1st ATRC Researcher Workshop

Anthony Kimpton

16 March, 2021

UQ’s Transport and Population Research Network (TPRN)

School of Civil Engineering [1]

Mark

Prof Mark Hickman: public transit planning and operations and remote sensing technology for traffic management.

School of Civil Engineering [2]

Carlo

Prof Carlo Prato: behavioral/choice modeling for understanding modal choice.

School of Earth and Environmental Sciences [1]

Jon

Prof Jonathan Corcoran: urban modeling, geo-visualization, geo-analytics, and prediction techniques

School of Earth and Environmental Sciences [2]

Yan

Assoc. Prof Yan Liu: cellular automata, agent based urban modeling, spatial data mining, and big data analytics for understanding social spatial issues

Developer/Postdoc

Ant

Dr Anthony Kimpton: land use and transport planning, smart cities, social equity, neighborhood effects, social sustainability/urban resilience, geo-spatial modeling, and data visualization.

Land Use and Transport Policy

fp

Predict and Provide, Multimodalism, and Demand Management

Planning
Location
Modal
Problem
Orientation Approach Residential Workplace Transport Link 1 Link 2 Link 3 Link 4 First Mile Last Mile
Pragmatist Predict and Provide suburban inner-city Private drive no no
suburban inner-city Public walk/cycle direct public transport walk yes yes
Multimodalism suburban inner-city Intermodal drive rapid public transport walk no yes
suburban inner-city Intermodal passenger rapid public transport walk no yes
suburban inner-city Transmodal walk/cycle feeder public transport rapid public transport walk yes yes
Demand Management transit node inner-city Public walk/cycle rapid public transport walk/cycle yes yes
Radical 24-hour city same location Active walk/cycle no no

Slack Time?

st

Ride-hailing, Rides-sharing, Bay-Sharing, and MaaS

bs

Cloud Kitchens

cc

Park `n’ Ride Overflow

of

Railheading Behaviour

rh

Visualising Australian Trip Chains [1]

236 combinations of three modes

tb

Visualising Australian Trip Chains [2]

  • 8 sets (i.e., taxi, tram, motorbike, bicycle, ferry, walk, bus, car, train)

  • 255 potential Venn intersections on a 2D page (i.e., 2^n – 1)

  • The ABS has observed 236 intersections

Visualising Australian Trip Chains [3]

2011 Inner-Sydney Worker Modal Choices:

Visualising Australian Trip Chains [4]

2016 Inner-Sydney Worker Modal Choices

Data Visualisations can…

  • Eliminate noise

  • Transfer ideas quickly

  • Provide policy feedback

  • Reduce the expertise barrier

South East Queensland Data

General Transit Feed Specification (GTFS) Data [1]

South East Queensland GTFS Data Model

General Transit Feed Specification (GTFS) Data [2]

  • Query 3-months of stops, schedules, routes, and trips (services)

  • Link with agency smart card data for flows

  • Opportunity for visualizing routes, flows, and services

General Transit Feed Specification (GTFS) Data [3]

Varying temporal coverage

General Transit Feed Specification (GTFS) Data [4]

Limits archiving

General Transit Feed Specification (GTFS) Data [5]

The South East Queensland Official GTFS
Name Length Date
agency.txt 133 2021-02-04 01:09:00
calendar.txt 8183 2021-02-04 01:09:00
calendar_dates.txt 5236 2021-02-04 01:09:00
feed_info.txt 191 2021-02-04 01:09:00
routes.txt 129865 2021-02-04 01:09:00
shapes.txt 49231500 2021-02-04 01:10:00
stops.txt 1571931 2021-02-04 01:09:00
stop_times.txt 135438882 2021-02-04 01:10:00
trips.txt 7838635 2021-02-04 01:09:00
The Victorian Official GTFS(s)
Name Length Date
1/ 0 2021-02-25 09:28:00
10/ 0 2021-02-25 09:28:00
10/google_transit.zip 4500 2021-02-25 09:15:00
11/ 0 2021-02-25 09:28:00
11/google_transit.zip 32124 2021-02-25 09:16:00
1/google_transit.zip 11594799 2021-02-25 09:05:00
2/ 0 2021-02-25 09:28:00
2/google_transit.zip 9696771 2021-02-25 09:07:00
3/ 0 2021-02-25 09:28:00
3/google_transit.zip 13241387 2021-02-25 09:10:00
4/ 0 2021-02-25 09:28:00
4/google_transit.zip 62867641 2021-02-25 09:28:00
5/ 0 2021-02-25 09:28:00
5/google_transit.zip 41937222 2021-02-25 09:12:00
6/ 0 2021-02-25 09:28:00
6/google_transit.zip 14182198 2021-02-25 09:15:00
7/ 0 2021-02-25 09:28:00
7/google_transit.zip 50417 2021-02-25 09:15:00
8/ 0 2021-02-25 09:28:00
8/google_transit.zip 581398 2021-02-25 09:15:00

