Clean M Miles Standar ard W Workshop 2018 B Base Y Year - - PowerPoint PPT Presentation

▶
clean m miles standar ard w workshop 2018 b base y year
SMART_READER_LITE
LIVE PREVIEW

Clean M Miles Standar ard W Workshop 2018 B Base Y Year - - PowerPoint PPT Presentation

Jan 09, 2023 322 likes •789 views

Clean M Miles Standar ard W Workshop 2018 B Base Y Year Emission ons I Inventor ory September 25, 2019 1 Purpose Establish 2018 base year emissions inventory Discuss data sources, methodology and assumptions Present

slide-1
SLIDE 1

Clean M Miles Standar ard W Workshop 2018 B Base Y Year Emission

  • ns I

Inventor

  • ry

September 25, 2019

1

slide-2
SLIDE 2

Purpose

  • Establish 2018 base year emissions inventory
  • Discuss data sources, methodology and assumptions
  • Present preliminary results
  • Solicit feedback on the assumptions

2

slide-3
SLIDE 3

Imp mpor

  • rtan

ant N Note

  • The data analysis and results presented in this workshop

are pr prelim limin inary ry

  • The analysis is conducted using March 2

2019 019 data submittals

  • Staff has worked closely with TNCs since March to QA/QC

the data – updates were incorporated

  • Preliminary results presented in this workshop may not

reflect all of the latest updates

3

slide-4
SLIDE 4

Outlin line

  • Introduction to SB 1014
  • Base year inventory methodology
  • Data description
  • Assumptions (overlap removal/occupancy/fuel efficiency)
  • Preliminary Results
  • Next Steps

4

slide-5
SLIDE 5

Clea ean Miles S es Standard

5

(Int ntroduc duction) n) (Methodology) (Data Description) (Overlap Removal) (Occupancy) (Fuel Efficiency) (Preliminary Results)

slide-6
SLIDE 6

Transportation N Network C Companies

  • California Public Utilities Commission (CPUC) Definition: “A

Transportation Network Company (TNC) is a company or

  • rganization operating in California that provides transportation

services using an online-enabled platform to connect passengers with drivers using their personal vehicles.”

  • In California, the CPUC oversees regulation and permitting of TNCs

such as charter-party carriers. There may up to 14 different companies providing services

6

(Int ntroduc duction) n) (Methodology) (Data Description) (Overlap Removal) (Occupancy) (Fuel Efficiency) (Preliminary Results)

slide-7
SLIDE 7

Princi ciples o

  • f the New R

Regulation

  • TNCs can facilitate state/local governments in meeting emissions reduction

goals

  • Decrease GHG emissions using a compliance metric of annual grams-CO2-

per-passenger-mile traveled

  • Transportation decarbonization by increasing zero-emission vehicles
  • Increase passenger-mile-traveled by increased pooling, active transport

(walk/bike), and transit usage

  • Forward-looking with automated vehicles or new modes that may arise in

future

  • Aligned with other State policies
  • Maximize transportation access equity

7

(Int ntroduc duction) n) (Methodology) (Data Description) (Overlap Removal) (Occupancy) (Fuel Efficiency) (Preliminary Results)

slide-8
SLIDE 8

Base Y Year Emission

  • ns I

Inventor

  • ry
  • SB1014 requires CARB to establish a GHG base year (2018)

emissions for TNCs on a per passenger mile

  • Base year GHG emissions provides a reference point to

establish future emission targets and evaluate effectiveness of future compliance scenarios

8

  • Vehic

ehicle S le Speed peed

  • Vehic

ehicle O le Occup upanc ncy

  • Trip

ip Mi Miles les: :  Vehicle trip miles/Ride VMT  Transit miles  Active miles

  • Fuel e

l effic icie iency b by vehic icle le specif ific icatio ions Vehicle T Technol

  • logy
  • gy

Grams o

  • f CO2 p

2 per p pass ssen enger er-mile e depen epend o

  • n

Vehicle O e Oper eration

(Int ntroduc duction) n) (Methodology) (Data Description) (Overlap Removal) (Occupancy) (Fuel Efficiency) (Preliminary Results)

