2014/08/28 webinar by Marcela A. Munizaga See more in: http://www.brt.cl/webinar-using-smart-card-and-gps-data-for-policy-and-planning-the-case-of-transantiago/

1. Using smart card and GPS data for
policy and planning: the case of
Transantiago
Marcela A. Munizaga
Universidad de Chile
Visiting CTS
Research Team: Universidad de Chile – Transantiago
Research grants: CONICYT PBCT, Milenio Scientific Initiative, FONDEF

2. Introduction. Transantiago: public
transport system Santiago, Chile
¤ Santiago. Capital City of Chile:
¤ Population: 6 million
¤ Area: 1,400 km2
¤ 34 Municipalities
¤ Modal Split: 50%

3. Introduction. Transantiago: public
transport system Santiago, Chile
¤ Transantiago
q Introduced in 2007
q 6.500 buses (65% low entry) with GPS
q 70 km of segregated busways
q 10.000 bus stops
q 125 bus stations (off-bus fare collection)
q 12 private bus operators
q 600 trunk and feeder services
q Metro: 5 lines, 100 km, 54 trains
q Only smartcard payment in buses
(global 97% penetration rate)

4. Map

5. Transantiago structure
¤ Operators: bus (private) + metro (public)
¤ Provide transport services, receive payment
(per passenger, per km, regularity,…)
¤ AFT (financial administrator)
¤ Collects and distributes money
¤ Collects and store data
¤ Transantiago authority DMTP
¤ Regulates (spatial coverage, fare, frequency)
¤ Controls

6. Quoting the Transantiago authority:
¤ “Before this project we were: “
§ Advancing slowly
§ With very little information
§ Lack of support tools
Carolina Simonetti, Director of Planning and Research,
DMTP. XVI Chilean Transport Conference,
Santiago, October 2013

7. The Data
AVL
Buses
GPS
Other
informa-tion
AFC
bip!
(metro &
buses)
OD trip matrices, buses speeds,
travel patterns, level of service
indicators…
• Buses GPS: 1 record
every 30s, 80–100 M
records per week
• bip! transactions: 35-40 M
records per week
• Other information:
• Routes paths
• Route assignments
• Position of bus stops
• Position of Metro
stations
• Position of bus
stations

8. What can we do with the data?
Observe buses

9. What can we do with the data?
Analyze transactions

10. What can we do with the data?
Analyze transactions

11. What can we do with the data?
Link vehicles and passengers through vehicle id

12. Processing
¤ Estimation of alighting stop
Second'transac,on'
of'the'day'
Last'transac,on'
of'the'day'
First'transac,on'
of'the'day'
Min'Tg'
Min'Tg'
Min'dist'
Boarding'point'
GPS'Point'
Bus'stop'
Metro'sta,on'
i min Tg = ti +
di−>xpost ypost
swalk
⋅(θ walk /θ travel)
s.t. dpost ≤ d

13. Post-Processing: Stages and Trips
Trip Trip t
Observed
boarding
Estimated
alighting
Transfer or
activity?
Determination of:
– Trips/stages
– Time. distance and speed of
transfers. stages and trips
– Walking and waiting time
Criteria to distinguish destination
from transfer
– Time elapsed
– Transaction sequence
– Land use
– Frequency of PT services
– Ratio: distance on the route /
Euclidean distance

14. Post-Processing: Visualization
¤ Passengers boarding and alighting in origin and destination

15. Validation
¤ We are able to estimate alighting location-time in 80% of trip
stages, generating over 20M trip observations in a week
¤ Validation with small OD survey:
¤ 84% correct estimation of alighting position-time
¤ Validation with a sample of volunteers:
¤ 90% correct estimation of trip/trip stage separation
q Disclaimers:
q Validation with large ODS to be conducted
q Fare evasion not included
q Exact Origin/Destination unknown
q Sociodemographic characteristics unknown

16. Commercial speed of buses
q Estimation of commercial speed of buses
q Associate position to linear route distance
q Define time-space disaggregation
q Monitor in time-space diagram
q Estimation of commercial speed for bus corridors
q Modelling

