Urban transport modal shift: an energy systems approach 1E_Steve Pye_WEB.pdf · Change in Bpkm...

Urban transport modal shift: an energy systems approach

Steve Pye, Hannah DalyUCL Energy Institute

IEW 2015, June 3rd 2015

Passenger transport growth in the UK

Modal shift in ESME, IEW2015Steve Pye, 03.06.15

• Growth in passenger km increased 4-fold in UK since 1950, driven by increased ownership and use of cars

• No question of positive benefits…….but also negative environmental impacts

Source: Transport Statistics Great Britain 2013

2

Going beyond technology: the need for behavioural representation


• A move to more sustainable transport systems will depend on consumer choices

• Schäfer (2012) highlights a deficit of behavioural representation in many E3-type models– Elastic transportation demand– Endogenous mode choice– Choice of no physical travel– Accounting for infrastructure capacity– Segmenting urban and intercity transport

• Daly et al. (2014) demonstrate an approach to incorporating modal shift in a TIMES model; this paper builds on that approach

3

How to represent urban modal shift?


• We want to consider the following –

1. Demand: focus on shorter trips in urban areas where application most relevant

2. Mode speed: consumers appear to have a travel time budget of ~1 hour per day; therefore mode speed matters

3. Rate and level of shift: there are constraints on timescale of shifts, and maximum levels

4. Costs of shift: different infrastructure costs need to be considered if we are exploring policy-optimal solutions

• So how can this be implemented in a bottom-up, optimisation model?

4

Standard model implementation


Mode-specific surface

passenger travel demand (pkm) i,t

Motorised techs

pkm

Fuel inputs

5

MS model implementation


Urban surface passenger

travel demand (pkm) i,t

Motorised techs

Non-motorised techs

Bus

Cars

Rail

Bikes

Walking

Travel time budget (TTB) i,t

Technologies control for –i. Time budget needed for different modes

(based on average speed)ii. Current mode shares (calibration)iii. Investments for additional mode capacityiv. Rate and level of shiftProxy

infrastructure techs (pkm)

Existing

New

pkmpkm (pr)TTB

Modes x5 (as per above)

Fuel inputs

6

How to represent urban modal shift?


• Four factors to consider -

1. Demand2. Mode speed3. Rate and level of shift4. Costs of shift

7

1a. Demand: Disaggregating UK surface passenger demand


3926

21078657

6726

0

1000

2000

3000

4000

5000

6000

7000

8000

9000

10000

Rural Urban All

Aver

age

mile

age

/ ca

p. /

yea

r, 20

10

>35

<35

NTS annual mileage per capita (by area type)

8

1b. Demand: Total urban and rural demand by mode, 2010


Total ESME demands disaggregated by area type and mode, 2010

• Urban (<35 miles) accounts for 42% of surface transport passenger demand; disaggregated by region.

0

50

100

150

200

250

300

350

400

450

500

Rural + Urban (>35 miles) Urban (<35 miles)

Pass

enge

r tra

vel d

eman

d, B

pkm

Walk Cycle Bus Rail Car

9

2a. Mode speed & the role of time budgets


• Reason for introducing time budget in model is to ensure mix of modes• The ‘need for speed’ means we can’t all switch to slower non-motorised

modes

• Evidence base for time budgets– Aggregate constant time budget observed ~1 hr/day/cap; stability of time budgets

as a concept (Zahavi and Ryan, 1980) – Reasons for this – ‘ideal’ travel time budget exists? (Mokhtarian and Chen, 2004)– Large differences when disaggregated – age, car ownership, gender, income, spatial

characteristics (Mokhtarian and Chen, 2004)– Different in other countries; rising in Netherlands (van Wee et al. 2006)– Sceptical positions; stability of concept is questioned (Goodwin 1981)

10

2b. Mode speed & the role of time budgets


• Problem of assuming constant budget for analysis• Overall demand increasing at higher rate than population; holding time

budget constant requires that average speeds have to increase (red dash line)• However, mode shift potential may require a higher share of slower modes

0.95

1.00

1.05

1.10

1.15

1.20

1.25

1.30

1.35

2010 2015 2020 2025 2030 2035 2040 2045 2050

Inde

x (2

010=

1)

Urban population

Urban demand(0-35)

Ave. urban speed(Const. TB)

Ave. urban speed(Adj. TB) Adj. time budget of

+7.5% by 2050 (1.02 hrs compared to 0.95)

11

3a. Rate & level of mode shift estimates


• Assess urban trip distance profile; replacement of car trips by other modes limited due to distance

Modal shift in ESMESteve Pye, 10.09.14 12

0

100

200

300

400

500

600

700

800

900

1000

<1 1-2 2-3 3-5 5-10 10-15 15-25 25-35Av

erag

e m

iles (

pre-

shift

)

