Your browser doesn't support the features required by impress.js, so you are presented with a simplified version of this presentation.

For the best experience please use the latest Chrome, Safari or Firefox browser.

Un fisico nel mondo dei BigData

Indaco Biazzo
Politecnico di Torino
SmartData@PoliTO - DISAT
personal page --
CityChrone --
Chi sono e perché mi ritrovo a farlo?

Oggi abbiamo i Dati


Generic placeholder image

High entrance barriers

You need a Ph.D. in the specific domain to do research. But sometimes even to understand the results.

Generic placeholder image

Overspecialization in Science

Very difficult to cross the branches of science.

Slow learning curve

Very difficult to perform self-education path

images:[ Branch Tree Drawing Clip art - branches clipart @kisspng ,]
Generic placeholder image

"Simplicity is the ultimate sophistication"

Data visualization

Interactive visualizations of results.

Generic placeholder image


Involve general audience to partecipate.

images:[ Jegi - flickr ]
Motivations 2

I was born in Rome

I had a very difficult childhood

Rome public transport are "not so good".
Ok. But how much compared to the other cities?

Where is the better served [by public transport] place in the city?
And in the world?




Sociality, trades ...

Urbanization and cities:
The birth of cities.

The history of foundation.

Created by: @galka_max

World City Populations.

Data, visualizations, algorithms

Open data sources used

Public Transports Schedules (GTFS format) - transitfeeds

Street graph - OpenStreetMap

Populations data - Eurostat Population Grid, SEDAC

city data and boundary - measuringurban - oecd

DataViz inspirations




Urban Accessibility measures
Urban Accessibility measures

Huge scientific literature

The first definition of accessiblity in urban context is done more than 50 years ago

Many different definitions of accessibility

But no attemp to compute it at large scale.

A science of city needs quantitative measurement.

We want easy to understand, easy to compute and meaningful quantities to measure public transport efficiency.

And we define and measure them.
Boundaries and Tessellation.
It is possible to compute isochrones
First step towards an accessibility measure:
The larger isochrones are, the faster you move.
Velocity Score
Average velocity taking a random direction
Paris Rome
... ... ...

interactive maps and more cities:
Sociality Score
Number of people is possible to reach in a typical day trip starting from a point.
Paris Rome
... ... ...

interactive maps and more cities:
City Rankings
City Velocity
Velocity Score per person
City Sociality
Sociality Score per person
City Sociality divided by total population
Interactive platform
Citizen Science [DataViz & Gamification]
SETI@home [1999]: analyze radio signals, searching for signs of extraterrestrial intelligence. People can partecipate using their PC, donating their computational resources.

foldit [2008]: fold the structures of selected proteins as perfectly as possible, using tools provided in the game. Nature paper with credits more than 57000 authors.

Quantum Moves [2012]: simulations of logical operations in a quantum computer. Played over 8 million times by more than 200,000 players worldwide. The 200 000 players were all beaten by the stochastic optimization method. :(
Algorithms: routing in urban context

Walking routing algortimh:

New class of public transport routing algorithms:
CSA [2013], RAPTOR [2011]

This new class of algoritms are easy to implements and fast, but they have some crucial limitations in urban context.

They needs to be closure by transitiveness in the walking path.

We modified the CSA and the RAPTOR algorithm in order to use it in urban context.

Interactive platform

Now I know how much Rome public transports suck

What we have to do to reach Paris?

What are the best interventions given a budget?

Let's Play!
Interactive platform for exploring new scenario
Budget: 5 Bilion €
Name Scenario: Gram Author: Pietro
After 1 year
Name Scenario: rer + circle Author: mat
Using users mind and computational resources

Prior knowledge of users.

Client-side computations.

Gamification aspect.

Useful Links

Interactive platform:

Open Source and Open Data:


theoretical results [arxiv]

Today presentation:



Direzioni Future:

costruisci la tua metro:

il futuro del trasporto pubblico:

qualche slide

L'intelligenza artificiale:

risorse online

The future of public transports in cities
Bad ending for my current research, but happing ending for public transport in the cities?
Cars per 1000 inhabitants

Italy togheter with USA has the highest level of car ownership.

Italy cars Europe cars
Rome 800 Paris 225
Milan 596 London 298
Turin 600 Barcellona 350
Catania 700 Berlin 297
Average person per car 1.2

95% of the time the cars are parked
Self driving cars (they are around us)
No property - No Parking
Boost in efficency

Sharing Trips
from taxy sharing to trip sharing\(^{1,2,3}\)
At least 50% less cars circulating

Public transport on demand
shrinking of the cost urban transportation of almost 10 times.

1. P. Santi, G. Resta, M. Szell, S. Sobolevsky, S. Strogatz, C. Ratti. Taxi pooling in New York City: a network-based approach to social sharing problems (2013).
2. hubcab
3. shared-mobility-innovation-liveable-cities