Affect PMT

slide-9
SLIDE 9

Period

  • ds D

Defined f for T TNC Miles es

9

(Int ntroduc duction) n) (Methodology) (Data Description) (Overlap Removal) (Occupancy) (Fuel Efficiency) (Preliminary Results)

slide-10
SLIDE 10

10

Grams CO2 PMT = VMT (miles) x Real World Fuel Consumption (gal/mi) x Conversion Factor (gCO2/gal) Ride VMT (miles) x Occupancy + Active/Transit PMT VMT in periods 1, 2, & 3 Only period 3 VMT Does not include driver Fuel Dependent Assumed - zero

  • Fuel e

el econo nomy Affects the numerator

  • nly
  • Increasing fuel economy reduces
  • TNC gCO2/PMT
  • CA Fleet GHGs
  • No affect on VMT
  • Deadhea

eading affects only the numerator

  • Decreasing deadhead VMT reduces
  • TNC gCO2/PMT
  • CA Fleet GHGs
  • CA Fleet VMT
  • Occupa

upancy affects only the denominator

  • Increasing occupancy reduces
  • TNC gCO2/PMT

Method

  • dolog
  • gy

(Introduction) (Methodology) (Data Description) (Overlap Removal) (Occupancy) (Fuel Efficiency) (Preliminary Results)

slide-11
SLIDE 11

Da Data De Description

  • CARB received approximately 1.4 billion trip records for

600k vehicles operating for TNCs

  • Data from 14 TNC companies received in March, 2019
  • Data issues: Faulty VINs; Self-overlapping trips

11

1. Unique trip ID 2. Unique driver ID 3. Vehicle ID number 4. Make and model name 5. Trip periods (P1,P2,P3) 6. Vehicle miles traveled(VMT) by period 7. Date/time for trips start and end 8. Zip codes 9. Latitude and longitude

  • 10. Average speeds
  • 11. Pooled or shared rides
  • 12. Surge period

Data F Field lds

(Introduction) (Methodology) (Data D Description) (Overlap Removal) (Occupancy) (Fuel Efficiency) (Preliminary Results)

slide-12
SLIDE 12

Some pr e prel eliminary statistics…

12

(Introduction) (Methodology) (Data D Description) (Overlap Removal) (Occupancy) (Fuel Efficiency) (Preliminary Results)

* * Total VMT/ Number of P3 Trips

TNC w C wide/Prop

  • por
  • rtion
  • n (

(%) California P Passen senger er V Vehi ehicles es (EMFAC2017) 2017) Total num umber er o

  • f vehi

vehicles es 600 thousand (2.3%) 25.2 million To Total V VMT 4.2 billion miles (1.2%) 342.3 billion miles Nu Number o r of t tri rips (Onl Only P P3 T Trips) s) 0.37 billion (0.9%) 41.4 billion Aver verage t e trip leng ength* 11.4 miles per trip 8.3 miles per trip Cars rs vs. T Tru ruck cks 79% Cars & 21% Trucks 63% Cars & 37% Trucks VMT W Wei eighted ed Ave verage e Model del Y Year 2010.5 2009

slide-13
SLIDE 13

13

Grams CO2 PMT =

VM VMT ( (miles) ) x Real World Fuel Consumption x Conversion Factor Ride VM VMT ( (miles) ) x Occupancy + Active/Transit PMT

Method

  • dolog
  • gy

Let’s start with VMT and Ride VMT

(Introduction) (Methodology) (Data Description) (Overlap Removal) (Occupancy) (Fuel Efficiency) (Preliminary Results)

  • Calculating actual VMT
  • Revising data to identify unique, real VMT for each vehicle
slide-14
SLIDE 14

Multi-Apping

  • A common practice of drivers being available for service on

multiple platforms at the same time.

  • To avoid double counting, instances of multi-apping should be

identified and removed accordingly (i.e., “combined”)

14

Company A Company B Combined

P1 P2 P3

?

