Your Federal Quarterly Tax Payments are due April 15th Get Help Now >>

Estimating Through Trips Using a Two-Phased Logit Model by vSTnB6

VIEWS: 0 PAGES: 20

									Workshop on Current and Emerging Practices in Collecting and Estimating
External Travel
TRB 90th Annual Meeting
23 January 2011
Background


                Texas Department of Transportation
                 (TxDOT) Travel Survey Program
                    Commercial Vehicle
                    Workplace
                    Household
                    External
                Research Project 0-6583
                 “Alternative Methods for Developing
                 External Travel Survey Data”


                                                       2
Research
Background


                Methods Reviewed
                  Data from other Surveys
                   • Commercial Vehicle
                   • Workplace
                   • Household
                  Statewide Model Extrapolation
                  Synthetic Models




                                                   3
Research Purpose


                                             Research Purpose: Develop an
                                              alternative to external surveys for
                                              estimating through travel that is
                                              applicable to urban areas in Texas.

                                             Research Approach: Develop a
                                              statistical model using data from
            MPO Boundary
            Network
                                              external surveys previously
            Intercept Survey Location

            High Volume Survey Location

                                              administered in Texas.


                                                                                    4
Model Overview


This research developed a
system of two models that can
be used to estimate through
travel
       1. The first model
           estimates the
           proportion through
           trips at an external
           station
       2. The second model
           estimates the
           distribution of
           through trips
           between external
           stations




                                  5
Model Overview


This research developed a
system of two models that can
be used to estimate through
travel
       1. The first model
           estimates the
           proportion through
           trips at an external
           station
       2. The second model
           estimates the
           distribution of
           through trips
           between external
           stations




                                  6
Data Collection


                      Study Area        Year   Survey       Total
                                               External   Responses
                                               Stations
                  Abilene               2005      11          3,329
                  Amarillo              2005      12          4,234
                  Austin                2005      22          8,298
                  Dallas – Fort Worth   2005      32         12,642
                  Longview              2004      30          8,426
                  Lubbock               2005      17          3,988
                  Midland – Odessa      2002      13          4,023
                  San Angelo            2004      11          4,031
                  San Antonio           2005      22          9,892
                  Sherman – Denison     2005      10          3,975
                  Tyler                 2004      18          5,124
                  Waco                  2006      15          4,557
                  Wichita Falls         2005      11          3,093

                                                                      7
Model
Development
Method


               Select candidate predictor
                variables
               Refine the selection using forward
                selection
               Check the model using model
                diagnostics
               Choose subset of variables to form
                final model
               Investigate variable interactions
                and transformations

                                                     8
Model 1 Details

                          exp[ g (x j )]
        Pj (Y  1) =
                        1  exp[ g (x j )]

where
                 j = indexes the external station for which the estimation is being made,
               Y = a variable for the coded response, where Y=1 for a through trip response,
                       and Y=0 for a non-through trip response,
        Pj Y  1 = the probability of a through trip at external station j ,

           g ( x j ) =  1.78  2.38 PINTTH j  0.235 INTTL1 j for commercial vehicles,

           g ( x j ) = 2.94  2.38PINTTH j  0.235 INTTL1j for non-commercial vehicles,

        PINTTH   j   = the proportion of through trips at the external station as predicted by the

                       interaction score, and
        INTTL1 j = a binary variable which indicates if the external station has any interaction

                       scores.
                                                                                                     9
Model Interaction
Score
                                                                         10



                                           Pv  Pw               
             1      INTij      6                              
                                                10  D 2  f vwij  
                                         
                           {vU , v  w}  wU        vw          

 250 Miles




                                           50 Miles
 Model 2 Details

                                   exp[ g (x sj )]
         Pj Y  s  
                                       exp[g (x
                         q 1, 2 ,, j 1, j 1,, S 1, S
                                                             qj   )]

where
             g (x ij )  2.24 PINT1ij  1.09TURN02 ij  0.572 PADTLG ij  0.814 ROUTE ij .


           PINT1ij  A binary variable that indicates if external station j has any interaction scores with a route
                    entering at external station i ;
        TURN02 ij  the square root of the number of turns on the least time route between external station i and
                    external station j ;
        PADTLG ij  the natural logarithm of the ADT at external station i as a portion of the total ADT across all
                    entry external stations; and
         ROUTE ij  a binary variable that indicates if the least time route between external station i and external
                         station j is valid.




