ideas + action for a better city...2017/02/28  · analytical modeling at the metropolitan...

Ideas + Action for a Better Citylearn more at SPUR.org

tweet about this event:

@SPUR_Urbanist

#MTCModeling

Analytical Modeling at the Metropolitan Transportation Commission

February 28, 2017

Lisa Zorn (lzorn@mtc.ca.gov) & Michael Reilly (mreilly@mtc.ca.gov)

Overview• Models in regional planning• Mike on UrbanSim, the land use model• Lisa on Travel Model One, the transportation model• Questions

3

Please note• Today’s talk is not a formal presentation in the Plan

Bay Area 2040 process• Scenarios are earlier versions

• Any comments will not be part of the EIR process• Please see http://planbayarea.org if you would like

to learn more or participate

4

What are Regional Models?• Complex, data-hungry computer programs• Use economics and statistics to forecast how

different parts of the city work and interact in an attempt to forecast the future

• At MTC, they use microsimulation• Explicit prediction of choices (e.g., Where do I

want to live? What time will I drive home?)5

Why Use Regional Models?• Forecast the future to better understand trajectory

and plan/evaluate transportation investments• Rigorous, consistent, and comprehensive

• Test the efficacy of transport and land use policies• Better understand how the region works and

what might ameliorate our problems• Evaluating alternate futures or scenarios

6

Regional Models at MTC• Various software forms an integrated model

• Regionwide total forecast: REMI• Local land use forecast: UrbanSim• Transportation behavior: CT-RAMP• Emissions: EMFAC• Other: health, benefit-c0st, equity assessment

7

Land Use Modeling• UrbanSim land use model

• Developed by Paul Waddell, UCB• Forecasts the intra-regional location

of households and jobs (and thebuildings that contain them) for a series of future years

8

Supply in UrbanSim• Start with map of all current buildings

• Attributes such as size, age, price• All households and jobs are explicitly assigned

using recent data on their locations

9

10

11

Demand in UrbanSim• Statistical equations developed from past behavior

explain the consistencies in location preferences• Every year new households (from REMI) and some

existing households choose a new housing unit• Very individual but there are correlations • Place them in these locations

• Jobs are similar12

Increasing Supply• Map of land use policies

• Mostly zoning, but also caps, fees, subsidies

• UrbanSim Developer Model simulates construction• Pro forma estimates profit = revenue – costs

• Costs from existing use, fees, constuction• Revenue starts with current prices, goes up in areas of

high demand

• Build the most profitable buildings13

14

15

16

17

Mini Pro Forma

18

Costs –$98m

Revenue $162m

Profit $64m

19

Scenarios• Built multiple scenarios with stakeholders

• Different visions relating to where growth ought to go• Use policies within the model to achieve

• Different transportation investments and policies

20

21

Four Initial PBA Scenarios• No Project and

PBA40 Scenarios Visions

22

23

24

Transportation ModelingTravel Model 1 forecasts the travel behavior of every resident on a typical weekday in the future

25

Demand in TM 1• Travel is a derived demand

• Start with land use model output: where do people live and where are their destinations?

• Explicit representation of people in households making many interrelated choices• Car ownership, where working, shopping, when

leave for a trip, what mode (car, walk, transit…)26

Supply in TM 1• Detailed representation of the travel network

• Roads with capacity, tolls• Transit with frequency, costs

• How do the trips generated by the demand model combine throughout the day to generate congestion and affect travel speeds/times

27

28

29

Scenario Output• Back to the scenarios introduced earlier• Travel Model used to assess alternate futures

• Different land use patterns means a different set of origins and destinations for trips

• Vary transportation investments and policies• Assess which one (or combinations) best achieve

regional goals30

31

32

11

Figure 2: Change in Roadw ay Lane Miles from 20 10

San

FranciscoSan Mateo Santa Clara Alam eda Contra Costa Solano Napa Sonom a Marin

-1.0 %

0 .0 %

1.0 %

2.0 %

3.0 %

4.0 %

5.0 %

6.0 %

7.0 %

8.0 %

9.0 %

10 .0 %

11.0 %

12.0 %

Gro

3313

Figure 3: Change in Transit Passenger Seat Miles from Year 20 10

Transit Technology

Local Bus Light Rail Ferry Express Bus Heavy Rail Com m uter Rail

0 %

5%

10 %

15%

20 %

25%

30 %

35%

40 %

45%

50 %

55%

60 %

Ch

3429

Figure 9: Year 20 35 Autom obile Ow nership Results

Zero autom obiles One autom obile Tw o autom obiles Three autom obilesFour or m ore

autom obiles

0 %

2%

4%

6%

8%

10 %

12%

14%

16%

18%

20 %

22%

24%

26%

28%

30 %

32%

34%

36%

38%

40 %

Sh

3534

Figure 12: Year 20 35 Autom obile Mode Shares for All Travel

Single occupant, No HOT

Single occupant, Pay touse HOT Tw o occupants, No HOT

Tw o occupants, Payto use HOT Three or m ore occupants

0 %

5%

10 %

15%

20 %

25%

30 %

35%

40 %

45%

50 %

55%

Sha

3637

Figure 14: Year 20 35 Typical W eekday Transit Boardings by Technology

Local Bus Light Rail Ferry Express Bus Heavy Rail Com m uter Rail

0

10 0 ,0 0 0

20 0 ,0 0 0

30 0 ,0 0 0

40 0 ,0 0 0

50 0 ,0 0 0

60 0 ,0 0 0

70 0 ,0 0 0

80 0 ,0 0 0

90 0 ,0 0 0

1,0 0 0 ,0 0 0

1,10 0 ,0 0 0

1,20 0 ,0 0 0

1,30 0 ,0 0 0

1,40 0 ,0 0 0

1,50 0 ,0 0 0

1,60 0 ,0 0 0

1,70 0 ,0 0 0

1,80 0 ,0 0 0

Typ

37

39

Figure 15: Year 20 35 Vehicle Miles Traveled per Hour by Tim e Period

Early AM (3 am to 6 am ) AM Peak (6 am to 10 am ) Midday (10 am to 3 pm ) PM Peak (3 pm to 7 pm ) Evening (7 pm to 3 am )

0 M

1M

2M

3M

4M

5M

6M

7M

8M

9M

10 M

11M

12M

13M

14M

VM

38

39

Questions?

lzorn@mtc.ca.govmreilly@mtc.ca.gov

40

41

42

43

25%

75%

24%

33%

43%

23%

77%

21%

33%

46%

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

outside PDA

in PDA

Inland, Coastal,

Delta

Bayside

Big 3 Cities

Compared to the Draft Preferred Scenario, the Final Preferred

Scenario boosts housing growth in the “Big 3” cities.

Where will the region plan for the 820,000new households?

30%

40%

30%

2010: 2.6 millionhouseholds

34%

38%

28%

2040: 3.4 million households

8

Draft

Draft

Draft

Draft

Draft

44

48%

52%

14%

46%

40%

45%

55%

17%

40%

43%

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

outside PDA

in PDA

Inland, Coastal,

Delta

Bayside

Big 3 Cities

New strategies included in the Final Preferred Scenario shifted

some job growth away from Bayside communities.

Where will the region plan for the 1.3 millionnew jobs?

33%

41%

26%

36%

41%

23%

2010: 3.4 million jobs

2040: 4.7 million jobs

9

Draft

Draft

Draft

Draft

Draft

Ideas + Action for a Better Citylearn more at SPUR.org

tweet about this event:

@SPUR_Urbanist

#MTCModeling