Transcript PowerPoint slide show

Analytics and Big Data —
Rail Public Transportation is a
Leader
Lyndon Henry
Blog Columnist, Railway Age Magazine
Co-Principal, Urban Rail Today
Technical Consultant , Light Rail Now Project
APTA 2013 Rail Transit Conference
Philadelphia
June 2013
What are …?
Big Data
Significantly large volumes
of data, particularly
involving human activities
and characteristics
Analytics
High technology applied to
data processing, complex
calculations, and automation
Common Examples
Private/Public Sector
Public Transportation
• Consumer behavior analysis
• Ridership forecasting
• Customer mailing lists/marketing
• Train signaling/dispatching
• Smartphone apps
• Route planning/scheduling
• GPS
• Automatic Vehicle Location (AVL)
• Financial market trading
• Passenger Information Systems
• Astronomical tracking/mapping
• Automated Fare Collection (AFC)
• Weather tracking/forecasting
• Automated Passenger Counting
(APC)
• Genome mapping
• Crowd surveillance
• Monitoring electronic
communications
• Data-mining online/wireless data
(Emails, texting, social media)
• Robots
• Driverless Automatic Train
operation (ATO)
Major Applications in
Public Transportation
Travel Demand Modeling &
Ridership Forecasting
• Forefront of Big Data &
Analytics for many decades!
• Project demographic
characteristics for each TAZ
• Project future growth in
population, economic
activity, travel patterns
• All trips from all
households in all zones to all
other zones
• Forecasting ridership
• Really Big Data!
Travel Analysis Zones in Baltimore Metro Area
Train Signal and Control Systems
• Legacy goes back to 19th century!
• High-tech, complex process for tracking location of trains,
estimating travel times, etc.
• From electro-mechanical to electronic-digital
• Automatic Block Signaling (ABS), Cab Signaling System
(CSS), Centralized Traffic Control (CTC), Automatic Train
Control (ATC), Communications-Based Train Control
(CBTC), Automatic Train Operation (ATO), Positive Train
Control (PTC), Driverless ATO
Route Planning and Scheduling
• Performs routine, tedious tasks — routing, developing
timetables, blocking into bus and train schedules, runcutting,
and other essential component tasks such as rostering
• Allows schedulers and transit planners to quickly develop
many different scheduling scenarios
• Significantly increases operational efficiency
Automatic Vehicle Location (AVL)
• Uses GPS-based data + Analytics to track both buses and trains
Passenger Information System
• Based on AVL or train signal/dispatching system
• Big hit with passengers
Automated Fare Collection (AFC)
• Uses ticket vending machines (TVMs)
• Smartcards provide access to all types of transit services
across multiple operating agencies and jurisdictions
• AFC Analytics provide details of how passengers are using
systems, identify trends, help improve service
Automated Passenger Counting (APC)
• Informs agency as to how many passengers are boardingdeboarding each vehicle, precisely where this happens
• Data can be used to provide better service and project
evolving ridership trends
Selected System
Examples
Bay Area Rapid Transit (BART)
• Analytics/Big Data is key element of train operations,
passenger services, power delivery, wayside facilities
• Operational analytics includes delay analysis, passenger
flow modeling (PFM), system performance analysis
• PFM ensures on-time service, optimizes train schedules
(ensuring high customer satisfaction) while constraining
service operating costs
Austin — Capital Metro's MetroRail
• CTC control center uses data radio as primary system,
cellular phone system as secondary backup
• GPS currently used mainly as component of passenger
information system (with PIDs in stations)
• APC used for gathering passenger statistics plus planning
and operations (e.g., adjusting schedules to accommodate
passenger flows, changes in traffic demand by time of day)
Philadelphia — SEPTA Regional Rail
• Wide variety of Analytics/ Big Data functions — operations control,
scheduling, AVL, train signaling/dispatching, AFC with TVMs, APC,
passenger information with PIDs, etc.
• Passenger information system provides PIDs in some larger
stations, plus app providing bus/train status to smartphones
• APC data to be used to adjust scheduled consists and track trips for
Job Access and Reverse Commute (JARC) services
Seattle – Sound Transit's Link, Sounder
• Analytics/ Big Data involved in signaling-controldispatching; passenger information with online and
smartphone train status information and station PIDs; APC;
GPS and AVL capabilities; AFC with TVMs in stations
• AFC system uses “seamless” regional, trans-agency ORCA
payment card
• Source of Big Data for local transit agencies — information
on individual passenger movements throughout region, plus
passenger flows at various locations and times of day
Current Issues and
Trends
Data Mining
• Poor public reputation, mainly because of privacy issues
• But data mining is typically just a computer-based process
of discerning patterns in sizable sets of data
• Especially with AFC and APC, opportunities abound to
utilize data mining of data flows to analyze operations,
passenger behavior, etc.
• Aim: improve services and performance
Cloud Computing
• Computing resources (hardware/ software) available over a network
(e.g., online — Email or blogging software)
• As Big Data and Analytics needs grow, Cloud is a means of getting
access to more powerful outside-vendor resources
• Apache Hadoop — NOT a database! Open-source framework for
distributed processing needed for data-intensive tasks
Sentiment Analysis
• Form of data mining applied to textual, verbal information
(from Emails, social media, open forums, etc.)
• Attitudes, opinions, intentions, etc. in source materials can
be identified and extracted for processing and scrutiny
• Useful to gauge public attitudes toward transit agency's
services, attitudes regarding new plans or services, sentiment
on a political issue such as a ballot measure
Security issues
• Concern is magnified with Big Data (far more data to lose)
• Issue has escalated with recent "cyber-attacks" on data and
functioning of large institutions
Privacy Concerns
• Very hot-button issue for general public
• Potential to extract great volumes of Big Data from fare
transactions, passenger counts, surveillance of passengers
in trains and stations opens opportunity for abuse
• Seattle ORCA card issue — employers could track
employees’ use of transit passes!
Predictive Analytics
• Applies Analytics to current and historical Big Data to
develop predictions about future
• Used to evaluate future role of public transit systems,
forecast ridership, and suggest need for new transit systems
and facilities
• Further benefits include passenger flow modeling (a la
BART), other operational projections and simulations
Robotics
• Automated machines or devices that can replicate human
activity or behavior, including operating a vehicle
• Rail transit has led the way with ATO and now driverless
trains
Summing Up
Rail Public Transportation Leads!
• Legacy of analytical and data-based high technology going
back to 19th century
• Rail transit probably has one of the heaviest utilizations of
Analytics/Big Data among modern industries
• Continuing to lead with automated systems, robotics, and
other leading-edge transportation technologies
Lyndon Henry
Blog Columnist, Railway Age Magazine
Co-Principal, Urban Rail Today
Technical Consultant , Light Rail Now Project
512.441-3014
[email protected]