Scott Kelley Dr. Michael Kuby Arizona State University

Scott Kelley

Dr. Michael Kuby

Arizona State University

NSF Grant 1025313: Spatial Refueling Patterns of Drivers of Alternative-Fuel and Conventional Vehicles

ESRI Southwest User Group Annual Conference

November 16, 2011 Mesa, Arizona

Energy Geography and Spatial Modeling/GIS

Our Energy Future?

New Infrastructure

Early Patterns?

Source: Energy Information Administration

Previous Work: ◦ Sperling & Kitamura (1986 & 1987)

◦ Diesel vs. Gasoline

Range Anxiety ◦ Gasoline: Higher Energy Density, Greater Range

Honda Civic GX Source: Honda

What are the refueling patterns of CNG drivers in Southern California?

◦ How do they compare to gasoline drivers?

◦ How can this help inform future alternative fuel infrastructure decision-making?

Survey Data

Collection

Fleet vs. Consumer

Find companion

station

Source: City of Burbank, CA

Network Analyst

Street Data Source: ◦ TeleAtlas data with ArcGIS 10

◦ Digitized and connected interstates

Goals: ◦ Shortest Path vs. Refueling Path

Primary Cost: Travel Time

Secondary Cost: Distance

◦ Closest Facility Analysis for CNG

◦ Service Areas

◦ Home locations important!

Build & calibrate

Digitize

Calculate

Automate

Export for Analysis ◦ Shortest Path

◦ Station Path

Review routes

QA/QC

Export to gdb

Immediate Route and Full Route ◦ Kitamura & Sperling used immediate

◦ Kept separate

Actual Trip Length, in Miles Time Deviation, in Minutes (%)

Mean Median Mean Median

Santa Monica

Gasoline 9.9 5.4 1.56 (12.2%) 1.35 (6.8%)

CNG 17.8 10.3 8.20 (101.2%) 6.39 (32.1%)

Anaheim

Gasoline 11.4 10.4 2.58 (13.4%) 1.43 (5.3%)

CNG 18.8 16.4 8.45 (23.7%) 5.39 (14.9%)

Burbank

Gasoline 54.0 12.3 2.76 (15.9%) 2.00 (6.53%)

CNG 44.1 21.0 8.02 (28.9%) 5.55 (21.4%)

* Single-purpose refueling trips (same origin and destination) removed

Burbank CNG Station - Consumer Stops Matrix (Origin - Destination)

DESTINATION

Home Work Social/Dining Shopping School Other TOTAL PERCENT

Home 0 6 3 3 2 2 16 30.19%

Work 19 1 1 1 0 0 22 41.51%

ORIGIN Social/Dining 6 0 0 0 0 0 6 11.32%

Shopping 2 0 1 0 0 0 3 5.66%

School 4 0 0 0 0 1 5 9.43%

Other 1 0 0 0 0 0 1 1.89%

TOTAL 32 7 5 4 2 3 53 PERCENT 60.38% 13.21% 9.43% 7.55% 3.77% 5.66%

Burbank GAS Station - Consumer Stops Matrix (Origin - Destination)

DESTINATION

Home Work Social/Dining Shopping School Other TOTAL PERCENT

Home 0 5 11 1 0 0 17 32.08%

Work 15 1 3 2 0 0 21 39.62%

ORIGIN Social/Dining 5 1 0 0 0 0 6 11.32%

Shopping 3 0 0 0 0 0 3 5.66%

School 6 0 0 0 0 0 6 11.32%

Other 0 0 0 0 0 0 0 0.00%

TOTAL 29 7 14 3 0 0 53

PERCENT 54.72% 13.21% 26.42% 5.66% 0.00% 0.00%

Closest Facility Analysis ◦ Data Source: NREL station data/Clean Car Maps

Calculate Route with Closest Facility Station

Closest Facility - % using closest to home

◦ Anaheim – 0%

◦ Santa Monica – 46.67%

◦ Burbank - 28%

Perceptions?

Traffic Patterns?

Cited:

“Convenient

Location”

Beyond GIS

◦ Statistical Analysis

Comparing populations of Gasoline/CNG

Near-home refueling for CNG or on-route?

Flow Recapturing Location Model (FRLM) - Kuby

◦ Patterns

Differences in geography?

Demographic differences?

Survey Drivers and Analyze Refueling Patterns

RQ: How do EV drivers use the new Phoenix infrastructure, and how can these results assist other cities in building their own networks?

Source: Blink Network

Ecotality/DOE network ◦ Residential

◦ Public

A different approach ◦ Level 1,2 chargers

◦ DC Fast-chargers

Stationary vs. quick-stop

Thanks to: ◦ PI: Dr. Michael Kuby

◦ School of Geographical Sciences and Urban Planning, Arizona State University

Joe Schoenemann, ASU - Digitizing

Patrick Zweifel, Jeff Martinez, ASU – Survey Collection

Mike McLean, ASU-ISSR

Clean Energy and gas station owners