[PPT] - Upcoming Work June 2010 John E Abraham with the HBA and UCDavis PowerPoint Presentation

SLIDE 1

PECAS Recent and Upcoming Work

June 2010 John E Abraham with the HBA and UCDavis PECAS Team

SLIDE 2

Part 1: Recent Work

Socioeconomic/Floorspace Rent Transport costs Setup of production Activity Allocation module for Long Distance Commercial Vehicle base matrices Software improvements ·

SLIDE 3

Part 2: Upcoming Work

Floorspace Synthesizer Disaggregate SD estimation Cluster analysis of technology options

SLIDE 4

Socioeconomic/Floorspace

Population Synthesizer Employment Synthesizer Floorspace type allocation Floorspace quantity calculation

Eric Lehmer, Shengyi Gao, John Abraham, Nicholas Linesch

SLIDE 5

Synthesizer Software

“Combinatorial Optimization”

Random changes to optimize objective function Add, remove or subtract PUMS households from TAZ’s

“Simulated Annealing”

Early on, accept changes that do not improve objective function with some probability Stops process from getting “stuck” at local optimum Based on annealing of crystals in metallurgy (Not found to be necessary with well chosen target weights)

SLIDE 6

Population Samples

Data source: census 2000 PUMS Synthesizing all people in each TAZ Sampling: taking all the samples within the PUMA in which a TAZ nests and 20% random samples from the adjacent PUMAs Duplicate unique IDs based on their weights to get a full population Control variables in the sample: household size, housing type, income group, age, number of cars, occupation, student status, group quarter flag Automated with Python

SLIDE 7

Population Targets

Data source: census SF tables Control variables: household size, housing type, income group, age, number of cars, occupation, student status, group quarter flag, number of rooms, number of bedrooms Targets: population by TAZ and each control variable Automated with Python

SLIDE 8

Population Synthesizer

Programmed in Java Maximum iterations: 1,000,000 per PUMA Outputs: the same unique person IDs as in the PUMS by TAZ Goodness of fit: simulated volume/targets > 0.99 Synthetic population by TAZ: each TAZ has all the population (unique person and household IDs) with all the census variables and aggregated PECAS household types, industries, occupations, and synthetic floorspace

SLIDE 9

Processing of Population Synthesis Outputs

Java code outputs a CSV file with 12208610 synthetic households. Each record has the TAZ and PUMS Household ID number (SN). This output was joined to a modified PUMS household table (hh) to get detailed information on each hh. The PUMS hh table was modified in the following ways: (1) ALL SENIOR HOUSEHOLDS: In order to get the population into the correct PECAS household types, households with all seniors had to be determined. This was done by examining the person records for the members of each household. Those households with all members > 65 yrs old were set to all senior. (2) HIGHRISE PUMA: In an attempt to capture the residential highrise spacetype we visually identified the PUMAS with existing highrise buildings and and this was then attached to the hh using the PUMA id. (3) URBAN PUMA: Visual representation of urban PUMAs. This was used as a criteria for the housing value calculations of the spacetype determination. (4) MILITARY HOUSEHOLD: Single person households where the person was active military living in group quarters were identified. This is so they can be put into gq spacetype.

SLIDE 10

Calculation of Residential SpaceType and SpaceQuantity

A SQL Server view was created based in part on work done by SANDAG.

This view split the hh into different spacetypes based on several criteria: (1) PUMS BLDGSZ attribute. The building size value determined whether it was mobile home, sfd, multi-family, etc. (2) ACRES attribute. Determined whether it was rural, acreage, or SFD. (3) The Urban PUMA combined with hh VALUE and hh RENT attributes determined whether it was Econ or Lux spacetype. (4) The Highrise PUMA attribute allong with the hh income category helped us determine which large apartment buildings might be highrise buildings. (5) Military single person hh were put into GQ The quantity of space was also assigned using the PUMS attributes of ROOMS, BEDRMS,YRBUILT and square foot estimates derived from SANDAG.