17. Bus following
Time-space diagram for buses of a route

18. No se puede mostrar la imagen. Puede que su equipo no tenga suficiente memoria para abrir la imagen o que
ésta esté dañada. Reinicie el equipo y, a continuación, abra el archivo de nuevo. Si sigue apareciendo la x roja,
puede que tenga que borrar la imagen e insertarla de nuevo.
No se puede mostrar la imagen. Puede que su equipo no tenga suficiente memoria para abrir la imagen o que
ésta esté dañada. Reinicie el equipo y, a continuación, abra el archivo de nuevo. Si sigue apareciendo la x roja,
puede que tenga que borrar la imagen e insertarla de nuevo.

19. Speed range definition sR=20[km/hr]
Condition Sijk [Km/h] Color
Very bad ≤ 15 Red
Bad >15 a ≤19 Orange
Regular >19 a ≤20 Yellow
Acceptable >20 to ≤25 Light green
Good >25 to ≤30 Dark green
Excelent >30 Blue
n.a.: Grey

20. Global results (all services)
September 2008
March 2009
April 2009

21. Other visualizations
¤ Spatial visualization by service
¤ Worst cases in a map
¤ Speed of a corridor
¤ For all services
¤ All times of day
¤ Divided into segments

22. N Santa Rosa corridor S
Exclussive way
7:30-10 & 17-21
Mixed traffic
3 lanes 10-17
Seggregated corridor
2 continue lanes per direction
Mixed traffic
2 lanes
Segment 7 6 5 4 3 2 1
Length (km) 0.99 1.39 1.97 1.63 1.18 1.41 0.97
Traffic light
controlled int/ km 3.03 4.32 3.55 3.68 2.54 3.55 3.09
Bus stops
service/km 2.02 2.88 2.54 2.45 2.54 2.13 2.06

23. Average commercial speed Santa Rosa corridor(km/h)
Segment
Period 1 2 3 4 5 6 7 Total
7:30 18.5 15.1 29.3 24.8 24.4 17.2 10.4 18.1
8:00 14.6 16.3 30.9 26.5 25.2 17.3 9.7 19.3
8:30 13.7 18.8 32.7 27.2 27.4 18.0 9.2 18.2
9:00 15.2 20.2 34.9 28.0 29.3 19.2 14.7 21.5
9:30 16.3 19.8 33.7 27.5 29.7 18.5 14.9 21.5
10:00 16.2 21.6 34.5 30.6 31.6 22.2 17.1 22.8
10:30 19.9 21.0 33.7 30.2 31.6 21.9 17.4 23.8
11:00 19.7 21.6 30.4 31.2 31.6 21.2 17.1 23.2
11:30 19.6 21.2 33.0 32.0 31.4 21.4 16.8 23.9
12:00 18.5 20.3 36.3 31.1 32.0 20.9 15.1 24.4
12:30 18.5 21.0 35.2 30.0 32.9 21.1 14.8 23.5
13:00 18.9 21.2 35.1 31.6 32.1 21.1 15.5 24.2
13:30 20.4 21.0 35.6 32.2 32.7 22.4 16.8 24.8
14:00 19.8 21.4 36.3 31.0 33.1 22.7 17.8 25.1
14:30 21.1 21.4 33.2 28.8 31.1 21.6 20.1 24.7
15:00 22.1 19.9 32.4 29.5 30.6 22.4 17.9 24.3
15:30 20.5 21.4 30.0 28.7 31.1 21.3 17.1 23.4
16:00 17.0 21.4 30.5 28.3 32.2 21.8 16.9 22.7
16:30 18.1 20.3 32.6 28.2 31.4 22.0 16.9 23.1
17:00 15.7 20.5 31.2 29.1 27.2 22.7 15.0 21.7
17:30 17.7 20.3 29.7 29.2 27.4 22.9 15.2 22.7
18:00 14.5 20.8 30.5 28.8 27.5 23.1 15.8 21.7
18:30 22.6 21.5 30.2 29.8 29.0 25.8 15.6 23.8
19:00 23.1 23.5 30.0 29.6 29.5 26.5 19.0 25.3
19:30 23.6 23.6 30.5 31.5 29.5 28.7 20.2 26.3
20:00 26.7 24.9 31.3 31.6 30.5 29.8 22.3 27.8
20:30 27.7 26.7 32.5 33.5 31.3 33.2 25.7 29.8
Total 18.8 20.8 32.2 29.5 29.9 22.1 16.0 23.1

