Materials developed by K. Watkins, J. LaMondia and C. Brakewood
Data Types & Sources
Unit 2: Describing Transit Systems with Data
Materials developed by K. Watkins, J. LaMondia and C. Brakewood
Outline
• Types of transit data
• Manual data collection
• Automated data collection
Materials developed by K. Watkins, J. LaMondia and C. Brakewood
TYPES OF TRANSIT DATAWe need to quantify and characterize our systems with different
Materials developed by K. Watkins, J. LaMondia and C. Brakewood
Service Is Only As Good As Your Data
• We need accurate data to make appropriate decisions for transit service.
• What if we were considering adding a new line to a transit system?– What information would we need to know to
make appropriate design decisions?
Materials developed by K. Watkins, J. LaMondia and C. Brakewood
What Data is Needed to Maintain Good Service?
7 Demands of Useful Service
How Transit Services Them
It takes me where I want to
go.
It takes me when I
want to go.
It is a good use of my
time.
It is a good use of my money.
It respects
me.
I can trust it.
It gives me freedom
(to change my plans).
Stops/ Stations
Connectivity
Frequency Span Speed or Delay Fares Civility Reliability
Simplicity /Presentation
Materials developed by K. Watkins, J. LaMondia and C. Brakewood
Three Main Types of Data
• System Level Data– Entire operational overview – Useful for long-range planning, comparing service to similar
regions
• Route Level Data– Characteristics of specific routes – Useful for long-range planning, service planning, route
optimization
• Trip Level Data– Specific details of trips made on transit system– Useful for service planning, route optimization
Materials developed by K. Watkins, J. LaMondia and C. Brakewood
Service Quality Improvement Cycles
Archive Data
Analyze Performance and Demand
Service Plan
Transit Operation
Automated Data
Gathering
Operational Control &
Passenger Info
Real-time loop
Off-line loop
Materials developed by K. Watkins, J. LaMondia and C. Brakewood
System Level: National Transit Database (NTD)
• Established by Congress (Title 49 USC 5335a)• Primary source for statistics on transit in US• FTA grantees required to submit data• >660 providers currently report to NTD • Data used to apportion > $5 billion FTA funds• Freely and publically available
Materials developed by K. Watkins, J. LaMondia and C. Brakewood
2011 National Transit Profile
Does this look familiar?
Materials developed by K. Watkins, J. LaMondia and C. Brakewood
What is in the NTD?
• Modules include:– Safety and Security – Financials• Capital funds, operating expenses, operator wages
– Assets• Stations and maintenance facilities, transit way
mileage, revenue vehicles
– Resources• Number of employees, energy consumption
– Transit Service (Supplied and Consumed)• Vehicle revenue miles, Vehicle revenue hours, unlinked
passenger trips, passenger miles traveled
Materials developed by K. Watkins, J. LaMondia and C. Brakewood
NTD Glossary (Expenses)
• Capital Expenses = purchase of equipment
– Useful life > one year and Greater than $5,000
– Includes Facilities (Guideway, Passenger Stations,
Administration Buildings, Maintenance), Rolling Stock
(vehicles), Other Equipment
• Operating Expenses = operation of the agency
– Function (Vehicle Operations, Vehicle Maintenance, Non-
Vehicle Maintenance, General Administration)
– Object Class (Salary&Wages+Fringe= Compensation, Services ,
Materials and Supplies, Utilities, Casualty and Liability,
Purchased Transportation, Other)
Materials developed by K. Watkins, J. LaMondia and C. Brakewood
NTD Glossary (Service Consumed)
• Unlinked Passenger Trips = boardings
• Passenger Miles = cumulative sum of the distances ridden by each
passenger
• Average Trip Length = avg distance ridden for = passenger miles /
unlinked passenger trips
• Average Passenger Load = avg # pass aboard a vehicle at any one
time
Materials developed by K. Watkins, J. LaMondia and C. Brakewood
NTD Glossary (Service Supplied)
• Average Speed = miles / hours in revenue service
• Revenue Service = operation when passengers can board and ride
on the vehicle
• Vehicle Total Miles = all miles from pull out to pull in, including
"deadhead"
• Vehicle Revenue Miles = miles in revenue service
• Vehicle Total Hours = hours from pull out to pull in, including
"deadhead"