SLIDE 11

Employment synthesizer

SLIDE 12

Employment Sample and Targets

Data sources: census 2000 CTTP and PUMS; infoUSA Synthesizing all jobs in each TAZ by PECAS industry Sample: all persons with a job in PUMS with CTPP industry and occupation Targets: 1) CTTP employment by PECAS industry or

ccupation; no crosswalk; weight 1; 2) InfoUSA

employment by TAZ, PECAS industry and occupation; weight 0.2 0.01

SLIDE 13

Employment Synthesizer

Programmed in Java Maximum iterations: 1.5 billion Output: PUMS unique person IDs by TAZ Goodness of fit (GOF): simulated volume/targets at TAZ level; 1) for CTPP targets, GOF > 0.99; 2) for InfoUSA, GOF = 0.65; 3) for the pooled CTPP and InfoUSA targets, GOF = 0.82 Synthetic employment: all employees with PUMS variables within each TAZ Floorspace use:

Floorspace use rate: surveyed average floorspace rate by PECAS industry A floorspace type is assigned based on existing land use type within each TAZ (if a space type exists, then assign it by allowable use; otherwise, randomly assign a space type)

SLIDE 14

Java program outputs a CSV with 14285953 synthetic employees. Each employee has a TAZ and PUMS person table id attached. The PUMS person table was augmented in the following ways:

(1) PECAS Industry and Occupation reclassification of PUMS records to fit

employees into correct PECAS activities.

(2) Blue and White Collar labor split. (3) Agriculture and Construction activities randomly distributed to finer

classification than PUMS using California Statistical Abstract

proportions. For example, PUMS does not differentiate cattle production.

We randomly assigned them to: CATTLE RANCHING AND FARMING production, DAIRY CATTLE AND MILK production, or NONCATTLE ANIMAL production.

Processing of Employment Synthesizer Outputs:

SLIDE 15

Based on the synthetic employee’s PECAS activity type we can determine

which spacetypes are possible for that employee to occupy.

Based on the synthetic existing land use layer we can determine which

spacetypes are available in the employee’s TAZ.

Unable to determine the proportions that employee’s of a particular

activity consume spacetypes we decided to randomly assign employees to spacetypes that were available or if no spacetype was available we randomly assigned them to a possible spacetype and call that non- conforming space.

Quantity was based on activity type with each employee consuming a

certain amount of floorspace. This was then summed up by TAZ and SpaceType to form FloorspaceI for Industry.

Calculation of Industry SpaceType and SpaceQuantity:

SLIDE 16

Rent Estimation

SLIDE 17

Rent Estimation

Separate rent (use value of space) into two parts

LUZ value based on accessibility and supply/demand relationship Modifiers based on site and structure specific attributes

SLIDE 18

Estimation Results (Residential Space)

Rural Economic Residential* Rural Luxury Residential Single Family Detached Economic Residential Single Family Detached Luxury Residential Max Distance Coefficient Function Max Distance Coefficient Function Max Distance Coefficient Function Max Distance Coefficient Function Age

.176

Power

.176

Power

.007

Power .180 Power Coast 1.5

Shift. Expo

1.5

Shift. Expo

1.5 .012

Shift. Expo

1.5 .218

Shift. Expo

Freeway 2

.198
Shift. Expo

2

.198
Shift. Expo

3

.005
Shift. Expo

3

.056
Shift. Expo

Highway 2

.050
Shift. Expo

2

.050
Shift. Expo

2 .030

Shift. Expo

2 .022

Shift. Expo

Park 1 .289

Shift. Expo

1 .289

Shift. Expo

.25 .002

Shift. Expo

.25 .003

Shift. Expo

Rail 1 .379

Shift. Expo

1 .379

Shift. Expo

.5

.018
Shift. Expo

.5

.012
Shift. Expo

Ramp .75

Shift. Expo

.75

Shift. Expo

.5

.002
Shift. Expo

.5

.013
Shift. Expo

School 1

.262

Expo 1

.262

Expo 1 .018 Expo 1 .035 Expo Water body 2 .072

Shift. Expo

2 .072

Shift. Expo

.5 .005

Shift. Expo

.25

.002
Shift. Expo

Transit Center 1.5

Shift. Expo

1.5

Shift. Expo

1

.036
Shift. Expo

1

.011
Shift. Expo

SLIDE 19

Estimation Results (Residential Space)

