The hidden role of people in understanding how cities work

[Applause]

why do we live in cities

and why do most cities worldwide

continue to grow

we might not have a clear answer to

these pressing questions as yet

but what is pretty clear is that cities

are the engines of

economic growth innovation and wealth

creation

and that to achieve this cities actually

need to connect us

their people cities bring together

people with different skills and

backgrounds

interests and cultures which then sparks

new ideas and

trends and creates new jobs

in fact empirical evidence shows that

not only wealth indicators like per

capita gdp

tend to increase with city population

size

we have found that human interactions

accelerate at

almost exactly the same pace in other

words

the interactions between the people are

key for the social

and economic functioning of cities

this fundamental role of human

interactions

is thus absolutely crucial for urban

planners

who are responsible for the design of

public spaces and infrastructures

that should facilitate the connectivity

between the people

in the past urban planning has indeed

often been quite successful

leading to attractive public spaces that

are now frequented by many people from

all walks of life

but unfortunately there are also

numerous cases where urban planning

interventions didn’t lead to the desired

outcomes

as is manifested in empty and sterile

plazas

traffic jams or even social segregation

what makes it then so hard to design

well-functioning cities well one of the

reason

is that in the past we didn’t

sufficiently understand

how people actually make use of urban

space

let’s just have a look at manhattan in

new york

we see people coming and going we see

taxi drivers going there for work

we see new yorkers who might go there

for shopping

or we see tourists who might go there

just once in a lifetime from very far

away

there is a myriad of different interests

and preferences that bring people here

and the result looks pretty random and

chaotic right

unfortunately it is exactly such

seemingly chaotic patterns

that urban planners need to deal with

when trying to avoid potential planning

pitfalls

but what if i told you that these

movements we just saw

are not at all random or chaotic but

that they are surprisingly structured

and predictable and that they follow

hidden regularities

that provide powerful guidelines for

urban planning

imagine you’re standing on a public

plaza which is full of people

coming and going now imagine you ask

all these people from how far away they

are actually coming from

this is exactly what my collaborators

and i did for boston in the u.s except

that we didn’t need to stand on a plaza

for several days

instead we could analyze millions of

anonymized mobile phone location data

that have been provided to us for

scientific purposes

and in an aggregate form so as to ensure

data privacy

so here is the result of our simple poll

for newbury street

a famous shopping area in boston on the

horizontal axis

we have from how far people visited

newbury street

and on the vertical axis we have the

number of visitors

just to be precise each data point is

the number of visitors

coming from a one square kilometer area

at the given distance away

and what we see is that if we go further

away from newbury street

we have less people coming this is of

course

nothing really surprising right who

wants to travel

all through the city just to do the

groceries

however our data now allow us to go one

step

further and to also ask how often people

are actually coming

here our same result except that we now

distinguish

between how often each person visits

newbury street

red is the number of people coming about

once per month

green is the number of people coming

about four times a month

and yellow is the number of people

coming about 10 times a month

and what we see is that the number of

visitors

also decreases the more often they are

coming

so we already start to see some very

systematic patterns

in this seemingly chaotic movement of

people right

but now comes the really surprising part

all these patterns

can actually be predicted by a simple

but

powerful mathematical travel law all

that really matters

is to just multiply the distance with

the number of visits

for instance the number of people coming

once per month

from 20 kilometers which is the red

square here

is about the same as the number of

people coming about four times from five

kilometers

which is the green square here and it is

also about the same as the number of

people

visiting ten times from two kilometers

since if we multiply a distance with the

number of visits

we always get 20.

so in our simple poll we can simply

multiply

distance with the number of visits and

all our data points beautifully line up

let me put it this way if i spend just

one day

measuring how many people come to

newbury street

i immediately know how many people will

come over the next weeks from

1 2 or 10 kilometers away and how many

of them will visit once

twice or 10 times a month and you know

what the best part is

this travel law isn’t well it only for

newberry street

it applies to virtually all locations in

the greater boston area

actually let me rephrase even this it

happens

all across the world we looked into more

data

and found that the very same travel law

holds for cities in europe

africa and asia regardless of the

detailed geographies

cultures or levels of development this

travel law is pretty amazing isn’t it

but how can it now help urban planners

to design great public spaces

well first of all having an idea of how

far

and how often people are willing to

travel helps to spot the best locations

to put the new public space such as a

park

and to have an estimate of how many

people such a new park can potentially

attract

second infrastructure planning

predictions of the population flows

especially to new urban developments is

essential

for the planning of public transport and

also for other infrastructures

we are actually applying exactly this

idea now

to support the electrification process

in a developing country

and third if we look at existing places

and if we compare the actual number of

visitors to our predictions

we can immediately identify those

locations

that attract less people than we would

actually expect

this signals a clear need to make such a

space more accessible or inviting

for additional population groups this

travel law is just one example of how

more science-based approach to cities

can help urban planning we are actually

just at the very beginning of revealing

many more of such exciting

and powerful regularities of how humans

connect to each other in urban space

cities are very complex there will

always be uncertainties

in doing such predictions for urban

planning think of bushwick brooklyn in

new york

neighborhoods that suddenly become hip

and trendy leading to escalating housing

prices

it is just very hard to predict such

dynamics

therefore it is not sufficient to

understand the basic laws of cities such

as the one i just showed

we additionally need to have early

warning indicators

that alert us of negative developments

such as social segregation

science-based approaches can certainly

help us here as well

and i’m pretty confident that we will

soon be able

to gain a much better understanding of

how to build

and maintain cities that truly enable

many diverse human interactions thank

you