• Vehicle Revenue Hours = hours in revenue service
Materials developed by K. Watkins, J. LaMondia and C. Brakewood
NTD Glossary (Service Supplied)
• Revenue Vehicle = vehicle in fleet available to operate in revenue
service including spares and out for maintenance
• Vehicles Available for Maximum Service = vehicles agency has
available to operate revenue
• Vehicles Operated Maximum Service = largest # vehicles operated
at any one time
• Base Period Requirement = vehicles needed to serve the base (all-
day) transit service
• Peak-to-Base Ratio = Max Service / Base Period
• Percent Spares = (Veh Available – Veh Operated) / Veh Operated
Materials developed by K. Watkins, J. LaMondia and C. Brakewood
NTD Performance Measures
• Service Efficiency– Operating Expense per Vehicle Revenue Mile– Operating Expense per Vehicle Revenue Hour
• Service Effectiveness– Operating Expense per Passenger Mile– Operating Expense per Unlinked Passenger Trip
• Service Effectiveness– Unlinked Passenger Trips per Vehicle Revenue Mile– Unlinked Passenger Trips per Vehicle Revenue Hour
Materials developed by K. Watkins, J. LaMondia and C. Brakewood
IN-CLASS EXERCISEHead to www.ntdprogram.gov to complete the
Materials developed by K. Watkins, J. LaMondia and C. Brakewood
Route and Trip Levels Are Similar
• We describe unlinked passenger trips with– Route Stop Locations– Route Scheduling & Efficiency– Volumes of Passengers– Access/ Egress Locations– Travel and Boarding Times– Trip Purposes
• Two ways to collect this detailed data– Manually– Automatically
Exact Characteristics
Depend on Application
Materials developed by K. Watkins, J. LaMondia and C. Brakewood
MANUAL DATA COLLECTION
Engineers can tailor route and trip analyses through
Materials developed by K. Watkins, J. LaMondia and C. Brakewood
Manual Procedure Considerations
• Instruments: – Pencil/paper – Hand-held units
• Route Selection: – 100 percent– Sample
• Type of “Checks”: – Ride check: on-board the vehicle– Point checks: at a specific location
• Surveys also used (at trip-level)
Materials developed by K. Watkins, J. LaMondia and C. Brakewood
Ride Checks
• Captures an entire route in detail• Data collection is done on-board• Checkers counts the number of boardings,
alightings, & thru-passengers at each stop• Distance between each stop may also be
recorded
• Output: Route-level & Stop-level ridership data
Materials developed by K. Watkins, J. LaMondia and C. Brakewood
Ride Check Data Sheet - Example
Materials developed by K. Watkins, J. LaMondia and C. Brakewood
Ride Check - Basic Calculations
1. Total Unlinked Passenger Trips on the Route = Sum of all Boardings
2. Load Profile for the Route = Cumulative Boardings @ Stop – Cumulative Alightings @ Stop
Peak (maximum) load along the route Average load along the route
3. Passenger Miles Traveled on the Route = Sum of Stop-Level Load * Distance between Stops
Materials developed by K. Watkins, J. LaMondia and C. Brakewood
Example Load Profile
Stop #1
Stop #1
Stop #1
Stop #1
Stop #1
Stop #1
Stop #1
Stop #1
Stop #1
Stop #1
0
Stop #1
0
Stop #1
0
Stop #1
0
Stop #1
0
Stop #1
0
Stop #1
0
Stop #1
0
Stop #2
0
Stop #2
0
Stop #2
0
Stop #2
0
Stop #2
0
Stop #2
0
Stop #2
0
Stop #2
0
Stop #2
0
Stop #3
0
Stop #3
0
Stop #3
0
Stop #3
00
5
10
15
20
25
30
35
40
45
50
Bus Route Load Profile
Load per StopAverage Load
Pass
enge
r Loa
d
Peak Load Point: 47 Pax
Average Load: 24 Pax
Materials developed by K. Watkins, J. LaMondia and C. Brakewood
Point Checks
• Data collection is done at one (or more) stops along a route
• Number of passengers on the vehicle is recorded
• Often conducted at high (peak) load points along a route
Materials developed by K. Watkins, J. LaMondia and C. Brakewood
Surveys
• Travel surveys are used to estimate complete origin-destination patterns
• Questions include boarding location, alighting location, transfer location
• More next lecture
Materials developed by K. Watkins, J. LaMondia and C. Brakewood
AUTOMATED DATA COLLECTION
Transit operators are increasingly relying on new technologies…
Materials developed by K. Watkins, J. LaMondia and C. Brakewood
Automated Data Sources
1. Automated Fare Collection (AFC)2. Automated Passenger Counters (APC)3. Automated Vehicle Location (AVL)
• Archived data from AFC and AVL systems is an important byproduct of installing these systems.