Acreage Economic Residential Acreage Luxury Residential High-rise Economic Residential High-rise Luxury Residential* Max Distance Coefficient Function Max Distance Coefficient Function Max Distance Coefficient Function Max Distance Coefficient Function Age

.036

Power .081 Power

.264

Power

.264

Power Coast 5 .002 Shift. Expo 1.5 .079 Shift. Expo 1.5 1.804 Shift. Expo 1.5 1.804 Shift. Expo Freeway 1.5 .070 Shift. Expo 1.5

.014

Shift. Expo 2

.398

Shift. Expo 2

.398

Shift. Expo Highway 2 .017 Shift. Expo 2 .001 Shift. Expo 1.5 .766 Shift. Expo 1.5 .766 Shift. Expo Park .25 .071 Shift. Expo .25

.013

Shift. Expo .25 .056 Shift. Expo .25 .056 Shift. Expo Rail 1

.037

Shift. Expo 1

.024

Shift. Expo 1

.255

Shift. Expo 1

.255

Shift. Expo Ramp .5

.042

Shift. Expo .5

.010

Shift. Expo 1 .436 Shift. Expo 1 .436 Shift. Expo School 1 .029 Expo 1 .036 Expo 1 .125 Expo 1 .125 Expo Water body .25 .079 Shift. Expo .25 .033 Shift. Expo .5 Shift. Expo .5 Shift. Expo Transit Center 3 .090 Shift. Expo 1.5 .041 Shift. Expo 1.5 .033 Shift. Expo 1.5 .033 Shift. Expo

SLIDE 20

Estimation Results (Residential Space)

Joint Economic Residential Joint Luxury Residential Low-rise Economic Residential Low-rise Luxury Residential* Max Distance Coefficient Function Max Distance Coefficient Function Max Distance Coefficient Function Max Distance Coefficient Function Age

.099

Power .099 Power .029 Power .029 Power Coast 1.5 .045 Shift. Expo 1.5 .162 Shift. Expo 1.5 .116 Shift. Expo 1.5 .116 Shift. Expo Freeway 3

.010

Shift. Expo 3

.069

Shift. Expo 3

.012

Shift. Expo 3

.012

Shift. Expo Highway 2 .005 Shift. Expo 2 .016 Shift. Expo 2 .070 Shift. Expo 2 .070 Shift. Expo Park .25 .006 Shift. Expo .25 .017 Shift. Expo .25 .033 Shift. Expo .25 .033 Shift. Expo Rail .5

.019

Shift. Expo 1

.017

Shift. Expo 1

.010

Shift. Expo 1

.010

Shift. Expo Ramp 1

.002

Shift. Expo 1

.017

Shift. Expo .5 .017 Shift. Expo .5 .017 Shift. Expo School 1 .043 Expo 1 .040 Expo 1 .028 Expo 1 .028 Expo Water body .5 .017 Shift. Expo .5 .023 Shift. Expo 1

.004

Shift. Expo 1

.004

Shift. Expo Transit Center .5 .003 Shift. Expo 1

.004

Shift. Expo 1.5

.021

Shift. Expo 1.5

.021

Shift. Expo

SLIDE 21

Estimation Results (Residential Space)

Urban Mobile Home Residential Max Distance Coefficient Function Age .001 Power Coast 1.5 .281 Shift. Expo Freeway 2 .134 Shift. Expo Highway 1.5 .042 Shift. Expo Park .25 .084 Shift. Expo Rail 1 .181 Shift. Expo Ramp .5 .063 Shift. Expo School 1 .051 Expo Water body 1 .079 Shift. Expo Transit Center 1 .944 Shift. Expo

SLIDE 22

Estimation Results (Non-Residential Space)

Downtown Retail Space Highway Retail Space Mall and Big Box Retail Space Neighborhood Retail Space Max Distance Coefficient Function Max Distance Coefficient Function Max Distance Coefficient Function Max Distance Coefficient Function Age