(Introduction) (Methodology) (Data Description) (Overlap Removal) (Occupancy) (Fuel Efficiency) (Preliminary Results)

slide-15
SLIDE 15

Combined – Trip ip O Overla lap Rem emoval

“Stick Painti ting a g algo gorith thm”

  • Assume that time is a long stick that can be painted on
  • Assign a color to each time period, i.e. P1, P2, P3
  • Start with a VIN; for each trip record using the assigned trip period colors paint a segment on the stick that

corresponds to the beginning and end times of that trip record

  • Start with painting P1 trip periods on the stick first, then move on to P2 trip periods, and at last to P3
  • Reconstruct the trip records by identifying the color of each painted segments on the stick and converting

the beginning and end locations to beginning and end of the trip periods

15

Beginning of 2018 End of 2018 TNC trips Stick Brush Paint P1 P2 P3

(Introduction) (Methodology) (Data Description) (Overlap Removal) (Occupancy) (Fuel Efficiency) (Preliminary Results)

slide-16
SLIDE 16

Trip Ov Overlap ap R Removal al: Ex Example (On One V VIN wi with Multiple T TNC Apps)

16

Company A Company B Combined

P1 P2 P3

?

(Introduction) (Methodology) (Data Description) (Overlap Removal) (Occupancy) (Fuel Efficiency) (Preliminary Results)

slide-17
SLIDE 17

Trip Ov Overlap ap R Removal al: E Example (On One V VIN wi with Multiple T TNC Apps)

17

Combined

P1 P2 P3

Step ep 1: 1: Paint P P1 1 trips o s on n the he “Combined” s stick

Company A Company B

(Introduction) (Methodology) (Data Description) (Overlap Removal) (Occupancy) (Fuel Efficiency) (Preliminary Results)

slide-18
SLIDE 18

Trip Ov Overlap ap R Removal al: Ex Example (On One V VIN wi with Multiple T TNC Apps)

18

Combined

P1 P2 P3

Step ep 2: 2: Paint P P2 2 trips o s on n the he “Combined” s stick

Company A Company B

(Introduction) (Methodology) (Data Description) (Overlap Removal) (Occupancy) (Fuel Efficiency) (Preliminary Results)

slide-19
SLIDE 19

Trip Ov Overlap ap R Removal al: Ex Example (On One V VIN wi with Multiple T TNC Apps)

19

Combined

P1 P2 P3

Step ep 3: 3: Paint P P3 3 trips o s on n the he “Combined” s stick

Company A Company B

(Introduction) (Methodology) (Data Description) (Overlap Removal) (Occupancy) (Fuel Efficiency) (Preliminary Results)

slide-20
SLIDE 20

Trip Ov Overlap ap R Removal al: Ex Example (On One V VIN wi with Multiple T TNC Apps)

20

Combined

P1 P2 P3

Step ep 4: 4: Rec econs nstruct t the r he rec ecords

Company A Company B

(Introduction) (Methodology) (Data Description) (Overlap Removal) (Occupancy) (Fuel Efficiency) (Preliminary Results)

slide-21
SLIDE 21

21

Grams CO2 PMT = VMT (miles) x Real World Fuel Consumption x Conversion Factor Ride VMT (miles) x Occup

upancy + Active/Transit PMT

Method

  • dolog
  • gy

Occupancy is a key factor determining PMT

(Introduction) (Methodology) (Data Description) (Overlap Removal) (Occupa upanc ncy) (Fuel Efficiency) (Preliminary Results)

  • Occup

upanc ncy d data is not provide ded by d by TNCs

slide-22
SLIDE 22

Occupan ancy

  • CA fleet average occupancy (PMT/VMT) is 1.68, estimated using

data from the 2010 – 2012 California Household Travel Survey

  • Scientific studies have found the following on average occupancy

for TNCs:

a) Circella et.al 2019 CA average occupancies: 1. 1.90 90 (N = = 1, 1,287 287) b) Henao et.al 2018 Colorado average occupancy = 1. 1.34 34 (N = = 416) 416)

  • CARB has also conducted an in-house study to collect occupancy

information

22

Weekday = = 1. 1.69, 69, Weeknig ight = = 1. 1.93, 93,Weekend D Day = = 1. 1.95, 95, Weekend N Night = = 2. 2.16 16

