Models : The Florida Experience

frequentverseUrban and Civil

Nov 16, 2013 (3 years and 9 months ago)

98 views

presented by

Krishnan Viswanathan

Cambridge Systematics, Inc.



Co
-
authors

Vidya Mysore, Florida Department of Transportation

Nanda Srinivasan, Cambridge Systematics, Inc.


May 7, 2007

Using LEHD Data for Travel Demand
Models : The Florida Experience

1

Outline

Motivation

Longitudinal Employer Household Dynamics (LEHD)
Description

Some Comparisons

FSUTMS Data Analysis

Summary

Future Directions

2

Motivation


Review and possible use of LEHD and Employment
data for travel demand modeling


Current and future directions of Census and other available
data sources for transportation

Review of QCEW [Quarterly Census of Employment and
Wages] (Employment data from AWI, state agency) and LEHD
Data

Data integration


3

LEHD Description

LEHD combines federal and state administrative data on employers
and employees with core Census Bureau censuses and surveys

Local Employment Dynamics (LED) is a voluntary partnership
between state labor market information agencies and the U.S.
Census Bureau to develop new information about local labor
market conditions

4

LEHD Description

The data for LEHD will be based on Quarterly Census of
Employment and Wages (QCEW/ES
-
202)

Number of people employed and the wages paid to the
employees each quarter

Multi
-
establishment employers also provide individual worksite
data

Preliminary annual data are available in June and final annual
data are available in October of each year. Quarterly data are
available six months after the end of each quarter

5

LEHD Description

Origin Destination Database

Work Profile


block group



# establishments



# workers



Worker distribution



age



earnings



industry



Demand/ Growth Indicators



Job creation/loss



Hires/separations



Earnings hires/separations


Home Block

(U/I Wage Data)

Workplace Block

(QCEW Data)

# of Workers (primary job) and # of Jobs

Home Profile


block group



# workers



Worker distribution



age



earnings



industry

Acknowledgments : US Census Bureau

6

LEHD Description

Origin Destination Database

Year

h_geocodefull

w_geocodefull

jobs_all

jobs_prim

jobs_priv_all

jobs_priv_prim

Year

YYYY

FIPS

state,

FIPS

county
;

Census

tract,

Census

block
;

home

FIPS

state,

FIPS

county
;

Census

tract,

Census

block
;

work

Number

of

jobs

to

which

workers

made

this

trip

Number

of

primary

jobs

to

which

workers

made

this

trip

Number

of

private

sector

jobs

to

which

workers

made

this

trip

Number

of

private

sector

primary

jobs

to

which

workers

made

this

trip

7

LEHD Description

OD with ACS: Possible Model for Integration

Home Block

(U/I Wage Data)

Workplace Block

(QCEW Data)

# of Workers (primary job) and # of Jobs

Home Profile


block group



# workers



Worker distribution



age



earnings



industry

Work Profile


block group



# establishments



# workers



Worker distribution



age



earnings



industry



Demand/ Growth Indicators



Job creation/loss



Hires/separations



Earnings hires/separations


ACS Data



Trip Time



Mode



Distance


Acknowledgments : US Census Bureau

8

Some Comparisons

An old well reviewed favorite


CTPP

New kid on the block


LEHD

North East Regional
Planning Model (NERPM)
region


Duval, St. Johns,
Clay and Nassau Counties
(Jacksonville is major city)

CBD



CTPP

LEHD

CTPP

1.00



LEHD

0.68

1.00







9

Some Comparisons

Travel Statistics








Region Travel

CTPP

JOBS_PRIMARY

Total : All resident workers in NERPM Region


519,280


500,702

Travel within NERPM region


509,105


385,951

Travel from NERPM region to CBD


49,460


46,366

Travel outside NERPM region


10,175


114,751







98 percent (CTPP) vs. 77 percent (LEHD) travel within region


Similar results in an earlier BTS Study by Cesar Singh

Under a quarter of LEHD travel is outside the region

Travel shares to CBD is approximately the same

10

Some Comparisons

Residual Plot

Total Workers Residual Plot
-1000
-800
-600
-400
-200
0
200
400
600
800
0
200
400
600
800
1000
1200
1400
1600
1800
Total Workers (CTPP Table T301C1)
Residuals
11

Some Comparisons

CBD Distance

ctppworkers
0 to 5 miles
5 to 10 miles
10 to 15 miles
15 to 20 miles
20 to 25 miles
25 to 30 miles
More than 30 miles
CTPP

jobs_prim
0 to 5 miles
5 to 10 miles
10 to 15 miles
15 to 20 miles
20 to 25 miles
25 to 30 miles
More than 30 miles
LEHD


Primary Jobs

jobs_prim
0 to 5 miles
5 to 10 miles
10 to 15 miles
15 to 20 miles
20 to 25 miles
25 to 30 miles
More than 30 miles
12

Some Comparisons

CBD Distance

Distance from CBD - CTPP vs LEHD
0%
5%
10%
15%
20%
25%
30%
35%
40%
45%
0 to 5
miles
5 to 10
miles
10 to 15
miles
15 to 20
miles
20 to 25
miles
25 to 30
miles
More than
30 miles
Distance
CTPP
JOBS_PRIMARY
13

FSUTMS Data Analysis

Models Used

QCEW Data Analysis

Comparison between InfoUSA and QCEW data

Results

14

Models Used

2005 NERPM CF Model

2000 Polk TPO Model

15

QCEW (ES202) Data Analysis

December 2004 data used for analysis

Data overlaid to TAZ

Geocoded the data for address matching

ZDATA2 data developed from QCEW Data


Model


Total QCEW
Employment

Lat/Long
Information
Available

Geocoded
to Model
TAZ

Address
Matched to
TAZ

NERPM

728,044

487,954

468,755


176,957

Polk

438,190

182,675

169,771


167,370

16

Comparison between InfoUSA and QCEW Data

NERPM

QCEW

InfoUSA

Model

MAN. INDUSTRIAL


113,003

89,318

54,903

OTHER INDUSTRIAL


34,699

42,644

137,255

COMMERCIAL


145,355

126,640

182,157

SERVICE


352,655


274,766

420,058

TOTAL

645,712

533,368

794,373

POLK

QCEW

InfoUSA


Model (2000)

INDUSTRIAL


41,133

35,761


41,684

COMMERCIAL


179,468

58,486


59,775

SERVICE


116,540

99,418


105,611

TOTAL


337,141


193,665


207,070

17

Results

Model HBW Trips


LEHD Primary Jobs


18

Results

Model HBW Trips


CTPP JTW Trips


19

Results

Model HBW Trips


LEHD Primary Jobs


20

Results

Model HBW Trips


CTPP JTW Trips


21

Summary

Under reporting of internal trips

Model results are reasonable

QCEW data good source for employment data

Issues regarding data suppression



22

Future Directions

Potential data source for model calibration, more review of long
distance trips to work

Coordination effort for data aggregation with AWI

Work with Census Bureau to produce LEHD for Florida model

Variability of data over time

23

Acknowledgments

Jim Baxter, Modeling Manager, FDOT District 1

Milton Locklear, Modeling Manager, FDOT District 2

Kathie Hughes, David Thurmond, AWI

Pam Schenker, ex
-
AWI