.036

Power

.008

Power

.217

Power .0005 Power Coast 1.5 .002 Shift. Expo 1.5 .151 Shift. Expo 1.5 .289 Shift. Expo 1.5 .183 Shift. Expo Freeway 1 .070 Expo 1 .005 Expo 1.5 .175 Expo 2 .047 Expo Highway 1 .017 Expo .5

.083

Expo 2

.147

Expo 1.5

.047

Expo Park .2 .071 Shift. Expo .25

.011

Shift. Expo .25 .032 Shift. Expo .25 .103 Shift. Expo Rail 1

.037

Shift. Expo 1

.196

Shift. Expo 1 .048 Shift. Expo 1 .011 Shift. Expo Ramp 1

.042

Expo .25 .069 Expo 1.5 ** Expo 1.5

.058

Expo School 1 .029 Expo 1

.014

Expo 1 .150 Expo 1 .005 Expo Water body 1 .079 Shift. Expo 1 .046 Shift. Expo 1

.163

Shift. Expo 1 .045 Shift. Expo Transit Center 1.5 .090 Shift. Expo 1.5 .066 Shift. Expo 1.5

.080

Shift. Expo 1.5

.035

Shift. Expo

SLIDE 23

Estimation Results (Non-Residential Space)

Low Density Office Space High Density Office Space Max Distance Coefficient Function Max Distance Coefficient Function Age .046 Power .097 Power Coast 1.5 .094

Shift. Expo

1.5 .111

Shift. Expo

Freeway 1

.023

Expo 1

.129

Expo Highway 1.5 .015 Expo 1

.069

Expo Park .2

.040
Shift. Expo

.2 .085

Shift. Expo

Rail 1

.037
Shift. Expo

1

.195
Shift. Expo

Ramp 1.5 .020 Expo 1.5 .055 Expo School 1

.011

Expo 1 .012 Expo Water body 1 .007

Shift. Expo

1 .029

Shift. Expo

Transit Center 1

.004
Shift. Expo

1 .010

Shift. Expo

* Estimations are affected by lack of observations. ** Drop the field due to high collinearity.

SLIDE 24

Estimation Results (Non-Residential Space)

SLIDE 25

Transport Cost Coefficients

SLIDE 26

Outputs of transport cost coefficients

the list of skims used from the transport model, indicating how these skims are converted into the commodity transport utilities used in the AA Module; the estimated values of the coefficients used in the conversion of transport skims into commodity transport utilities for the specification of the setup and use of year that the PECAS model is to be calibrated to reproduce in terms of cross-sectional distributions

SLIDE 27

Four commodity categories

goods commodities household obtained service commodities labor commodities worker delivered service commodities

SLIDE 28

Goods commodities

Tranc,j,k = κDc · Distj,k + κTc · Timej,k + κMc · Tollj,k

κDc = $ value of transport utility per unit commodity per mile

= -1 · ($ value vehicle operating costs per vehicle-mile)/(commodity load in money per vehicle)

κTc = $ value of transport utility per unit of commodity per minute

= -1 · (driver wage)/(commodity load per vehicle)

κMc = $ value of transport utility per unit of commodity per $ of toll charge

= -1/(commodity load per vehicle)

SLIDE 29

Goods commodities

Commodity Name κDc κTc κMc $ / $-mile $ / $-minute $ / $-$ Agriculture Animals output

1.041E-04
1.932E-05
7.661E-05

Agriculture Plants output

1.266E-04
1.517E-05
6.015E-05

Agriculture Forestry and Fishing output

8.907E-05
1.213E-05
4.490E-05

Mining and Extraction output

1.070E-04
1.973E-05
7.302E-05

Manufacturing Food output

8.387E-05
1.312E-05
5.601E-05

Manufacturing Textiles output

7.530E-06
1.775E-06
7.577E-06

Manufacturing Wood Products Printing Furniture Misc output

6.334E-06
9.585E-07
4.092E-06

Manufacturing Petro Paper Chemicals Plastic Rubber Glass Cement output

1.066E-05
1.636E-06
6.985E-06

Manufacturing Metal Steel Machinery output

8.039E-06
1.412E-06
5.226E-06

Fuels

2.441E-04
3.082E-05
1.141E-04

Scrap

9.404E-04
1.412E-04
6.028E-04

Commodity Name commodity load in tons per vehicle value to weight ratio commodity load in money per vehicle $ value vehicle