Walking Cycling Bus Rail Car

0

20

40

60

80

100

120

140

160

180

200

<1 1-2 2-3 3-5 5-10 10-15 15-25 25-35

Aver

age

trip

s (pr

e-sh

ift)

Walking Cycling Bus Rail CarTrips

Greater London: Average miles & trips /capita by MODE & TRIP DISTANCE

Miles

12

3b. Rate & level of mode shift estimates


• Determine maximum mode shift by 2050. Already limited by distance, additional information from other analysis / international experience

• Rates of mode shift over time based on linear interpolation, achieving maximum in 2050

13Greater London; max. permitted change in per capita miles by mode

10%

647%

25%

23%

-44%

0

200

400

600

800

1000

1200

1400

1600

1800

2000

Walking Cycling Bus Rail Car

Aver

age

mile

s

Pre-shift

Post-shift

Trip mileage share by mode

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

% pre-shiftshare

% post-shift share

% pre-shiftshare

% post-shift share

Mileage Trips

Mile

age

/ tr

ip sh

ares Car

Rail

Bus

Cycling

Walking

13

4a. Cost factors in mode shift


• Within constraint of mode shift potential and rate, optimisation will play role• Mode costs considered given inter-modal competition, and non-mechanised

modes• Inclusion of infrastructure costs; no repr. of other key factors (value of time,

convenience etc.)

0

0.05

0.1

0.15

0.2

0.25

Bus Car Rail Rail (exist) Cycle

£/pk

m, 2

030

INVEST FIXOM FUEL INFRA.14

Model analysis


• Runs focus on exploring model behaviour and future application of approach• All runs under LT climate policy scenario

Run DescriptionRef (v3.3) ESME v3.3 standard run for comparison

MS-Ref Modal shift ref. case (as presented)

MS-High Strong push on sustainable transport, increasing MS potential

MS-NoTTB Innovation erodes assumptions of time budgets

MS-HighCC External costs penalising car travel

MS-HighCC NoTTB Combined sensitivity case

15

-80

-60

-40

-20

0

20

40

60

80

MS-

Ref

MS-

High

MS-

NoT

TB

MS-

High

CC

MS-

High

CC N

oTTB

MS-

Ref

MS-

High

MS-

NoT

TB

MS-

High

CC

MS-

High

CC N

oTTB

MS-

Ref

MS-

High

MS-

NoT

TB

MS-

High

CC

MS-

High

CC N

oTTB

MS-

Ref

MS-

High

MS-

NoT

TB

MS-

High

CC

MS-

High

CC N

oTTB

2020 2030 2040 2050

Chan

ge in

Bpk

m tr

avel

led,

com

pare

d to

Ref

.

Car

Rail

Bus

Cycle

Walk

Levels of shift: key sensitivities • Increasing levels of mode shift over time

• Levels of shift – 5-15% of demand by 2040/50

Cycle always maximises in model; however, issue w/ trade-off

Bus stronger role where time constr. relaxed / car

costs hiked

Strong car reduction only if cost penalty introduced,

and time constraint relaxed

16

Technology impacts: change in car capacity


-5

-4

-3

-2

-1

0

1

2

3

4

MS-

Ref

MS-

High

MS-

NoT

TBM

S-Hi

ghCC

MS-

High

CC N

oTTB

MS-

Ref

MS-

High

MS-

NoT

TBM

S-Hi

ghCC

MS-

High

CC N

oTTB

MS-

Ref

MS-

High

MS-

NoT

TBM

S-Hi

ghCC

MS-

High

CC N

oTTB

MS-

Ref

MS-

High

MS-

NoT

TBM

S-Hi

ghCC

MS-

High

CC N

oTTB

2020 2030 2040 2050

Chan

ge in

car

cap

acity

, mill

ion

vehi

cles

Car PHEV

Car ICE

Car hydrogenFCV

Car hybrid

2020-2040: Carbon constraint sees ICEs drop out

of mix

2050: Reducing mitigation costs through MS headroom (reflected

in marginal CO2 cost)

17

Reductions in passenger transport emissions

Modal shift in ESMESteve Pye, 10.09.14

-8.0

-7.0

-6.0

-5.0

-4.0

-3.0

-2.0

-1.0

0.0

1.0

2.0

MS-

Ref

MS-

High

MS-

NoT

TB

MS-

High

CC

MS-

High

CC N

oTTB

MS-

Ref

MS-

High

MS-

NoT

TB

MS-

High

CC

MS-

High

CC N

oTTB

MS-

Ref

MS-

High

MS-

NoT

TB

MS-

High

CC

MS-

High

CC N

oTTB

MS-

Ref

MS-

High

MS-

NoT

TB

MS-

High

CC

MS-

High

CC N

oTTB

2020 2030 2040 2050

Chan

ge in

TO

TAL s

urfa

ce tr

ansp

ort p

asse

nger

em

issi

ons,

M

tCO

2

Car

Rail

Bus

• Highest % reductions where non-motorised demand increases / motorised decreases• Percentage reductions small in 2050 due to strongly decarbonised sector & reduction in

