using mobile apps to measure spatial travel behavior changes of carsharing users sisinnio concas, s....

13
Center for Urban Transportation Research Using Mobile Apps to Measure Spatial Travel Behavior Changes of Carsharing Users Sisinnio Concas, S. Barbeau, P.L. Winters, N.L. Georggi, J.M. Bond Center for Urban Transportation Research Session 765 Technology and Travel Surveys

Upload: clara-worcester

Post on 31-Mar-2015

216 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Using Mobile Apps to Measure Spatial Travel Behavior Changes of Carsharing Users Sisinnio Concas, S. Barbeau, P.L. Winters, N.L. Georggi, J.M. Bond Center

Using Mobile Apps to Measure Spatial Travel Behavior

Changes of Carsharing Users

Sisinnio Concas, S. Barbeau, P.L. Winters, N.L. Georggi, J.M. Bond

Center for Urban Transportation Research

Session 765

Technology and Travel Surveys

Page 2: Using Mobile Apps to Measure Spatial Travel Behavior Changes of Carsharing Users Sisinnio Concas, S. Barbeau, P.L. Winters, N.L. Georggi, J.M. Bond Center

Center for Urban Transportation Research, University of South Florida

What is Carsharing?

• Short-term auto rental option• Hourly flat rate• Annual membership fee• No gasoline costs • Unlimited mileage• No parking costs

Page 3: Using Mobile Apps to Measure Spatial Travel Behavior Changes of Carsharing Users Sisinnio Concas, S. Barbeau, P.L. Winters, N.L. Georggi, J.M. Bond Center

Center for Urban Transportation Research, University of South Florida

Carsharing Program

• Launched at the University of South Florida, Tampa Campus, on July 2009

• Short-term rental option with hourly flat rate– Annual membership fee– Unlimited mileage– Non parking costs– No gasoline costs

• Four hybrid vehicles available 24/7

Page 4: Using Mobile Apps to Measure Spatial Travel Behavior Changes of Carsharing Users Sisinnio Concas, S. Barbeau, P.L. Winters, N.L. Georggi, J.M. Bond Center

Center for Urban Transportation Research, University of South Florida

Program Participation

Sep-

09

Nov-0

9

Jan-

10

Mar-1

0

May-1

0

Jul-1

0

Sep-

10

Nov-1

0

Jan-

11

Mar-1

1

May-1

1

Jul-1

1

Sep-

110

20

40

60

80

100

120

140

160

180

0

500

1000

1500

2000

2500

Rentals by Month

RentalsCumul Rentals

Num

ber

of

Renta

ls

Cum

ula

tive N

um

ber

of

Renta

ls

Do not own a car/truckDo not want to depend on others for

transportationSave money

Protect the environment

Did not bring my car/truck to campus

0% 20%

40%

60%

80%

Why did you join WeCar?(n=34)

Shopping/errands

Visit friends/relatives

Visit public place: historical site/museum/park/library

Social event

Family personal business/obligations

Attend business meeting/trip

Buy services: video rentals/dry cleaner/post office/car service/bank

Coffee/ice cream/snacks

Pet care: walk the dog/vet visits

Day care

Go to gym/exercise/play sports

Where do you go with WeCar?(n=34)

Page 5: Using Mobile Apps to Measure Spatial Travel Behavior Changes of Carsharing Users Sisinnio Concas, S. Barbeau, P.L. Winters, N.L. Georggi, J.M. Bond Center

Tracking Spatial Behavior

TRAC-IT GIS Database

ServerData Storage Tier

GIS Web or Desktop

ApplicationsClient Tier

TRAC-ITParticipant

Raw GPS Data

Personalized Email withlink to Data

TRAC-IT Toolkit Software

TRAC-IT Web Server

Hosts files for download

Formatted GPS Data(KML, GPX, ICS, etc.)

1)

2)

3)

4) User downloads KML file from server

Date Range 2/10/2011 to 4/29/2011

Total Number of Users 30

Total Number of Sessions 1,857

Avg. Session Length (hrs) 15.44

Total Survey Time (days)* 1,194.80

Avg. Survey Time per User (days) 39.83

Total Number of GPS fixes Received 4,023,917

Avg. Number of GPS fixes per Session 2,166.89

Avg. Number of GPS fixes per User 134,130.57

“This is my trip to campus, via the Bull Runner, to pick up the WeCar. I then drove the WeCar to the CVS on Fowler to pick up medication and then drove to the grocery store on Bears Ave. After shopping, I dropped the groceries off at home and then drove back to campus to return the WeCar. I then took the Bull Runner back home.”

