Date post: | 01-Jul-2018 |
Category: |
Documents |
Upload: | trinhkhanh |
View: | 214 times |
Download: | 0 times |
1
Understanding people on the move in London
Charles BuckinghamImpacts Monitoring Manager
(with thanks to….,.)
Lauren Sager WeinsteinHead of Oyster Development
Contents
•
TfL’s responsibilities•
About Oyster
•
Using Oyster to understand people on the move
•
Congestion Charging –
some examples
2
About TfL
•
Mayor of London’s transport authority
•
Finances/procures/operates/maintains public transportation
–
London Underground
–
Buses
–
Docklands Light Railway
–
Croydon Tram
–
TfL Road Network – 580km of arterial roads
–
Congestion Charging
3
TfL’s responsibilities
4
TfL travel facts
Every weekday in Greater London:
6 million journeys are made on London’s buses
3.4 million on the Tube
11 million car / motorcycle trips
155k + passengers on DLR
9.5 million walking or cycling trips
70% of National Rail journeys begin or end in London
5
About Oyster
Source: TfL Fares & Ticketing Directorate
•
TfL’s multi-modal smartcard–
National rail–
London Underground–
Buses–
Tram–
DLR•
Contactless: 0.2s read/write at the reader•
3 x tickets + £90 PAYG with daily capping•
Distributed to customers free with a period travel product or a £3 returnable deposit: >16m issued to date
•
Concession & discount variants–
Freedom Pass for over-60s–
Various child & student schemes–
Bus & tram adult discount card
6
Penetration of Oyster
2004 2005 2006 2007
Source: TfL Fares & Ticketing Directorate
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
2004 2005 2006 2007
Oyster share of all TfL trips
Eliminated magnetic 7-
day bus passesEliminated TfL
magnetic weekly
TravelCards
(TCs)
Reduced Oyster
PAYG fares relative to
cash
Oyster PAYG
introduced
7
Key Oyster Benefits
Change in customer behaviour
Old: purchase a ticket and then travel
New: streamlined travel for customer
Reduces queues
Minimises cash handling
Tackles fraud
Speeds customers through gate
Source: TfL Fares & Ticketing Directorate
8
Using Oyster to Understand travelling behaviour
Key Oyster statistics
• As of January 2008, 17m + Oyster cards issued• 5.6m cards were in use during the previous 4 week period
•
During the week of 25 November -1 December 2007, on an average weekday there were:
3.1 million Oyster journeys a day on the Tube and DLR5.4 million Oyster journeys a day on buses and trams
•
In November 2007, Oyster card journeys represented around 74% of bus and Tube journeys.
9
Oyster Card Personal Data
• Oyster cards can be registered or unregistered
• Registered cards can be protected if lost or stolen
• Mandatory registration on monthly and annual tickets
•
Detailed journey history kept for 8 weeks for customer service purposes (eg
refunds)
• After 8 weeks, personal data is anonymised
10
Understanding travel patterns using anonymised
Oyster data
Analysis work supported through TfL partnership with MIT, with TfL guidance on crafting research questions
Sample research
11
Using Oyster to measure Variation of OD Journey Time
Oyster Journey
Time Percentile
10th 25th 50th 75th 90th
Minutes
Range of
journey time
experienced by
the middle
80% of
passengers
Range of
journey time
experienced
by the middle
50% of
passengers
•
Example ranges only
12
Can we use Oyster data to measure variability of journeys between stations?Research by Joanne Chan, MIT MST 2007
Results – Victoria Line (AM Peak Northbound)
Origin Stations to All Northbound Destinations
•
Skewed distribution•
Victoria, the only Zone 1 station in the graph–
Largest average excess minutes–
Largest variation in excess minutes
-1
0
1
2
3
4
5
Exce
ss M
inut
es
PC90
PC75
PC50
PC25
PC10
AVG
Brixton
(4,277)
Vauxhall
(3,365)
Victoria
(2,780)
Finsbury
Park
(287)# of Oyster Observations
Research by Joanne Chan, MIT MST 2007
Using Oyster to Measure Crowding
Can we use Oyster data to capture effects of crowding?Research by David Uniman, MST candidate 2008 The theoretical model:•
Imbalance b/w Travel Demand ↔ Transport Supply–
Platform Crowding
–
On-Train Crowding–
In-Station Crowding
•
Leads to Increased User Travel Times–
↑
Wait Times
–
↑
On-Train Times•
Through…–
Full Trains
–
Headway Variations
Dwell Times
Schedule Adherence
Platform Crowding
Full Trains14
Analysis: 30-min AM Peak Oyster
Victoria --> Oxford Circus
4
5
6
7
8
9
10
11
12
13
14
7:00-7:30am(54)
7:30-8am(134)
8-8:30am(460)
8:30-9am(315)
9-9:30am(296)
9:30-10am(180)
AM Peak
TT [m
in]
10th Percentile25th Percentile50th Percentile75th Percentile90th PercentileRP Corr2
RP Corr2 = TfL rail plan modelling tool, corrected to take into account Oyster journeys are gateline
to gateline
David Uniman, MST Candidate MIT 2008
Travel time increases at the peak; after the peak many journeys are still longer than early morning
15
16
Research Question
•
By focusing on bus passenger interchange behaviour, can Oyster data be used to help improve the public transport network in London?