                                                                                                                       11
Route Validity




       Urban area

       External station

       Study area boundary




                             12
Model Evaluation


    Study Area Groups

Group 1
Amarillo, San Antonio, and                    13 Study Areas
Waco

Group 2
Austin, Lubbock, and Sherman-
Denison                         Group       Group       Group        Group
                                  1           2           3            4
Group 3
Dallas-Fort Worth, Longview,
San Angelo, and Wichita Falls

Group 4                                              Fit the Model
Abilene, Midland-Odessa, and
Tyler


                                   Fit the Model (repeat)

                                                                             13
Model Evaluation
                                  14



Through Trip Split (Percentage)
Model Prediction Errors for
Each Quartile of the Prediction
Values
Model Evaluation
                               15



Distribution of Model
Prediction Errors for Each
Quartile of the Prediction
Values (for Observations not
equal to zero)
Model Application

                                                                                          Exit Zone
                                                                                                                                                                        Total
                        350   351   352   353   354   355   356   357   358   359   360     361       362   363   364   365   366   367   368   369   370   371   372
                  350   0     0     0     0     0     0     0     0     0     0     0        2        0     9     1     0     0     0     0     0     0     0     0      13
                  351   0     0     3     21    77    6     0     0     355   0     0        51       0     9     0     0     0     0     0     0     0     0     0     522
                  352   0     3     0     0     8     3     0     0     9     0     0        7        0     2     2     0     0     0     0     6     0     0     0      39
                  353   0     21    0     0     2     0     0     0     17    0     0        16       0     2     2     0     0     0     0     7     0     0     0      65
                  354   0     77    8     2     0     4     0     0     6     0     0        33       0     19    0     0     0     0     0     9     0     0     0     157
                  355   0     6     3     0     4     0     0     0     3     0     0        2        0     1     0     0     0     0     0     0     0     0     0      20
                  356   0     0     0     0     0     0     0     0     0     0     0        3        0     0     0     0     0     0     0     0     1     0     0       4
                  357   0     0     0     0     0     0     0     0     6     0     0        2        0     0     0     0     0     0     0     0     0     0     0       8
                  358   0     355   9     17    6     3     0     6     0     0     0        7        0     8     0     0     0     0     0     0     0     0     0     411
                  359   0     0     0     0     0     0     0     0     0     0     0        0        0     0     0     0     0     0     0     0     0     0     0       0
  Entry Zone




                  360   0     0     0     0     0     0     0     0     0     0     0        0        0     0     0     0     0     0     0     0     0     0     0       0
                  361   2     51    7     16    33    2     3     2     7     0     0        0        0     2     0     0     0     0     0     2     0     0     0     129
                  362   0     0     0     0     0     0     0     0     0     0     0        0        0     0     0     0     0     0     0     0     0     0     0       0
                  363   9     9     2     2     19    1     0     0     8     0     0        2        0     0     2     0     0     0     0     0     0     0     0      54
                  364   1     0     2     2     0     0     0     0     0     0     0        0        0     2     0     0     0     0     0     0     0     0     0       7
                  365   0     0     0     0     0     0     0     0     0     0     0        0        0     0     0     0     0     0     0     0     0     0     0       0
                  366   0     0     0     0     0     0     0     0     0     0     0        0        0     0     0     0     0     0     0     0     0     0     0       0
                  367   0     0     0     0     0     0     0     0     0     0     0        0        0     0     0     0     0     0     0     0     0     0     0       0
                  368   0     0     0     0     0     0     0     0     0     0     0        0        0     0     0     0     0     0     0     0     0     0     0       0
                  369   0     0     6     7     9     0     0     0     0     0     0        2        0     0     0     0     0     0     0     0     0     0     0      25
                  370   0     0     0     0     0     0     1     0     0     0     0        0        0     0     0     0     0     0     0     0     0     0     0       1
                  371   0     0     0     0     0     0     0     0     0     0     0        0        0     0     0     0     0     0     0     0     0     0     0       0
                  372   0     0     0     0     0     0     0     0     0     0     0        0        0     0     0     0     0     0     0     0     0     0     0       0
               Total    13    522   39    65    157   20    4     8     411   0     0       129       0     54    7     0     0     0     0     25    1     0     0     1,455