Limits interoperability

General Transit Feed Specification (GTFS) Data [6]

South East Queensland
stop_id stop_code stop_name stop_desc stop_lat stop_lon zone_id stop_url location_type parent_station platform_code
1 1 Herschel Street Stop 1 near North Quay NA -27.5 153 1 http://translink.com.au/stop/000001/gtfs/ 0
10 10 Ann Street Stop 10 at King George Square NA -27.5 153 1 http://translink.com.au/stop/000010/gtfs/ 0
100 100 Parliament Stop 94A Margaret St NA -27.5 153 1 http://translink.com.au/stop/000100/gtfs/ 0
Perth
location_type parent_station stop_id stop_code stop_name stop_desc stop_lat stop_lon zone_id supported_modes
0 NA 10000 10000 Albany Hwy After Armadale Rd -32.1 116 4 Bus
0 NA 10001 10001 Albany Hwy After Frys L -32.1 116 3 Bus
0 NA 10002 10002 Albany Hwy After Clarence Rd -32.1 116 3 Bus
Victoria
ï..stop_id stop_name stop_lat stop_lon
17204 Wallan Railway Station (Wallan) -37.4 145
19980 Melton Railway Station (Melton South) -37.7 145
19981 Rockbank Railway Station (Rockbank) -37.7 145

GTFS libraries fail and/or lose information

ABS Journey to Work [1]

  • Australia-wide

  • Long or matrix format

ur pow Train Bus Ferry Tram Taxi Car..as.driver Car..as.passenger Truck Motorbike.scooter Bicycle Other Walked.only Worked.at.home Total
LGA10050 LGA10050 0 68 10 3 32 9968 977 118 110 252 53 815 601 13004
LGA10110 0 0 0 0 0 0 0 0 0 0 0 0 0 0
LGA10150 0 0 0 0 0 0 0 0 0 0 0 0 0 0
LGA10200 0 0 0 0 0 4 0 0 0 0 0 0 0 11
LGA10250 0 0 0 0 0 0 0 0 0 0 0 0 0 0
LGA10300 0 0 0 0 0 0 0 0 0 0 0 0 0 0
LGA10350 0 0 0 0 0 0 0 0 0 0 0 0 0 0
LGA10470 0 0 0 0 0 0 0 0 0 0 0 0 0 0
LGA10500 0 0 0 0 0 0 0 0 0 0 0 0 0 0
LGA10550 0 0 0 0 0 0 0 0 0 0 0 0 0 0

ABS Journey to Work [2]

Changing geographic standards and coding

Household Travel Survey [1]

  • Explore trip stops and purpose(s)

  • Household and personal characteristics for explaining and as weightings

  • Suitable for visualizing OD flows that can be filtered by trip purpose

Household Travel Survey [2]

Queensland Travel Survey

Household Travel Survey [3]

Victorian Integrated Survey of Travel and Activity (VISTA)

Household Travel Survey [4]

Limited supporting documentation and so more readily available as fact sheets

Data Visualisations

Routes [1]

TRAVIC

Routes [2]

Suitable for Visualizing:

  • GTFS services and stops

Flows [1]

flowmap.blue

Flows [2]

Suitable for Visualizing:

  • GTFS service flows

  • GTFS ridership flows if paired with smart cards

  • ABS Journey to Work data

  • Household Travel Surveys

Services, Traffic, and Disruptions [1]

TRAVIC

Services, Traffic, and Disruptions [2]

Suitable for Visualizing

  • Real time transit and traffic

  • Real time disruptions

  • GTFS as GPS-type data

  • Variability between schedules and services

  • Explore how transit riders and motorists respond to disruptions?