•
Key contribution:–
Methodology for describing passenger interchange behaviour in London using Oyster card data
Catherine Seaborn, MIT MST candidate 2008
17
Journey Segments Per Passenger
0.0%
5.0%
10.0%
15.0%
20.0%
25.0%
30.0%
35.0%
40.0%
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Perc
enta
ge o
f Tot
al P
asse
nger
s
Number of Daily Journey Segments
Daily Journey Segments Per PassengerAll Oyster Card Modes
Source: 5% Oyster data for 2007 Period 2 (April 29 – May 26)
17.1%
34.9%
14.7% 14.1%
7.0%
4.8%
2.7%1.8%
1.1% 0.7% 0.4% 0.3% 0.1%0.2% 0.1%
•
Question: how many journey segments do Oyster customers take on
a given day?
Catherine Seaborn, MIT MST Candidate 2008
18
Weekday Journey Segment Patterns
Mode 1 Mode 2 Mode 3 Mode 4 Mode 5 Mode 6 Passengers ShareCumulative Share
U U 416,082 16.3% 16.3%
B B 401,356 15.7% 32.0%
B 266,561 10.4% 42.4%
B B B 150,781 5.9% 48.3%
B B B B 144,275 5.6% 54.0%
U 125,528 4.9% 58.9%
B U U B 77,353 3.0% 61.9%
B B B B B 72,943 2.9% 64.8%
U U U 65,190 2.6% 67.3%
B B B B B B 50,485 2.0% 69.3%
Source: 100% Oyster data for Wednesday, November 14, 2007
•
What are the modes for these journey segments?
•
Top 10 shown
•
Total patterns: 15,802
Catherine Seaborn, MIT MST Candidate 2008
19
Research Question
•
What are the characteristics of interchanges to bus at London Underground/bus interchange locations?–
How long does it take for passengers to transfer between modes?
–
Function of walk time, frequency of service, reliability
Catherine Seaborn, MIT MST Candidate 2008
Catherine Seaborn10 January 2008
20
Example Interchange Stations
21
Potential Interchange Time: Underground-Bus
Time Difference Between Underground Station Exit and Bus BoardingAll Oyster Card Passengers, Single Weekday
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
0 5 10 15 20 25 30 35 40 45 50 55
Time from Station Exit to Bus Boarding (minutes)
Perc
enta
ge o
f Pas
seng
ers
Boa
rdin
g B
uses
With
in 6
0 M
inut
es
.
Oxford Circus (38)Angel (19)Holloway Road (16)Burnt Oak (11)Northwood Hills (3)
Source: 100% Oyster data for Wednesday, November 14, 2007
Note: Number denotes total bus routes serving station, including night buses.
Catherine Seaborn, MIT MST Candidate 2008
Congestion charging in central London
22
Extended Central London charging zone
23
A transport success
24
•
Traffic entering charging zone: (4+ wheels)
Down 21%
•
Chargeable vehicles: Down 31%
•
Initial impact on congestion high: 30% decline (first yr) Averaging 21% over scheme lifetime
•
Nitrogen oxides (NOx
) emissions: Down 13% 8% due to Congestion Charging
•
Particulate matter (PM10
) emissions: Down 16%6% due to Congestion Charging
•
Carbon Dioxide (CO2
) emissions: Down 16%
Substantial traffic change
25
0
20,000
40,000
60,000
80,000
100,000
120,000
140,000
160,000
180,000
200,000
Cars and minicabs
Vans Lorries and others
Taxis Buses and coaches
Powered two-
wheelers
Pedal cycles
Total flow
Spring 2002Autumn 2002January 2003Feb/Mar 2003Spring 2003Autumn 2003Spring 2004Autumn 2004March 2005Spring 2005Autumn 2005November 2005Spring 2006Autumn 2006Spring 2007
Camera-based enforcement (1)
26
Camera-based enforcement (2)
27
Colour Contextual ImageColour Contextual Image
Monochrome Image from ANPR cameraMonochrome Image from ANPR camera
ANPR system outputANPR system output
Evidential Record SummaryEvidential Record Summary
Number Plate image from ANPRcamera, Lane 1
Number Plate image from ANPRcamera, Lane 1
Unique opportunity to study traffic characteristics and behavioural change
28
Potent data source:
Vehicle population profilesFrequency of travel etc.Some routeing/journey time information (congestion)Match with licensing data –
vehicles registered not same as vehicles
‘in the zone’
BUT:
Cameras capture vehicles NOT peopleOnly captures vehicles ‘there’
–
not those who have gone away
Do not capture whole tripData Protection imposes some (necessary) limitationsCan’t really use as sample frame for follow-on surveys Cameras optimised for enforcement NOT researchTend to be defeated by ‘easier’
things like data processing
So:Potential only partially fulfilled
Understanding our chargepayers
29
Measuring congestion
30
0.0
0.5
1.0
1.5
2.0
2.5
3.0
MarApr
MayJun
JulAug
SepOct
NovDec
JanFeb
MarApr
MayJun
MarApr
MayJun
SepOct
MayJun
NovDec
Exce
ss d
elay
(min
/km
)
Automatic number plate reading camera dataMoving car survey data
2003 2004 2005 2006
Understanding the effects of charging
31
Potential future developments - Tag & Beacon
Tag and beacon technology is already providing high capture rates for schemes where charges vary across the day, for example cordon charging varying by time of day
Stockholm 2006
The future? Satellite/mobile positioning systems
Satellite and mobile phone location systems for distance-based charging need further development for affordable use in urban areas
Example position reports from multiple different GPS and GSM mobile devices
Example from GPS trials - ‘use’ of zone varies by time of first entry
34
Thank You !
35
www.tfl.gov.uk