ULEV vehicles• Broadly speaking, shifts supply-side options down the technology cost curve

18

Findings


• Cost optimal solutions favour sustainable transport modes if included. Relative strength is contingent on disincentivising car travel, and affecting travel time considerations

• Demonstrates application of approach to mode shift in energy systems models, and sub-optimality of supply-side focused approaches

• Strengths of approach– Demonstrates approach in full systems model, and considers non-motorised modes– Begins to capture infrastructure requirements explicitly– Allows for endogenous shift, capturing technology trade-offs

• Limitations– Lack of ‘choice’ considerations (engineering perspective only)– Data needs and model re-configuration not insigificant

19

Further research


• Extend scope of approach to capture longer distance trips; this is where most emissions are derived

• Key questions across constraints– Level of time budget?– Maximum shift achievable, and rate of change?

• Development of consumer level choice parameters

• Develop infrastructure capacity constraints, impacting on mode speed (depending on investment)

• Consideration of externalities, particularly relevant for urban transport (pollution, noise, congestion etc.)

20

Thanks for listening. Any questions?

[email protected]@st_pye

Modal shift in ESME, IEW2015Steve Pye, 03.06.15 21

mailto:[email protected]

References

Transport demands in ESMESteve Pye, 08.01.14

• Banister, D. (2008). The sustainable mobility paradigm. Transport policy, 15(2), 73-80.• Daly et. al (2014), Incorporating travel time and behaviour into TIMES energy system models, Applied Energy, In Press.• DfT (2011). Realising the potential of GB rail: final independent report of the rail value for money study. Department for Transport. May

2011. http://assets.dft.gov.uk/publications/report-of-the-rail-vfm-study/realising-the-potential-of-gb-rail-summary.pdf• EC (2011). Flash Eurobarometer “Future of transport” (No 312). European Commission, Brussels. March 2011.

http://ec.europa.eu/public_opinion/flash/fl_312_en.pdf• GLA (2013). The Mayor’s vision for cycling in London. Greater London Authority. March 2013.

http://www.london.gov.uk/sites/default/files/Cycling%20Vision%20GLA%20template%20FINAL.pdf• Goodwin, P. (2013). Get Britain Cycling - Report from the Inquiry. All Party Parliamentary Cycling Group (APPCG). April 2013.

http://allpartycycling.files.wordpress.com/2013/04/get-britain-cycling_goodwin-report.pdf• Goodwin, P. (2007). Carbon abatement in transport – Review of demand responses. A review on behalf of the Committee on Climate

Change. December 2007. http://archive.theccc.org.uk/aws3/CCC%20transport%20demand%20review%20Goodwin.pdf• Goodwin, P.B., 1981. The usefulness of travel budgets. Transportation Research 15A, 97–106.• Kenyon, S., & Lyons, G. (2003). The value of integrated multimodal traveller information and its potential contribution to modal

change. Transportation Research Part F: Traffic Psychology and Behaviour, 6(1), 1-21.• Mann, E., & Abraham, C. (2006). The role of affect in UK commuters' travel mode choices: An interpretative phenomenological

analysis. British Journal of Psychology, 97(2), 155-176.• McKinsey (2011). Keeping Britain moving: the United Kingdom’s transport infrastructure needs. March 2011.

http://www.mckinsey.com/global_locations/europe_and_middleeast/united_kingdom/en/latest_thinking/keeping_britain_moving_full• Mokhtarian, P. L., & Chen, C. (2004). TTB or not TTB, that is the question: a review and analysis of the empirical literature on travel time

(and money) budgets. Transportation Research Part A: Policy and Practice, 38(9), 643-675.• Schäfer, A (2012). Introducing Behavioral Change in Transportation into Energy/Economy/Environment Models. World Bank Policy

Research Working Paper 6234. http://elibrary.worldbank.org/doi/book/10.1596/1813-9450-6234

22

http://assets.dft.gov.uk/publications/report-of-the-rail-vfm-study/realising-the-potential-of-gb-rail-summary.pdf

http://ec.europa.eu/public_opinion/flash/fl_312_en.pdf

http://www.london.gov.uk/sites/default/files/Cycling%20Vision%20GLA%20template%20FINAL.pdf

http://allpartycycling.files.wordpress.com/2013/04/get-britain-cycling_goodwin-report.pdf

http://archive.theccc.org.uk/aws3/CCC%20transport%20demand%20review%20Goodwin.pdf

http://www.mckinsey.com/global_locations/europe_and_middleeast/united_kingdom/en/latest_thinking/keeping_britain_moving_full

http://elibrary.worldbank.org/doi/book/10.1596/1813-9450-6234

References

Transport demands in ESMESteve Pye, 08.01.14

• Schwanen, T., Banister, D., & Anable, J. (2012). Rethinking habits and their role in behaviour change: the case of low-carbonmobility. Journal of Transport Geography, 24, 522-532.