(Introduction) (Methodology) (Data Description) (Overlap Removal) (Occupa upanc ncy) (Fuel Efficiency) (Preliminary Results)

slide-23
SLIDE 23

CARB B Occup upancy Projec ect

  • Goa
  • al:

l: Determine occupancy rates for pooled and non-pooled rides, and collect activity information to develop drive cycles

23

  • 2 week study period
  • TNC drivers maintain a daily trip diary
  • Data loggers record on-board vehicle data

(Introduction) (Methodology) (Data Description) (Overlap Removal) (Occupa upanc ncy) (Fuel Efficiency) (Preliminary Results)

slide-24
SLIDE 24

Trip D p Diary a and D d Data L Logger P Parameter ers ( (Exampl ples es)

24

Engine RPM Vehicle Speed Catalyst Temperature Fuel Consumption Engine Load Road Grade Altitude etc.

(Introduction) (Methodology) (Data Description) (Overlap Removal) (Occupa upanc ncy) (Fuel Efficiency) (Preliminary Results)

Date: Vehicle: Driver: Odometer: Event Code TNC Type Number of S: Start the App Passengers O: App turned Off AP: Pooled (not including W: Waiting AN: Non Pooled driver) R: Receive Call BP: Pooled P: Pick Up BN: Non Pooled D: Drop Off CP: Pooled CN: Non Pooled Circle one for each line: Circle one: Circle one: Time (AM / PM) Event TNC Type No of Passengers S O W R P D F AP AN BP BN CP CN 1 2 3 4 5 6 S O W R P D F AP AN BP BN CP CN 1 2 3 4 5 6 S O W R P D F AP AN BP BN CP CN 1 2 3 4 5 6 S O W R P D F AP AN BP BN CP CN 1 2 3 4 5 6 S O W R P D F AP AN BP BN CP CN 1 2 3 4 5 6

Daily Trip Diary

slide-25
SLIDE 25

Preliminary Occupancy A Analysis

  • 22 trip diaries have been returned (almost 2,000 fares)
  • Staff expecting to receive a total of ~40 trip diaries by the

end of October

  • Occupancy data from this study is used for this analysis

25

Pooled R d Ride de

1.48 48

No Non-Pooled R d Ride de

1.50 50

(264 f 264 far ares) (1,754 f fares es)

(Introduction) (Methodology) (Data Description) (Overlap Removal) (Occupa upanc ncy) (Fuel Efficiency) (Preliminary Results)

slide-26
SLIDE 26

26

Grams CO2 PMT = VMT (miles) x Fuel C

Consum umpt ption n x Conversion Factor

Ride VMT (miles) x Occupancy + Active/Transit PMT

Method

  • dolog
  • gy

Now let’s move on to fuel efficiency

(Introduction) (Methodology) (Data Description) (Overlap Removal) (Occupancy) (Fuel el E Efficien ency) (Preliminary Results)

  • Re

Real-world TNC f fuel efficie iency is not the same a as CA f fle leet a average

slide-27
SLIDE 27

Rated F Fuel Efficiency

  • First staff assigned each vehicle with a Fuel efficiency rating

derived from Federal Fuel Economy Data

27

Vehic hicle Ide dentif ificatio ion N n Num umber ( (VIN)

Make Model Model Year etc. Series Name

(Introduction) (Methodology) (Data Description) (Overlap Removal) (Occupancy) (Fuel el E Efficien ency) (Preliminary Results)

slide-28
SLIDE 28

28

Fuel E Econ

  • nom
  • my

Data P Proces essi sing Fl Flow C Cha hart

DMV MV VINtell ellig igence ce

VIN, Make Name, Model Name, Series Name, Model Year, etc. VI VIN Vehicle Specs incl. transmission type, # of cylinders, fuel type, advance vehicle type, drivetrain type All vehicle specs from DMV and VINtelligence including VIN EPA F Fuel E Econ

  • nom
  • my

(FuelE lEconomy.GOV OV)

Adv Advanced string match ching algorithm Fuel economies and vehicle specs

VIN and fuel Economies

TNC NC V VINs Ns TNC NC V VINs Ns + Rated ed F Fuel el Effici ciency cy

Note te: CO2 emissions are calculated assuming complete conversion of fuel carbon to CO2 and using conversion factors: 8,887 and 10,180 grams of CO2 per gallon of fuel, for gasoline and diesel, respectively