Casestudy: Greater Melbourne

Longitudinal Transit Data

  • Merge historical GTFS

  • Merge most recent GTFS with historical GTFS

  • Scheduled for regular updates

  • API access and AURIN download button

  • Supporting documentation for service, route, stop, spatial and temporal queries e.g. with R’s tidytransit and Python’s GTFSpy

Quick Example

Quick PTV GTFS Sample

GPS-type Data [1]

id shape_id trip_id trip_number route_type shape_pt_lon shape_pt_lat departure_time stop_id stop_sequence dist cumdist cumtime
1 2-GLW-C-mjp-1.10.R 5027.T5.2-GLW-C-mjp-1.10.R 1 2 145 -37.8 05:27:00 19906 1 0.0 [m] 0.0 [m] 0.000 [s]
2 2-GLW-C-mjp-1.10.R 5027.T5.2-GLW-C-mjp-1.10.R 1 2 145 -37.8 05:27:00 19907 2 55.4 [m] 55.4 [m] 0.104 [s]
5 2-GLW-C-mjp-1.10.R 5027.T5.2-GLW-C-mjp-1.10.R 1 2 145 -37.8 05:27:01 19908 3 119.7 [m] 586.4 [m] 0.929 [s]
26 2-GLW-C-mjp-1.10.R 5027.T5.2-GLW-C-mjp-1.10.R 1 2 145 -37.8 05:27:02 19843 4 435.4 [m] 2993.2 [m] 2.273 [s]
35 2-GLW-C-mjp-1.10.R 5027.T5.2-GLW-C-mjp-1.10.R 1 2 145 -37.8 05:27:03 19842 5 470.4 [m] 3822.7 [m] 3.115 [s]
48 2-GLW-C-mjp-1.10.R 5027.T5.2-GLW-C-mjp-1.10.R 1 2 145 -37.8 05:27:05 19841 6 600.5 [m] 4972.9 [m] 4.633 [s]
50 2-GLW-C-mjp-1.10.R 5027.T5.2-GLW-C-mjp-1.10.R 1 2 145 -37.8 05:27:05 22180 7 460.7 [m] 5550.2 [m] 5.092 [s]
84 2-GLW-C-mjp-1.10.R 5027.T5.2-GLW-C-mjp-1.10.R 1 2 145 -37.8 05:27:06 19854 8 60.6 [m] 7339.3 [m] 6.092 [s]

GPS-type Data [1]

Ready for queries and reveals service scheduled and/or density

Interactive Figures

  • Victoria already uses Tableau

  • Dashboards contextualize the graphics and can contrast Victoria to other Australian cities

Journey to Work Data [1]

  • Cleaned e.g., grouped as active, public, and private transport

  • Harmonized i.e., with concordance files

  • Visualized

  • AURIN download buttons and API access

Journey to Work Data [2]

flowmap.blue

Conclusion

Project Aims

Support the Australian transport decision makers and practitioners, and the international transport research community:

  • Static and real-time visualizations i.e., routes, flows, services, and disruptions

  • Cleaned and harmonized SEQ longitudinal data

  • Tutorials and shared scripts e.g., in R and Python

  • Visualizations paired with APIs and download buttons

  • Reduce the data and software expertise barrier

  • Minimize research “cold starts”

Next 6 Months [1]

Data:

  • Containerize i.e., Docker

  • Deploy i.e, NeCTAR

  • Integrate e.g., ATRC Platform and APIs

Next 6 Months [2]

Visualizations:

  • Developing as HTML widgets

  • Containerize i.e, Docker

  • Deploy i.e, NeCTAR

  • Integrate e.g., ATRC Platform

Time for Questions