• TfL (2010). Analysis of Cycling Potential: Policy Analysis Research Report. Transport for London. December 2010. https://www.tfl.gov.uk/cdn/static/cms/documents/analysis-of-cycling-potential.pdf

• van Wee, B., Rietveld, P., & Meurs, H. (2006). Is average daily travel time expenditure constant? In search of explanations for an increasein average travel time. Journal of transport geography, 14(2), 109-122.

• Zahavi, Y., & Ryan, J. M. (1980). Stability of travel components over time.Transportation research record, (750). Transportation ResearchBoard ISSN: 0361-1981.

23

https://www.tfl.gov.uk/cdn/static/cms/documents/analysis-of-cycling-potential.pdf

Additional slides

Transport demands in ESMESteve Pye, 08.01.14 24

Per capita transport demand by area type and mode


• Average mileage dominated by car travel– 87% (Rural + Urban, >35)– 78% (Urban, <35)

• Greater London profile distinctive; more bus and rail travel, and lower overall demand

• Urban <35 travel dominates trip profile (97%)

• Distinctive trip profile for Greater London

Mileage

Trips

0

1000

2000

3000

4000

5000

6000

7000

8000

9000

Rural Urban (>35) Urban (<35) G. London (<35) Other Urban(<35)

Aver

age

mile

age

per y

ear,

2010

Walk Bus Cycle Rail Car

0

50

100

150

200

250

300

350

400

Rural Urban (>35) Urban (<35) G. London(<35)

Other Urban(<35)

Aver

age

trav

el ti

me

(hrs

) per

yea

r, 20

10

Walk Cycle Bus Rail Car 25

Mode shift: factors


Source: Flash Eurobarometer “Future of transport” (EC 2011)

26

Mode shift factors


Literature suggests many factors at play -

– Lifestyle stability. Change of home, workplace often key factor, known as ‘churn’ (Goodwin 2007).

• ‘9 year period, 50% of commuters changed main mode at least once’

– Lack of information (Kenyon and Lyons 2003).– Habit of mode choice (Schwanen et al. 2012).– Public acceptability and demonstration (Bannister 2008).– Travel as valued activity, not just derived demand e.g. use of ICT on public transport

(Bannister 2008).– The role of affective (emotional) considerations. Car travel provides autonomy,

personal space and ownership / identity (Mann and Abraham 2006).– Land use patterns. Future planning of communities has strong bearing on transport

choices (Bannister 2008).

27

Mode shift rates and potential


• Projected urban passenger demands from ESME projections (blue continuous line)• Max shift multiplier applied to 2050 demands (excl. cars), and linear extrapolation back

to 2010• Shift above projected demand (purple shaded area)• Any additional growth (total shaded area) subject to infrastructure costs

0

10

20

30

40

50

60

70

80

2010 2015 2020 2025 2030 2035 2040 2045 2050

Bpkm

Shiftpotential

Proj.potential

Rail (proj.)

Rail (shiftpot.)

28

Mode shift costs: infrastructure considerations


• Infrastructure costs considered for different modes, to ensure greater cost comparability

Cycling Under Get Britain Cycling report, consensus around £10-20 / capita year-on-year spend being able to deliver trip mode shares of 20-40% (Goodwin 2013); London Strategy funded at £18 / head to deliver 400% increase by 2026 (GLA 2013).Cost of bikes not included.

Cars Future investment needs based on McKinsey (2011) Keeping Britain moving: the United Kingdom’s transport infrastructure needs.

Rail Costs of infrastructure investment and operation added, based on report Realising the potential of GB rail: final independent report of the rail value for money study(DfT 2011). Excluded rolling stock investment (as already in ESME).Future investment needs based on McKinsey (2011) Keeping Britain moving: the United Kingdom’s transport infrastructure needs.

Bus As for cars but lower costs in model due to occupancy factor, as McKinsey estimates in vkm.

29

Urban transport modal shift: an energy systems approach 1E_Steve Pye_WEB.pdf · Change in Bpkm...

Documents

Transcript of Urban transport modal shift: an energy systems approach 1E_Steve Pye_WEB.pdf · Change in Bpkm...