(Introduction) (Methodology) (Data Description) (Overlap Removal) (Occupancy) (Fuel el E Efficien ency) (Preliminary Results)

slide-29
SLIDE 29

Rated Fuel E Eco conomies

TNC NC-wide de v

  • vs. C

Californi nia L LDV ( V (MY2008+ 008+)

29

TNC fleet tend to have higher rated fuel efficiency values than California average passenger vehicles

(Introduction) (Methodology) (Data Description) (Overlap Removal) (Occupancy) (Fuel el E Efficien ency) (Preliminary Results)

1.0 0.8 0.6 0.4 0.2 0.0 Cumulative Probability 140 120 100 80 60 40 20 Fuel Economies (MPG) Statewide TNC-wide

slide-30
SLIDE 30

5 10 15 20 25 30 35 2.5 7.5 12.5 17.5 22.5 27.5 32.5 37.5 42.5 47.5 52.5 57.5 62.5 67.5 Miles per Gallon Average Vehicle Speed

Real MPG (EMFAC2017*)

In In-Us Use F e Fuel Effici ciency Varies es by Dr Driving g Condition

  • Fuel efficiencies are corrected for various driving conditions

30

City Fuel Efficiency

*MY2012 – 2018 Passenger Vehicles

(Introduction) (Methodology) (Data Description) (Overlap Removal) (Occupancy) (Fuel el E Efficien ency) (Preliminary Results)

slide-31
SLIDE 31

EMFAC2017 v 2017 vs. TNC 4 NC 40-Vehicle S Study

31

0.00 0.50 1.00 1.50 2.00 2.50 5 10 15 20 25 30 35 40 45 50 55 60 65 70 Normalized CO2 Speed Bin

CO CO2 Normalized d at 30 m 30 mph ph: : TNC v

  • vs. E

. EMFAC

average TNC EMFAC 2017

TNC fleet has lower emission rates compared to EMFAC at higher speeds

(Introduction) (Methodology) (Data Description) (Overlap Removal) (Occupancy) (Fuel el E Efficien ency) (Preliminary Results)

slide-32
SLIDE 32

How Elec ectric V Vehi hicles a es are e iden dentified ed?

  • Decoded VIN numbers to identify Battery, Plug-in hybrid, and Fuel Cell

Electric vehicles

  • Assumed 15% eVMT for PHEVs (preliminary assumptions)
  • 0.7% eVMT (30.2 million miles) as compared to total TNC VMT (i.e., 4.2

billion miles)

32

Plug ug-in in H Hybrid id Electric ic V Vehic icle les Battery Electric ic V Vehic icle les Fue Fuel Ce Cell Electric ic V Vehic icle les

Popula pulatio ion (P (Proportion of

  • f TNC fl

fleet) 5,400 (1%) 2,800 (0.5%) 50 (0.01%) VMT MT ( (millio llion m miles les) 43.6 23.6 0.17 eVMT ( (millio llion m n miles les) 6.5 23.6 0.17

(Introduction) (Methodology) (Data Description) (Overlap Removal) (Occupancy) (Fuel el E Efficien ency) (Preliminary Results)

slide-33
SLIDE 33

Prel eliminary R Resu esults

33

(Introduction) (Methodology) (Data Description) (Overlap Removal) (Occupancy) (Fuel Efficiency) (Preliminary R Results)

slide-34
SLIDE 34

VMT B By T y Time Per erio iod

34

(Introduction) (Methodology) (Data Description) (Overlap Removal) (Occupancy) (Fuel Efficiency) (Preliminary R Results)

slide-35
SLIDE 35

Cha Chang nge i in n VMT Be

Before And A After T Trip O Overlap R Removal

35

Tr Trip Periods ds VMT Befor

  • re Remov
  • val

(mil iles) VMT After Remova val (mil iles) Per ercen ent Change ge All Periods 4.49 billion 4.22 billion 6%

(Introduction) (Methodology) (Data Description) (Overlap Removal) (Occupancy) (Fuel Efficiency) (Preliminary R Results)

slide-36
SLIDE 36

VMT D Distribution B By S Speed B Bin

36

0% 2% 4% 6% 8% 10% 12% 14% 16% 18% 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 % of VMT MT Spe peed Bi Bins ns ( (MPH)