perating costs

per vehicle-mile tons / truck $ / ton $ / truck $ / mile Agriculture Animals output 13.6 960.2 13053.2 1.4 Agriculture Plants output 21.0 790.0 16626.2 2.1 Agriculture Forestry and Fishing output 19.8 1122.7 22272.2 2.0 Mining and Extraction output 14.6 934.9 13694.0 1.5 Manufacturing Food output 15.0 1192.4 17853.4 1.5 Manufacturing Textiles output 9.9 13281.0 131973.1 1.0 Manufacturing Wood Products Printing Furniture Misc output 15.5 15787.2 244387.4 1.5 Manufacturing Petro Paper Chemicals Plastic Rubber Glass Cement output 15.3 9381.9 143170.2 1.5 Manufacturing Metal Steel Machinery output 15.4 12439.9 191338.1 1.5 Fuels 21.4 409.7 8766.5 2.1 Scrap 15.6 106.3 1658.8 1.6

Values Variables

SLIDE 30

Household obtained service commodities

Tranc,j,k = κHp,c · LogSump,j,k

κHp,c= $ value of transport utility per unit of commodity per unit of logsum

utility for person trip = 2/((util / $) · (units of commodity per visit))

Units of commodity per visit= (total amount of commodity from aggregate economic flow table) / (total trips per day * total days per year ) util / $ means transport money cost coefficient which comes from mode choice estimation of travel model.

SLIDE 31

Household obtained service commodities

Commodity Name κHp,c $ / $-util Retail Food output 0.044236 Retail Automotive output 0.044236 Retail General Merchandise output 0.044236 Restaurant output 0.065630 Amusement services and media consumption 0.111746 health services output 0.031704 Retail based services 0.031704 Religion output 0.031704 Higher education output 0.021031 Gradeschool Attendance 1.881468 Parks attendance 1.881468 Commodity Name Travel Segment total amount of commodity obtained model area wide per year visits per day amount of commodity obtained per visit Skim Name $ / year visits / day $ / visit Retail Food output HBO_SHOPPING 18,092,240,918 1,181,583 42.5 Retail Automotive output HBO_SHOPPING 50,477,817,182 3,296,648 42.5 Retail General Merchandise output HBO_SHOPPING 66,100,154,205 4,316,924 42.5 Restaurant output HBO_EAT 43,216,689,399 4,187,521 28.7 Amusement services and media consumption HBO_RECREATION 19,358,139,399 3,193,723 16.8 health services output HBO_PER BUSINESS 119,272,487,305 5,582,815 59.3 Retail based services HBO_PER BUSINESS 17,726,768,721 829,741 59.3 Religion output HBO_PER BUSINESS 6,041,439,400 282,783 59.3 Higher education output HBS 19,813,993,439 1,476,556 89.5 Gradeschool Attendance HBS 5,784,832 1 Parks attendance HBO_PARK 321,763 1

Values Variables

SLIDE 32

Labor commodities

Tranc,j,k = κLp,c · LogSump,j,k

κLp,c = transport disutility per unit of labor commodity per unit of logsum utility for

person trip = 2/((util / $) · (units of commodity per visit))

Units of commodity per visit= (total amount of commodity from aggregate economic flow table) / (total trips per day * total days per year ) util / $ means transport money cost coefficient which comes from mode choice estimation of travel model.