                                                                                                                                                                                16
Sample Results

                         Commercial                  Non-Commercial              Total (com+non-com)
  External Station                        % Diff                        % Diff                         % Diff
                     Survey      Model             Survey      Model             Survey       Model
        427           721          696     -3.47    530         508      -4.15   1,251        1,204     -3.76
        428           100          136     36.00    162         188      16.05    262          324      23.66
        429            0             0      N/A      0           0        N/A      0             0       N/A
        430            0             6      N/A     23           18     -21.74    23            24       4.35
        431            0             0      N/A      0           12       N/A      0            12       N/A
        432            0             0      N/A      0           6        N/A      0             6       N/A
        433           86           202    134.88    178         432     142.70    264          634     140.15
        434            0            16      N/A      0           36       N/A      0            52       N/A
        435          1,595         938    -41.19   1,160        554     -52.24   2,755        1,492    -45.84
        436           419          738     76.13    296         404      36.49    715         1,142     59.72
        437            0             6      N/A     35           20     -42.86    35            26     -25.71
        438            0             2      N/A      0           2        N/A      0             4       N/A
        439            0             0      N/A      0           0        N/A      0             0       N/A
        440           668          480    -28.14    638         486     -23.82   1,306         966     -26.03
        441           130           12    -90.77    52           28     -46.15    182           40     -78.02
        442            0             6      N/A      0           12       N/A      0            18       N/A
        443           257          454     76.65    253         476      88.14    510          930      82.35
        444            0             6      N/A      0           8        N/A      0            14       N/A
        445            0             0      N/A      0           0        N/A      0             0       N/A
        446            0             0      N/A      0           0        N/A      0             0       N/A
        447          1,231        1,324     7.55   1,056        822     -22.16   2,287        2,146     -6.17
        448           16            36    125.00    80           48     -40.00    96            84     -12.50
        449            0             2      N/A      0           8        N/A      0            10       N/A
        450            0             0      N/A      0           0        N/A      0             0       N/A
       Total         5,223        5,060   -3.12    4,463        4,068   -8.85    9,686        9,128    -5.76
                                                                                                                17
Results


                          Commercial           Non-Commercial      Total (com + non-com)
 Study Area                                                                                % Diff
                     Survey       Model     Survey       Model     Survey         Model
 Abilene              7,569        5,784     5,539        9,072    13,108        14,856    13.34
 Amarillo             5,223        5,060     4,463        4,068     9,686         9,128    -5.76
 Austin              18,713       15,964    31,838       26,032    50,551        41,996    -16.92
 Dallas/Fort Worth   20,609       15,864    19,139       21,404    39,748        37,268    -6.24
 Longview            13,944       18,544    21,650       23,980    35,594        42,524    19.47
 Lubbock              2,001        1,824     2,240        3,308     4,241         5,132    21.01
 Midland/ Odessa      2,555        3,208     956          2,452     3,511         5,660    61.21
 San Angelo           1,497        1,664     2,908        3,732     4,405         5,396    22.50
 San Antonio          9,710        9,816    23,570       14,940    33,280        24,756    -25.61
 Sherman/ Denison     4,727        8,212    15,423       13,616    20,150        21,828     8.33
 Tyler               10,095       13,596    29,752       20,020    39,847        33,616    -15.64
 Waco                11,833       14,896    35,273       26,336    47,106        41,232    -12.47
 Wichita Falls        3,063        4,284     5,437       10,416     8,500        14,700    72.94
 Total               111,539      118,716   198,188      179,376   309,727       298,092   -3.76


                                                                                                    18
Conclusions


                 Model 1 – variables based on
                  interaction scores
                 Predictions by vehicle type
                 Model 2 – use of turn and route
                  variables
                 Readily available data
                 Only provides external through
                  estimates

                                                    19
To Learn More


Alternative Methods for
Developing External Travel          Eric Talbot
Survey Data
                                     Resource Systems Group, Inc.
TTI Technical Report 0-6583-1        eric.talbot@rsginc.com
http://tti.tamu.edu/documents/
0-6583-1.pdf                         Estimating Through Trip Travel Without
                                     External Surveys (11-2832)
                                     Wednesday Poster Session – Hilton Intl.
                                     Ballroom (9:30 a.m – 12:00 p.m.)

                                    Steve Farnsworth
                                     Texas Transportation Institute
                                     s-farnsworth44@tamu.edu

                                                                               20

								
To top