EMFAC2017 TNC data

(Introduction) (Methodology) (Data Description) (Overlap Removal) (Occupancy) (Fuel Efficiency) (Preliminary R Results)

  • Average Speeds are lower for TNCs
slide-37
SLIDE 37

Model el Y Year Di Distribution

37

200,000 400,000 600,000 800,000 1,000,000 1,200,000 1,400,000 10,000 20,000 30,000 40,000 50,000 60,000 70,000 80,000 90,000 100,000 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 Calif lifornia ia Passenger Vehic icle P Popula latio ion TNC F Fleet Vehicle le P Popula latio ion Vehic icle le Mo Model Y l Years TNCs EMFAC2017

Vehicles operating for TNC fleet are in general newer than California average passenger vehicles

(Introduction) (Methodology) (Data Description) (Overlap Removal) (Occupancy) (Fuel Efficiency) (Preliminary R Results)

slide-38
SLIDE 38

Flee eet A Aver erag age I In-Use Fu Fuel Effi ficiency (C (Corrected f for Dr Driving C g Condition

  • ns)

38

26. 26.0 28. 28.0 20 21 22 23 24 25 26 27 28 29 30 CA CA TNC

Fuel el Efficien ency ( (MPG) G)

(Introduction) (Methodology) (Data Description) (Overlap Removal) (Occupancy) (Fuel Efficiency) (Preliminary R Results)

  • The efficiency gain of newer model years vehicle might get slightly
  • ffset by slower trip average speeds
slide-39
SLIDE 39

Ou Out of

  • f 2.53

Billion M Miles

  • f
  • f P3

P3 V VMT

Po Pooling

39

(Introduction) (Methodology) (Data Description) (Overlap Removal) (Occupancy) (Fuel Efficiency) (Preliminary R Results)

  • Pool ride do not necessarily reflect the rides that were matched to

another passenger. It may include rides that were not matched too.

slide-40
SLIDE 40

Base Y se Year g g CO2 per er P PMT

40

(Introduction) (Methodology) (Data Description) (Overlap Removal) (Occupancy) (Fuel Efficiency) (Preliminary R Results)

  • Although TNC fleet has better in-use fuel efficiency, lower occupancy and higher

deadhead VMT drive higher emissions (+ 75% emissions per PMT)

  • For California fleet average; PMT/VMT of 1.68 was used

BAU i in CY2018 2018 Occupancy = =1.5 Deadheading= g=40% CA CA-like a assumpti tions Occupancy = = 1.68 Deadheading= g=0% 50 100 150 200 250 300 350 400 CA TNC g CO2 per PMT

189 189 202 202 355 355

slide-41
SLIDE 41

Bas Base Y e Year ear g CO2 per per PMT Sensitivity A Analysis

41

150 200 250 300 350 400 0.1 0.2 0.3 0.4 g CO2 p per er P PMT MT Fraction of deadh dhea eading V ing VMT MT CA fleet TNCs

slide-42
SLIDE 42

Next xt S Steps

  • Finalized data analysis using latest data submittal by TNCs
  • Late 2019 - Release draft emissions inventory

documentation

  • Early 2020 – Potential informational board item on initial

inventory and program overview

  • Q4 2019 – Workshop on regulatory concepts

42

slide-43
SLIDE 43

Tak ake H Home Q Ques estio ions

  • How can we further improve our estimates of occupancy? Are

there other data sources than those presented here?

  • Are there available data on %eVMT for PHEVs operating in

ridesharing business?

  • What assumptions should we make for active transportation in

the base year emissions inventory?

  • What are better ways to reflect the impact of

congestion/driving conditions on real world fuel efficiency?

43

slide-44
SLIDE 44

Comments ts a and Q Questi tions

  • Please submit your comments related to 2018 base year

emissions by October 25, 2019

  • Questions and comments can be emailed to

cleancars@arb.ca.gov

  • For more information on the Clean Mile Standards, please

visit our website at: https://ww2.arb.ca.gov/our- work/programs/clean-miles-standard

44