SLIDE 33

Labor commodities

Commodity Name

person type κLp,c travel segment skim $ / $-util Agriculture workers < 16 years old Agriculture workers No cars hh and >= 16 years old 0.010115 Agriculture workers Cars less than workers hh and >= 16 years old 0.029677 Agriculture workers Cars greater equal than workers hh and >= 16 years old 0.029892 Assembly and Fabrication workers < 16 years old Assembly and Fabrication workers No cars hh and >= 16 years old 0.005509 Assembly and Fabrication workers Cars less than workers hh and >= 16 years old 0.017632 Assembly and Fabrication workers Cars greater equal than workers hh and >= 16 years old 0.023838 Business and financial operation workers < 16 years old Business and financial operation workers No cars hh and >= 16 years old 0.001039 Business and financial operation workers Cars less than workers hh and >= 16 years old 0.004179 Business and financial operation workers Cars greater equal than workers hh and >= 16 years old 0.021334 Construction workers < 16 years old Construction workers No cars hh and >= 16 years old 0.004579 Construction workers Cars less than workers hh and >= 16 years old 0.011428 Construction workers Cars greater equal than workers hh and >= 16 years old 0.024867 … … … Service workers < 16 years old Service workers No cars hh and >= 16 years old 0.007668 Service workers Cars less than workers hh and >= 16 years old 0.019563 Service workers Cars greater equal than workers hh and >= 16 years old 0.030693 Transport workers < 16 years old Transport workers No cars hh and >= 16 years old 0.004985 Transport workers Cars less than workers hh and >= 16 years old 0.014249 Transport workers Cars greater equal than workers hh and >= 16 years old 0.025568

Values

SLIDE 34

Labor commodities

Commodity Name person type total amount of commodity provided model area wide per year visits model area wide per day amount of commodity

btained per visit

travel segment skim $ / year visits / year $ / visit Agriculture workers < 16 years old

Agriculture workers

No cars hh and >= 16 years old 490,930,550 7,474,650 65.68 Agriculture workers Cars less than workers hh and >= 16 years old 1,519,955,164 22,527,755 67.47 Agriculture workers Cars greater equal than workers hh and >= 16 years old 2,049,152,810 26,251,890 78.06 Assembly and Fabrication workers < 16 years old

Assembly and Fabrication workers

No cars hh and >= 16 years old 2,471,778,060 28,408,415 87.01 Assembly and Fabrication workers Cars less than workers hh and >= 16 years old 7,912,527,086 90,928,015 87.02 Assembly and Fabrication workers Cars greater equal than workers hh and >= 16 years old 22,170,554,890 176,977,640 125.27 Business and financial operation workers < 16 years old

Business and financial operation workers

No cars hh and >= 16 years old 867,552,420 5,684,900 152.61 Business and financial operation workers Cars less than workers hh and >= 16 years old 3,711,204,350 23,581,495 157.38 Business and financial operation workers Cars greater equal than workers hh and >= 16 years old 29,418,376,914 150,010,860 196.11 Construction workers < 16 years old

Construction workers

No cars hh and >= 16 years old 1,953,097,450 18,596,430 105.03 Construction workers Cars less than workers hh and >= 16 years old 4,558,961,604 44,886,660 101.57 Construction workers Cars greater equal than workers hh and >= 16 years old 17,540,228,744 129,873,920 135.06 … … … … … Service workers < 16 years old

Service workers

No cars hh and >= 16 years old 2,095,685,238 32,216,600 65.05 Service workers Cars less than workers hh and >= 16 years old 5,712,840,150 84,962,650 67.24 Service workers Cars greater equal than workers hh and >= 16 years old 21,357,512,416 205,768,640 103.79 Transport workers < 16 years old

Transport workers

No cars hh and >= 16 years old 1,525,451,890 17,959,465 84.94 Transport workers Cars less than workers hh and >= 16 years old 5,263,275,060 56,398,105 93.32 Transport workers Cars greater equal than workers hh and >= 16 years old 17,485,910,102 137,700,680 126.98

Variables

SLIDE 35

Worker delivered service commodities

Tranc,j,k = κSp,c · LogSump,j,k

κSp,c = $ value of transport utility per unit of commodity per unit of logsum

utility for person trip = 2/((util / $) · (units of commodity per visit))

Units of commodity per visit= (total amount of commodity from aggregate economic flow table) / (total trips per day * total days per year ) util / $ means transport money cost coefficient which comes from mode choice estimation of travel model.

