Mobility

Lets a lot of Digital Footprints

Sensors everywhere ...

for whom, for what ?

Sensors :

surveys
counters
smart-card (transit, bike-share,...)
wifi
video
smartphone (apps, telecom-provider, ...)
credit-card

Utopy ?

Open the black box ?

Transparency / Trust / Personal data

Utopy ?

Better urban governance ?
Anatagonist interests ?

Distopy ?

Better urban governance ?
Observation limits ?

Quite voluminous !! Must be analysed

! communicate results

reappropriation

System, city perspective

Unsupervised learning, clustering, Data-Mining

Try to let the data speak themself

Digital Revolution and Mobility

Paradigm Shift From Transport to Mobility

Mobility viewed through the prism « purposes » : access to human activities
Trip generations or centers : work, training, health, culture, leisure,….
Link between Mobility and territories (urban shapes, location of persons and services, rythmes of organization of human activities)

Lifestyle transformation through digital

Emergence of new concepts: e-commerce, smart meters, mobile work, MOOC,...
Sharing Economy, in the Mobility domain: bike sharing, car sharing, carpooling, shared parking,...
Adaptation of activities generating movements, a better organization of lifestyles in the proximity areas (co-working place, working time arrangement...)

Digital Revolution and Mobility

Digital and mobility actors

Historic Transport Operators and new entrants
Better planification of mobility (less expensive, more energy efficient, reliable, shorter) than « go faster » (speed)
offer new services of mobility (multimodality), Digital Information platerform

Digital and Citizen (ou user)

User as provider of data on his movements (Ex. Tranquilien sncf API, PT blogs, Replica, a next-generation urban planning tool sidewalk labs-google)
User as co-producer of mobility services (Ex. carpooling)

Digital Revolution and Mobility

Data mining

Conventional tools : Travel Surveys (Household Travel Surveys, Origin-destination surveys, ...)
Non public data as being of general interest (Jutand report)
Combined exploitation of many sources of data, creation of value-added services for the community

Trends

Mobility renewed by digital revolution
Innovative and Sustainable Mobility

Operationnal objectives

For urban stakeholders (operators, transport authorities, city managers)

Better understanding of the dynamics of mobility and users,
Predict passenger flows
Help to adapt infrastructure
and networks (existing and emerging needs)
Achieve a better planning of transport supply and level of service

For citizens

Trip Planning tools
Provide better mobility services
Better organization of activities

Replica

We believe this powerful data source can help do just that. Meet Replica: a user-friendly modeling tool that uses anonymized mobile location data to give planning agencies a comprehensive portrait of how, when, and why people travel in urban areas. Replica provides a full set of baseline travel measures that are very difficult to gather and maintain today, including the total number of people on a highway or local street network, what mode they’re using (car, transit, bike, or foot), and their trip purpose (commuting to work, going shopping, heading to school, etc). By updating these measures every three months, Replica also provides the ongoing ability to detect changes in these measures over time — helping planners answer questions about land use and transportation from a regional level all the way down to a city block.

Replica

Replica uses this anonymized data from about 5 percent of the population to learn about travel patterns and create a travel behavior model — basically, a set of rules to represent who’s moving where, when, why, and how. But models aren’t perfect. So we gut check these rules using on-the-ground data (such as manual traffic counts or transit boardings) to make sure Replica is consistent with real-world movement patterns.

Replica

We then match these models with what planners often call a “synthetic” population. That’s a very technical term, but the basic idea is that planners can use incomplete samples of census demographic data to create a broad new data set that is statistically representative of the full population.

Replica

Smartphone app data huge dataset
5% user every 3 months
Anonymization
Model Bayesian Networks (data exchange public-private data)
Population synthesis
what if scenario

Replica

Status :

https://replicahq.com/
Data-gouvernance and tools?
https://www.bbc.com/news/technology-56168306

Other company examples

Forcity https://www.societe.com/societe/forcity-799957840.html status: dead
Cit.io (RATPdev) https://cit.io/ status : teenager
Modality https://with-modality.com/fr/ status : new born

Special data

Smart cards data

Smart Card Data designed for tarification purpose

spatially and temporally precise, longitudinal, quite exaustives

Secondary use of Smart Card Data for urban mobility analyzis

BUT

Missing Data (only origin, information about destinations is missing)
Large volume of data (storage and processing problems)
No socio economic data on the user (personal privacy)
Data enrichment (contextual data, socio-demographic data ….)
Decision-making tools, analysis and visualisation tools

Bike-sharing systems analysis

Visualize and analyze Vélib' data

2 data sources derived from smart-card data:

Stocks

in open-data (Real-Time Apps)

Flows

Origines / Destinations sometime in open-data (London, New-York, Boston,...) frequently not ()

Animated bikes stocks

A problem of balance !
A regulation challenge
A move // a repetitive pulse

Stocks data : vlsstat

Data can be used for more adavanced analysis :
$\Rightarrow$ historical data
Aggregation / Comparison
http://vlsstats.ifsttar.fr/rawdata/

Clustering of stocks

The Discriminative Functional Mixture Model
for the Analysis of Bike Sharing Systems

Origin/Destination data clustering

http://www.comeetie.fr/galerie/velib/

Station clustering model

$X_{sdt}$ (observed): number of bikes comming/leaving
$Z_s$ (latent): station s cluster
$W_d$ (observed): (week / week end), meteo, ...

