 5 42,588 28 285 ‐ 10 60,116 39 6710,1 ‐ 20 47,888 32 99>20 1,883 1 100

total 152,475 100

Park Location-allocation general model; Impedance cutoff 20 minutes

Market service areas

•Table 2: Population within each market time-interval

100.0152,475Total

100.05.08,072>24

95.046.070,69112–24

49.036.054,2686–11

13.013.019,4441–5

Cumulative %%PopulationInterval time in minutes

Market location-allocation general model; 20 minutes

Cerdà hospitals allocation of demand within 30 minutes distance

Cerdà hospital service areas 10, 20, 30, and >30 minutes

Demand uncovered by the hospitals

Table 1: Population within each hospital service area

100.0152,475Total

100.022.033,224>30

78.033.050,50020–30

45.034.052,50010–19

11.011.016,2511–9

Cumulative %%PopulationInterval time in minutes

Annoyance Negative Externalities

Future research: Solar radiation and Air flow

• Solar radiation, grid positioning, health• Air flow analysis and ventilation, health• Energy efficiency• Taking into account capacity of services

– Introducing different population densities• Introducing variability of services• Considering different levels of income

– Transferable utility– Spatial justice

- One of the Cerdà’s objectives was to positioning the grid in order to get maximum sunshine and natural ventilation for housing

-More sustainable cities

-Old technique of house ventilation and natural air recycling and cooling inside the house

- He considered the streets as “aerial channels”, which had the function for the city what lungs do for humans: “Por lo que toca a salubridad, siguiendo en estaparte a los highienistas, podemos considerar las calles como canales aereos (…) que vienen a ser para las ciudades como lo que para el cuerpo humano son los pulmones.” Cerdà, 185, p. 376 (1991)

Source: Cerdà, 1855 p. 374 (1991).

Figure 14. Cerdà’s discussion about sunlight’s good influence for human health

Source: Cerdà, 1855 p. 375 (1991).

Examples of St. Petersburg and Paris: mortality rates and quantity of sunlight

Fig 2. Summer solstice, June 21rst. Passeig de Gràcia section current geographic coordinates

THE END

THANK YOU VERY MUCH

MONTSERRAT PALLARES-BARBERA

0168314,00TOTAL

27971,66Error

4490126,68Non-paying taxesLow Income class14

7320,44PoorLow Income class13

7566744,96Day laborersLow Income class12

3840,23Unemployed workers11

4880,29FarmersLow Income class10

6820,41Retired armyMiddle income class9

161979,62ArmyMiddle income class8

25681,53Employed workersMiddle income class7

11410,68ClergiesMiddle income class6

25471,51OwnersHigh income class5

122897,30IndustrialsHigh income class4

15650,93ManufacturersHigh income class3

18951,13MerchantsHigh income class2

44602,65Land ownersHigh income class1

Number% ProfessionIncome class

Table 3. Categories of population by income: high income class, medium income and lower income

Questions about spatial justice• Is this situation always possible?• Could we have transferability without any

strategic policy to guide where and who gets wellbeing?

• How does livability can be increased in cities? • Can urban policies be transformed and adjusted

to fulfill population needs? • Might mechanisms to distribute resources in

space help to achieve a more even spatial justice?

The Neolithic site of Çatalhöyük (Anatolia, Turkey)

http://www.catalhoyuk.com/history.htmlhttp://www.sfgate.com/c/pictures/2005/04/18/mn_catalhoyukgrf.jpg

http://www.felsefeekibi.com/dergi7/grafik/catalhoyuk.JPG

Çatalhöyük (Anatolia, Turkey)

Network Analyst Location-Allocation Models

• . Minimize impedance. Minimize weighted impedance (P-Median). The option solves the warehouse location problem. It chooses facilities such as the total sum of weighted impedances (demand allocated to a facility multiplied by the impedance to the facility) is minimized.

• Maximize Coverage• . This option solves the fire station location problem. It chooses facilities such that all or the

greatest amount of demands is within a specified impedance cutoff.• Maximize coverage/Minimize facilities• . This option solves the fi station location problem. It chooses the ,minimum number of facilities

needed to cover all or the greatest around of demand within a specified impedance cutoff.• Maximize attendance• . This option solves the neighborhood store location problem where the proportion of demand

allocated to the nearest chosen facility falls with increasing distance. The set of facilities that maximize the total allocated demand is chosen. Demand further than the specified impedance cutoff does not affect the chosen set of facilities.

• Maximize market share• . This option solves the competitive facility location problem. It chooses facilities to maximize

market share in the presence of competitive facilities. Gravity model concepts are used to determine the proportion of demand allocated to each facility. The set of facilities that maximizes the total allocated demand is chosen.

• Target market share• . This option solves the competitive facility location problem. It chooses facilities to reach a

specified target market share in the presence of competitive facilities. Gravity model concepts are used to determine the proportion of demand allocated to each facility. The minimum number of facilities needed to reach the specified target market share is chosen.

Parks, interval time in minutes

ArcGIS_10 ArcScene

Winter solstice

Summer solstice