Materials developed by K. Watkins, J. LaMondia and C. Brakewood
Automated Fare Collection (AFC)
• Magnetic stripe & smart cards• Smart card systems have unique
ID that provides entry (exit) information at the individual level – Data not available in real-time
(coming soon!)• Magnetic stripe and smart cards
together can produce station/stop level data– Rail gateline counts– Bus fare boxes (operator often
punches a key indicating other fare types, such as free passes)
Materials developed by K. Watkins, J. LaMondia and C. Brakewood
Smart Card Example: OD Estimation
• AM Inbound trip: tap-in at “origin”
• PM Outbound trip: tap-in at “destination”
• Infer that passenger is traveling from Brookhaven to Midtown and back
AM Tap at Brookhaven
PM Tap at Midtown
Materials developed by K. Watkins, J. LaMondia and C. Brakewood
Automated Passenger Counters (APC)
• Sensors near vehicle doors count passenger on and offs, usually using infrared beams
• Some vehicles also determine weight/steps on-board vehicle
• Typically data not available in real-time • Usually only on a sample (%) of bus fleet;
redistribute buses to determine counts
Materials developed by K. Watkins, J. LaMondia and C. Brakewood
Automated Vehicle Location (AVL)
• Originally, Signpost-beacon-based used to track the location of buses
• Post-2000, GPS-based technology• Provided in real-time• Usually matched with schedule through Computer
Aided Dispatch (CAD) systems• Also combined with other data sources (AFC & APC)
to capture detailed ridership• Valuable for service planning (i.e. determining
running times, schedule adherence)
Materials developed by K. Watkins, J. LaMondia and C. Brakewood
Signpost-based AVL
Materials developed by K. Watkins, J. LaMondia and C. Brakewood
GPS-based AVL
Materials developed by K. Watkins, J. LaMondia and C. Brakewood
Combining AVL & APC
• Detailed Route Level Analysis
Materials developed by K. Watkins, J. LaMondia and C. Brakewood
DATA COLLECTION COMPARISON
Materials developed by K. Watkins, J. LaMondia and C. Brakewood
Manual vs. Automated Data
Manual Data• Low capital costs• High marginal (labor) costs• Small sample sizes– Limited spatial and
temporal variation • Often unreliable (errors by
checkers)• Longer data
collection/processing times
Automated Data
• High capital costs• Low marginal (labor) costs• Large sample sizes
– Detailed temporal and spatial data
• Biases can (usually) be corrected• Available in (quasi) real-time
Materials developed by K. Watkins, J. LaMondia and C. Brakewood
Passenger Counting Technologies (2008)
Technology / Procedure # Systems PercentageCombination of manual and automatic 44 51.2%Manual (paper and pencil) only 18 20.9%APCs only 12 14.0%Other automated methods (farebox, hand-held units)
12 14.0%
Total systems 33 100.0%
Materials developed by K. Watkins, J. LaMondia and C. Brakewood
Future Trends
• Increasing use of automated data collection systems– Including mixed modes (manual &
automated)
• More disaggregate data used for planning and decision-making (refined units of analysis)
Materials developed by K. Watkins, J. LaMondia and C. Brakewood
DATA STANDARDS
Materials developed by K. Watkins, J. LaMondia and C. Brakewood
Transit Data Consumption
The changing landscape
Schedule
Paper Schedules
109:36
Digitization Interactivity
Materials developed by K. Watkins, J. LaMondia and C. Brakewood
GTFS
General Transit Feed Specification
routes.txt
stops.txt
trips.txt
stop_times.txt calendar.txt
agency.txt
shapes.txt
Materials developed by K. Watkins, J. LaMondia and C. Brakewood
How Does Open Data Help?
Data access models
Agency responds to special requests by
developers
Small subset of riders find this specific tool useful.
Transit Agency
App Developers
Riders
DATA
DATAAnyone can access data
Many riders access a diverse market of tools powered by GTFS.
Agency produces data and opens it
once.
Materials developed by K. Watkins, J. LaMondia and C. Brakewood
Transit Open Data Timeline
Source: Rojas, Francisca (2012) Transit Transparency: Effective Disclosure through Open Data
Materials developed by K. Watkins, J. LaMondia and C. Brakewood44
Open Schedule Data (GTFS) Adoption
Source: Wong, James. (2013). Leveraging the General Transit Feed Specification (GTFS) for Efficient Transit Analysis. Proceedings of the 2013 Transportation Research Board Annual Meeting.
Materials developed by K. Watkins, J. LaMondia and C. Brakewood
Conclusions
• We need accurate data to make appropriate decisions for transit service.
• The National Transit Database (NTD) is the largest source of transit data in the US.
• Manual data collection consists in recording transit information in person.
• Automatic data collection uses captors in buses to record vehicle location, vehicle loads, etc.
• Future trend of opening up data based on standardized formats.
Materials developed by K. Watkins, J. LaMondia and C. Brakewood
Reference
Materials in this lecture were taken from:• Walker, J. (2011). Human transit: How clearer thinking about
public transit can enrich our communities and our lives. Island Press.
• Furth, Hemily, Muller, Strathman (2006). “Using Archived AVL-APC Data to Improve Transit Performance and Management. TCRP Report 113." Transportation Research Board.
• National Transit Database Sampling Manual (2009)• "Sampling Tests Automatic Passenger Counters." The Inside
Lane. 26 Sept. 2011.• Boyle, D (2008). "TCRP Synthesis 77: Passenger Counting
Systems.” TRB, National Research Council, Washington, D. C.
Top Related