Simple generative model :

$$Z_s\sim\mathcal{Multinomial}(1,\pi)$$ $$X_{sdt}|\{Z_{sk}=1,W_{dl}=1\}\sim\mathcal{Poisson}(\alpha_s\lambda_{klt})$$ + constraints $\sum_{l,t}D_l\lambda_{klt}=DT, \forall k \in\{1,...,K\}$,
with $D_l$ number of days in $l$.

Station clustering model

independent poissons distristributions
stations scaling factors $\alpha_s$
categorical information on days (day of week, meteo,...)

Urban dynamics through the observed flows (2 month of O/D data in Paris)

http://www.comeetie.fr/galerie/velib/

Urban dynamics through the observed flows

An interesting viewpoint
Easy interpretation
Crossing with contextual data
...

Crossing with contextual data // socio-eco

	hab/ha	emp/ha	serv/ha	com/ha
*	162	237	4.2	3.7
Leisure (1)	367	189	6.3	4.4
Leisure (2)	261	322	7.7	6.9
Parks	172	90	2	1.7
Stations	209	206	2.4	1.8
Housing	375	108	3.8	2.7
Jobs(1)	138	409	4.5	2.8
Jobs(2)	157	456	5.7	5.6
Average	301	163	3.8	2.8

Origin/Destination matrix analysis
with Latent Dirichlet Allocation

Latent Dirichlet Allocation

For dynamic Origine-Destination matrices analysis

Local stationarity of BSS behaviour / OD
Small bags of successive trips $\approx$ stationarity of OD
Documents (bags of words) = bags of successive trips (5000)
With :

Words = Origin/Destination couples
Topics = Latent activities

Latent Dirichlet Allocation

For dynamic Origine-Destination matrices analysis

For each latent activity $a$, draw its template : $\Lambda_a\sim\mathcal{Dirichlet}(\beta)$
For each bag of trips
Draw the proportion of activities : $\pi_t \sim \mathcal{Dirichlet}(\alpha)$
For each trips
    Draw its activity
    $A \sim \mathcal{Multinomial}(1,\pi_t)$
    Draw an OD using the activity template
    $D \sim \mathcal{Multinomial}(1,\Lambda_A)$

OD decomposition results

Model selection with perplexity analysis
(clear drop for K=5)

Stations Balances : home→work

Stations Balances: work→home

Stations Balances : evening

Gravity-LDA

LDA extension for taking stations context into-accounts

Replace the O/D matrix templates $\Lambda_k$ by a parametric form which depends on :

context of departure station, $\mathbf{x_u}$
context of arrival station, $\mathbf{x_v}$
distance between $u$ and $v$, $\mathbf{x_{uv}^{da}}$

Inspiration from gravity models for O/D matrices

Gravity-LDA

LDA extension for taking stations context into-accounts

Replace the O/D matrix templates $\Lambda_k$ by a parametric form which depends on :

context of departure station, $\mathbf{x_u}$
context of arrival station, $\mathbf{x_v}$
distance between $u$ and $v$, $\mathbf{x_{uv}^{da}}$

\begin{equation} \Lambda_{\,uv}(\Theta_k) \,=\, \frac{ \exp(\mathbf{\theta_{k}^{d}}^\top \mathbf{x_{u}} + \mathbf{\theta_{k}^{a}}^\top \mathbf{x_{v}} + \mathbf{\theta_{k}^{da}}^\top \mathbf{x_{uv}^{da}}) } {\sum\limits _{u,v}\; \exp(\mathbf{\theta_{k}^{d}}^\top \mathbf{x_{u}} + \mathbf{\theta_{k}^{a}}^\top \mathbf{x_{v}} + \mathbf{\theta_{k}^{da}}^\top \mathbf{x_{uv}^{da}}) } \end{equation}

Estimation by Collapsed CEM

Conclusion on OD analysis

Results

Easy segmentation of time
Compact model of the system few template
Evidences of clear cycles
Also used for other type of data Bluetooth, smartcard,....
Can be interesting for simulation
Only satified demand is taken into account (users that do not find bikes!)

Extension

Socio-economic factors + od factors (distances, type of roads) can be integrated
$\Rightarrow$ Usefull for new station planning (system extension)

Bonus : cities portraits at Night

Bonus : cities portraits : atNight

Transit networks analysis

Smart-card data form

anonymous user id (changed every 3 month)
type of card
stop id (+line and direction for the bus)
date and time

Open-Data ? (ex: "Ile de France Mobilité" in aggregated form)

Smart-card data form

anonymous user id (changed every 3 month)
type of card
stop id (+line and direction for the bus)
date and time

A particular field : user id

A massive dataset

anonymous user id (changed every 3 month)
type of card
stop id (+line and direction for the bus)
date and time

2 year of data

Outline

What to do without user ids ?

Analyzing in-flow volumes
Days clustering
→ outliers detection, mid term prediction,...
Stations clustering
→ spatial analysis of the demand, ...

What to do with user ids ?

Destination inference
→ transfers analysis, dynamical OD matrices,...
Short term OD matrix prediction
Users clustering

Analyzing in-flow volumes

Profiles of the demand with spatio-temporal variations

Between day variations clearly visible (CAH)

Which is mainly explained by calendar effects

Which can be used to detects outliers

#Rennes #metro #Star des chaises jetées sur la ligne aérienne de métro à Villejean. Dégâts importants. Trafic interrompu pendant 2h?
— Samuel Nohra (@SamuelNohra) 29 mars 2016