Page 6: Using Mobile Apps to Measure Spatial Travel Behavior Changes of Carsharing Users Sisinnio Concas, S. Barbeau, P.L. Winters, N.L. Georggi, J.M. Bond Center

Communicating with Users

TRAC-IT GIS Data StorageDatabase Server

TRAC-ITMobile

ApplicationCell Phone

TRAC-ITSurvey

Participant

Web or Desktop ApplicationsData Viewing/

Analysis

TRAC-ITParticipant

LEGEND

UDP

JDBC

SMTPHTTP(S)

Application Data

Location DataLocation Data

Application Data

TRAC-IT Database ToolkitAutomated/Manual

Batch Data Processing

TRAC-IT Web Application

Web Services

TRAC-IT File

StorageWeb Server

TRAC-ITData Analyst

TRAC-IT Emails

Email Server

POP3/IMAP/Exchange

FTP

TRAC-ITAdministrator /Data Analyst

SOAP

Toolkit Monitors Web Application

Application & Location Data Processed

Data

Personalized Emails

KML, GPX, ICS files

PersonalizedEmails

KML, GPX, ICS files

User Feedback

File URL

TCP

Adaptive Location Data Buffering

Page 7: Using Mobile Apps to Measure Spatial Travel Behavior Changes of Carsharing Users Sisinnio Concas, S. Barbeau, P.L. Winters, N.L. Georggi, J.M. Bond Center

Center for Urban Transportation Research, University of South Florida

Trips by Mode and Purpose

Mode Count Percent

Car 666 40.8%

Bicycle 155 9.5%

Bus 132 8.1%

Scooter 18 1.1%

Walking 410 25.1%

Carsharing 31 1.9%

Not Reported

221 13.5%

Total 1,633 100.0%

84; 5%417; 26%

303; 19%120; 7%

93; 6%

110; 7%

211; 13%

28; 2%

267; 16%Dining Out

Return Home

School / Religious Activity

Shopping / Errands

Social / Recreational

Transport Someone

Work or Work-Related

Other

Not Reported

Page 8: Using Mobile Apps to Measure Spatial Travel Behavior Changes of Carsharing Users Sisinnio Concas, S. Barbeau, P.L. Winters, N.L. Georggi, J.M. Bond Center

Center for Urban Transportation Research, University of South Florida

Communicating Pricing Changes

• Incentives to encourage off-peak travel• 50-percent reduction in hourly rates• Rate changes communicated by email to

treatment:– Scheduling constraints accounted for– Discount applied on actual rental starting time

Implementation Schedule Starting Date Details

Pricing Strategy A March 1 to March 21, 2011 USF to communicate schedule to carsharing on 02/21/2011

Pricing Strategy B March 28 to April 28, 2011 USF to communicate schedule to carsharing on 03/21/2011

Page 9: Using Mobile Apps to Measure Spatial Travel Behavior Changes of Carsharing Users Sisinnio Concas, S. Barbeau, P.L. Winters, N.L. Georggi, J.M. Bond Center

Center for Urban Transportation Research, University of South Florida

Measuring Spatial Patterns of Activity-Travel

  Trip Length (miles) SDE (square miles) User Type Average Carsharing

TripNon

CarsharingAverage Carsharing

TripNon-Carsharing

TripCarsharing 2.6 8.0 1.7 0.5 0.2 0.5Non-Carsharing 4.2 - 4.2 7.8 - 7.8

Activities Mean Center

SDC

Minor Axis

Major Axis

SDE

𝑥

Y

X

Y

X

Page 10: Using Mobile Apps to Measure Spatial Travel Behavior Changes of Carsharing Users Sisinnio Concas, S. Barbeau, P.L. Winters, N.L. Georggi, J.M. Bond Center

Center for Urban Transportation Research, University of South Florida

Rental Cost and Rental Time

Before Variable Pricing

After Variable Pricing

Page 11: Using Mobile Apps to Measure Spatial Travel Behavior Changes of Carsharing Users Sisinnio Concas, S. Barbeau, P.L. Winters, N.L. Georggi, J.M. Bond Center

Center for Urban Transportation Research, University of South Florida

Discussion (1)

• User interaction and real-time data monitoring contributed to continued participation throughout the study.

• Participants responded positively to this interaction and expressed a sense of confidence in the GPS tracking tool.

• Data collection allowed studying users’ travel behavior responses to variable pricing experiments.

Page 12: Using Mobile Apps to Measure Spatial Travel Behavior Changes of Carsharing Users Sisinnio Concas, S. Barbeau, P.L. Winters, N.L. Georggi, J.M. Bond Center

Center for Urban Transportation Research, University of South Florida

Discussion (2)

• The research examined changes in travel behavior through pricing mechanisms based on time-of-day and day-of-week hourly rate discounts.

• The study found that overall carsharing users are price sensitive

• Changes in rental rates of the order of 20 to 50 percent have a significantly large impact on daily rentals.

• The analysis shows that carsharing users’ activity space contracts while using carsharing as a mode of transport.

Page 13: Using Mobile Apps to Measure Spatial Travel Behavior Changes of Carsharing Users Sisinnio Concas, S. Barbeau, P.L. Winters, N.L. Georggi, J.M. Bond Center

Center for Urban Transportation Research

Sisinnio Concas, PhD

(813) 974-7760

[email protected]

Center for Urban Transportation Research

Contact