Transcript Presentation - 15th TRB National Transportation Planning

0
GIS Applications for OriginDestination Surveys
Greg Spitz and John Lobb
Resource Systems Group
[email protected]
[email protected]
Daniel Jacobs
MTA Bridges & Tunnels
Wayne Bennion
Wasatch Front Regional Council
1
Origin-Destination (OD) Data

Critical basis of knowledge for transportation
operations, modeling, and future planning
 Essential for all transportation modes
 Often collected by DOTs and transit agencies,
but not (yet) often analyzed with GIS
 OD data are inherently spatial, and
visualization is key to understand them, thus
the need for GIS
2
Origin & Destination Surveys

Provide insights to:
 Who the customers are
 Why customers are traveling
 How long they are traveling, and
 Where they are traveling, etc.
 Labor intensive data collection efforts
 Used to understand current demand and
plan for future demand
 Critical data for calibrating travel models
3
GIS Tools in action

MTA Bridges and Tunnels— largest toll
agency in US
 Wasatch Front Regional Council &Utah
Transit Authority— transit on-board survey
 Two different agencies, modes, and
application purposes
 GIS the common denominator to solve
issues for these different purposes
4
2004 MTA B&T Origin & Destination
Survey

Conducted roughly every 8-10 years
 2004 Study:

304,000 surveys distributed in CASH lanes
 329,000 surveys mailed to E-ZPass customers
 Control Data collected in both CASH and EZPass lanes with Pocket-PCs
 E-ZPass sample from B&T, PANYNJ, NYSTA,
NYSBA, and NJRSC comprising 99% of all
transactions
 Survey period 6AM to Midnight; Weekday,
Saturday, and Sunday
5
MTA B&T Survey TAZ Structure

All zones are based on zip codes


Geocoding of survey data only necessary
to the zip code level
Aggregations of zones used to form
“super zones”
6
GIS using static maps

Static analysis with better presentation
 Easy to put together quickly
 Becoming more and more typical
7
Survey Zone Map: Close-up
DRAFT
8
Super Zone Map
9
Example of Static Map using super zones:
Bronx-Whitestone Bridge – Bronx Bound
•All O-D pairs with greater than
4% of traffic
10.2%
•All payment types
•Weekday
•Truck/taxi trips excluded
10.2%
19.1%
•6am to Midnight
4.5%
•Total Auto Traffic (6AM to
Midnight) is 47,030
20.2%
4.1%
•95% Confidence interval for
these OD pairs is plus or minus
2.0%
Order
1
2
3
4
5
6
Origin Super
Zone
Queens
Queens
LI
Queens
LI
Queens
Destination
Super Zone
Bronx
Westchester
Westchester
Connecticut
Bronx
NJ
Adjusted Volume
9,520
8,989
4,817
4,816
2,096
1,914
% of Total
Traffic
20.2%
19.1%
10.2%
10.2%
4.5%
4.1%
10
GIS Tool Purpose

Take geospatial data and make it easier to
visualize, analyze, and interpret

Allow more in depth analyses beyond static reports

Make the tool easy to use, so even a (skilled)
monkey can use it

Leverage GIS experts’ skills without taking too
much of their time
11
How was the tool made?

ArcMap extension
 Access DB backend
 Currently developed to be a stand alone tool
(could be made to be network/web enabled)
 Simple user interface created using VB
development environment
12
Income Distributions by Origin Zones
13
Vehicle Occupancy by Origin Zone
14
Specific Trip Analysis
15
Travel Pattern Analysis
16
2006 WFRC On-Board Transit Survey


3 surveys conducted in last 15
years
Data primarily used for travel
model development and
forecast refinement

2006 survey


Surveyed riders on 90 bus routes
and the TRAX rail system
Collected 5,600 surveys
17
OD Data Mining is Cumbersome

Survey database has
over 100 columns and
5600 rows (difficult to
view all relevant
variables at once)

TAZ number and/or
address text may not
be immediately
recognizable (data is
spatial)
18
GIS Tool Purpose (WFRC)

Preliminary review of
the data suggested that
the 2006 survey
transfer rates were
unreasonable
Interest in examining
individual survey
records efficiently
Frequency Distribution of Transit Transfers
(unweighted sample)
70
2006
2002
60
50
% of Trips

40
30
20
10
0
0
1
2
3
Transfers
4
5
19
GIS tool (WFRC)
Goals
Features

Local QA/QC

Custom ArcGIS application

Efficiently examine each
record

Visualize transit path details
for one record at a time

Easy data editing/entry

Routes

Gain confidence in the data
used for model calibration
(e.g. transfer rates)

Origin/Destination

Boarding/Alighting

Data viewing/editing window
20
Screen capture of GIS Tool
Reported data
21
Survey Errors Identified
1.
2.
3.
4.
People reported
multiple paths
they sometimes
take, rather than
simply their
current path
(question 7)
Inconsistency
between routes
and OD pair
Geocoding errors
Illogical route
sequence
22
Example of Multiple Paths
between same OD Pair
23
Another example of multiple
paths between same OD pair
24
Summary

Application clearly
identified obvious
inconsistencies and
errors, resulting in a
more reliable database.
Visualization of these
records would also be
extremely helpful in
assessing survey
design and reducing
respondent error.
Frequency Distribution of Transit Transfers
(unweighted sample)
2006 Before
2006 After
2002
70
60
50
% of Trips

40
30
20
10
0
0
1
2
3
Transfers
4
5
25
User-friendly GIS tools developed for
planners and analysts

GIS tools allow planners and others to drill
down on the data
 Tool interface confines the problem (both a
good and bad thing)
 Interface makes analysis much easier and
doesn’t require in-depth software/data
knowledge
 Tools allow analysts to get more out of
their data collection investment
26
Questions ?

Please use the Microphone.