SLIDE 36

Worker delivered service commodities

Commodity Name κSp,c $ / $-util Agriculture Services output 0.004354 Agriculture Management 0.004354 Mining and Extraction Management 0.000807 Construction Management 0.004695 Manufacturing Management 0.000179 Transportation Service output 0.004095 Utilities and Communications Management 0.000748 Wholesale output 0.000182 Finance Insurance Legal output 0.000995 Real Estate output 0.000444 Hotel output 0.005899 Professional services output 0.004603 Govt Enterprises output 0.001785 Military output 0.003560 Federal Governance output 0.001947 State and Local Governance output 0.001947 Media output 0.005731 Industrial based services 0.022267 Retail and Restaurant Management 0.001596 Office based services 0.000453 Onsite Business services 0.000453 Onsite personal services 0.000453

Values

SLIDE 37

Worker delivered service commodities

Commodity Name total amount of commodity produced model area wide per year visits per day amount of commodity delivered per visit $ / year visits / day $ / visit Agriculture Services output 4,833,976,000 31,970 579 Agriculture Management 12,523,377,139 82,825 579 Mining and Extraction Management 6,281,763,729 7,704 3124 Construction Management 53,225,012,107 379,589 537 Manufacturing Management 303,928,422,327 82,651 14089 Transportation Service output 34,842,441,529 216,763 616 Utilities and Communications Management 58,246,002,226 66,176 3372 Wholesale output 146,539,990,482 40,599 13829 Finance Insurance Legal output 172,245,468,877 260,436 2534 Real Estate output 218,359,368,677 147,304 5680 Hotel output 15,622,712,581 140,013 428 Professional services output 99,356,094,669 694,833 548 Govt Enterprises output 15,977,640,506 43,316 1413 Military output 18,131,690,000 98,056 708 Federal Governance output 15,571,990,000 46,048 1296 State and Local Governance output 102,657,860,000 303,573 1296 Media output 39,655,530,940 345,269 440 Industrial based services 11,092,913,175 375,240 113 Retail and Restaurant Management 57,731,038,321 140,013 1580 Office based services 102,775,986,706 70,700 5570 Onsite Business services 33,348,542,871 22,941 5570 Onsite personal services 12,682,981,900 8,725 5570

Variables

SLIDE 38

Demargining IMPLAN

Setup of AA production module

SLIDE 39

Margin Issues in IMPLAN

In IMPLAN households’ consumption, households purchase goods from both producers and retailers.

EX. Households buy food from Agricultural Plants
utput, Agricultural Animal outputs Retail General

Merchandise outputs, and Retail Unprepared food outputs. In reality, household buy food from Retail General Merchandise outputs, and Retail Unprepared food outputs. Thus, the total consumptions of households are

verestimated for 10 goods while underestimated for 3

retailer goods. They should be demargined so that households purchase the goods from the right retailers.

SLIDE 40

Goods to be demargined

Agriculture Plants output Agriculture Animals output Agriculture Forestry and Fishing output Mining and Extraction output Fuels Manufacturing Food output Manufacturing Textiles output Manufacturing Petro Paper Chemicals Plastic Rubber Glass Cement output Manufacturing Wood Products Printing Furniture Misc output Manufacturing Metal Steel Machinery output

SLIDE 41

Household Consumption Margins in IMPLAN

SLIDE 42

Household Consumption Demargin in PECAS

SLIDE 43

Method of Demargining Goods

SLIDE 44

Method of Demargining Goods

SLIDE 45

Method of Demargining Capital

SLIDE 46

Recent Software Improvements

Data loader/map maker Density Shaping Functions Floorspace Synthesizer performance/functionality

SLIDE 47

Data Loader / Map Maker

AA Data Loader Spatial Database PostgreSQL/PostGIS SQL Server Spatial Data selector/ Crosstabber Spatial Comparison Views Web map preview Shape File Arc Map Map Info QGIS Map Server Google Earth Every thing Arc SDE