24. Post processing
¤ Load profiles
Built using
¤ Bus trajectory
¤ Observed boarding with expansion
factors
¤ Estimated alighting with expansion
factors
à Aggregated at bus or route level

25. Route load profile

26. 0.9
0.8
0.7
(million)
0.6
0.5
0.4
0.3
0.2
0.1
0
Regular
card
–
days
used
in
a
week
1
2
3
4
5
6
7
cards
days
sep.08
ago.09
jun.10
abr.11
abr.12
200
180
160
(thousands)
140
120
100
80
60
40
20
0
The
most
frequent
user
is
the
unfrequent
traveller,
but…
there
is
an
important
number
of
regular
users
Student
card
–
days
used
in
a
week
1
2
3
4
5
6
7
cards
days
sep.08
ago.09
jun.10
abr.11
abr.12
The
most
frequent
behavior
for
students
is
frequent
traveller
Travel patterns

27. Zone of residence estimation for frequent users
Day
1.
07:18
am
Day
2.
07:38
am
Day
3.
10:53
am
Day
4.
09:02
am

28. Zone of residence estimation for frequent users
R
=
500
m
Day
1.
07:18
am
Day
4.
09:02
am
Day
2.
07:38
am
Day
3.
10:53
am

29. Other developments
¤ Trip purpose
¤ Level of service indicators
¤ Time use patterns
¤ Fare evasion

30. Applications
¤ OD matrix at different levels
of aggregation (XY, bus
stop, zone, municipality)
¤ Route/service design
¤ Infrastructure decisions
¤ Design of information
campaigns
Recoleta
203-208
Fusion
Lira-Carmen
204

31. Applications
¤ Speed profiles
¤ Operational interventions
¤ Bus priority decisions
¤ Infrastructure investment
decisions
B
F
F
C
C
A
D
D
E
E
G
G

32. Applications
¤ Load profiles
¤ Frequency optimization
¤ Design of express or short
variations of services
800
700
600
500
400
300
200
100
0
Perfil
de
carga
Servicio
104
Puente
Alto
-­‐ Providencia
(7:30)
subidas bajadas Carga

33. Quality of the information?
b. Santiago 2001 ODS data
One
example:

34. Applications
¤ Third party applications
(open access policy)

35. Conclusions
q Quantum leap on information
availability and cost
q Many tools can be developed to improve
planning, operation and control
q We can advance on understanding
behavior and test hypothesis
q Solid grounds to formulate new policies

36. Further research
¤ Additional information:
¤ Vehicle detectors
¤ Private GPS equipment
¤ Mobile phone traces
¤ Online applications (waze)
¤ Surveys!
¤ New age for transport engineering

37. Thanks!
Cortés, C., Gibson, J., Gschwender, A., Munizaga, M.A.,
Zúñiga, M. (2011) Commercial bus speed diagnosis based on
GPS-monitored data. Transportation Research C 19(4),
695-707.
Devillaine, F., Munizaga, M.A., Trepanier, M. (2012) Detection
activities of public transport users by analyzing smart card
data. Transportation Research Record 2276, 48-55.
Gschwender, A., Ibarra, R., Munizaga, M., Palma, C. (2012)
Monitoring Transantiago through enriched load profiles
obtained from GPS and smartcard data. CASPT Santiago,
Chile 23-29 Julio.
Munizaga, M.A., Palma, C. (2012) Estimation of a
disaggregate multimodal public transport origin-destination
matrix from passive Smart card data from Santiago, Chile.
Transportation Research 24C(12), 9-18.
Munizaga, M.A., Devillaine, F., Navarrete, C., Silva, D. (2014)
Validating travel behavior estimated from smartcard data.
Transportation Research 44C, 70-79.