WFS/WMS WFS/WMS

SLIDE 48

SLIDE 49

SLIDE 50

SLIDE 51

SLIDE 52

SLIDE 53

SLIDE 54

SLIDE 55

Data Loader / Map Maker

AA Data Loader Spatial Database PostgreSQL/PostGIS SQL Server Spatial Data selector/ Crosstabber Spatial Comparison Views Web map preview Shape File Arc Map Map Info QGIS Map Server Google Earth Every thing Arc SDE

WFS/WMS WFS/WMS

SLIDE 56

Density Shaping Functions

Want to control intensity of development with finer detail than desirable with space-type categories. Modifiers to account for different intensities

A-spatial (i.e. not parcel specific) Not specific about whether they are rent modifiers (demand) or cost modifiers (supply) Calibration parameters to match observed distributions

SLIDE 57

From the Parcel and the Construction Cost Calculation System Space type specific Density Shaping Modifiers

SLIDE 58

Continuous utility function

SLIDE 59

SLIDE 60

SLIDE 61

Improvements to Floorspace Synthesizer

Database related performance improvements

Each thread uses separate connections to database
Allows database to operate in parallel too
Consolidated database queries
One per TAZ instead of multiple per parcel
Moved database updates into separate threads

New features

Added the option to allow the synthesizer to assign

space to parcels in neighboring TAZs.

SLIDE 62

Future Improvements

Expand the flexibility of the match coefficient table

FAR (Intensity) targets modified by TAZ

Retrofit the synthesizer to output the SD database structure

Less manual transformation

Add repeatable random numbers to break ties for parcels with equal suitability

Random utility theory Better visuals

SLIDE 63

Construction Capacity Theory

SD Module is constrained by the amounts of construction produced by AA

SD considers the full list of parcels in random order Rates of development are compared and adjusted during processing to arrive to a result consistent with AA constraint.

high rates of construction increases the construction cost high construction costs  lowers the rates of construction

Amounts of space produced by construction activities

AA SD

estimates in year t feeds SD in year t+1

SLIDE 64

Construction Capacity Software

SD implements Capacity

Constraints function in V2.4

Two tables are added to the

database

4 fields were added to the

space_types_i table

SLIDE 65

Upcoming Work

SD Disaggregate Estimation PUMS Cluster Analysis Running the Floorspace Synthesizer

SLIDE 66

Nested logit structure

No change Demolish Derelict Add space New space type Quantity Quantity

multi-level nested discrete-continuous logit

Renovate

SLIDE 67

SD Disaggregate Estimation

Purchased statewide development permit data Use PECAS SD to assign a construction cost and modified rent to each record in the dataset

For each potential space type Based on parcel attributes Aggregated to grid cells

Probably most difficult part of work

Also to a selection of unchanged grid cells Set up logit model estimation software to estimate dispersion parameters/constants U = Rent –Θcost, get future rent discount and dispersion parameters

SLIDE 68

PUMS Cluster Analysis

Middle level of AA is choice of “technology” Primarily options regarding occupation, work/no work, quantity of space and type of structure Need to span full set of occupations and space types

To have appropriate elasticities in labor and space markets

Currently using full combinatorial options Cluster observed PUMS records “Caricature” them

SLIDE 69

technology selection buying and selling allocations production allocation production activity to zones production of commodities consumption of commodities produced commodities to selling locations consumed commodities to buying locations

3-level nested logit

SLIDE 70

PUMS Records in 2 Dimensions

Labor Quantity Housing Quantity Combinatorial Options Span available space Activities in AA can “blend” the options to select middle ground

SLIDE 71

Clustered Options

Labor Quantity Housing Quantity Less options (3 instead of 4, much more important with more than 2 dimensions!) “Caricature” the options to span space Move them away from the mean to make them more extreme Activities in PECAS blend options to select less extreme versions

SLIDE 72

Running the Floorspace Synthesizer

Setup match coefficients Recent experience here in San Diego provides more guidance Draft report from San Diego here, available soon System for rating the appropriateness of grid cells different types of space and quantities of space

Quantities and type by TAZ have been developed from population, employment and other sources, as described.

Simple in concept, but can get confusing because all ratings are relative to each other