<%BANNER%>

Development of a Rural Freeway Level of Service Model Based upon Traveler Perception

xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E20101130_AAAACS INGEST_TIME 2010-11-30T16:28:26Z PACKAGE UFE0010297_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES
FILE SIZE 24109 DFID F20101130_AABVIS ORIGIN DEPOSITOR PATH kirschner_d_Page_31.QC.jpg GLOBAL false PRESERVATION BIT MESSAGE_DIGEST ALGORITHM MD5
be59bef7122783de08d12081d77482bb
SHA-1
86c8a89a30467bc4d0db864b546f780af81fbb9c
2006 F20101130_AABVDU kirschner_d_Page_21.txt
4c919603abdd7918c7fca41ac01c02e7
48eeaf852ea7182ed7ef5682feae276bedb1e4d5
25271604 F20101130_AABUWX kirschner_d_Page_05.tif
d5fe81f0c0bf62110a6e89c8e90f6d56
0d5751c6cb366c368de2aafa8b96126425b91885
22913 F20101130_AABVIT kirschner_d_Page_34.QC.jpg
6e7769b4b5a64562d6da717dc2e2bffc
b6792dc3da95e9ff2e13dede942ce0d213f09987
2011 F20101130_AABVDV kirschner_d_Page_22.txt
dd6451f2acf913ebfdcc2ff8a174f6c6
f8992271984112ecc5e79ef49f6ab33b73569892
104058 F20101130_AABUUA kirschner_d_Page_12.jp2
42dd3e13cd0afded908d9515cbc014c7
cfc1f833d9cf1c1e55b362685626bef6cc2c439d
F20101130_AABUWY kirschner_d_Page_06.tif
066e6bc45cca7682557e200ffdee69e8
a55780e60cdc47509072543ea64ad1a8e8e72834
3234 F20101130_AABVIU kirschner_d_Page_43thm.jpg
e0fa1b9aabb7aceaa1b48c0669e33103
f1c7183cd2da6349cef9a723557c80bcd96dac5b
1949 F20101130_AABVDW kirschner_d_Page_23.txt
1a100d978edd9105650e3ab395a479f5
0bcd78ef8e86057841afed14c978e201cc226257
78807 F20101130_AABUUB kirschner_d_Page_13.jp2
8d2e360759640adb72740cbf179800c4
3a53a275249b27d783bff5cd1ce942450b08cc19
F20101130_AABUWZ kirschner_d_Page_07.tif
b4af81ff7dffa187cde5e5971d5281c9
785410f82f271dae94824e2839ba94560a48007b
1274 F20101130_AABVDX kirschner_d_Page_24.txt
28cee20d966eb7594eb83c7d84a87b14
a6b70d288a5bd1b1fb7c5e0c100bd70165eef7fe
6006 F20101130_AABVIV kirschner_d_Page_50thm.jpg
f750279c6746d1cce8f858590db4750c
653a0f579eca6a1c8292fceb6878def7b2a88bcd
F20101130_AABUZA kirschner_d_Page_60.tif
0ed60ec46edde562bbd19072eb36bdd4
01b6368efe52284ecb111c8075211d5548fbc0a0
50127 F20101130_AABVBA kirschner_d_Page_31.pro
c3353077a5cd822e22a74c415b1929f3
e566dc02bc28d27818112c46551f1f3fdbda168e
1847 F20101130_AABVDY kirschner_d_Page_25.txt
90314cdcbbe2fb338e086c86dcae93ab
056ede9a772d75546a166f422de891de76dd763e
10761 F20101130_AABUUC kirschner_d_Page_14.jp2
5ee87bab90d8c699a9b6c5cc67876379
710e064be0d8203663deeb44da7fe7b6889f4c44
17816 F20101130_AABVIW kirschner_d_Page_52.QC.jpg
7801afd91fd94d7dd3b2e66971c5afb2
762c83450ff535d0010fc7e47bd2f006eccaeb41
F20101130_AABUZB kirschner_d_Page_61.tif
baf1d46ad52f3678d71d2ccc3ad94bdd
d7c8024c3e99251e0d1434e5bc5b5ab914f35a28
22961 F20101130_AABVBB kirschner_d_Page_32.pro
1f643e891e6bd6cf732b7a19ec1fe2b6
d76e6133148cb5ee500842dd1303e7e1d65eb773
1867 F20101130_AABVDZ kirschner_d_Page_26.txt
3d9eed48b93f35800c869f5be8bd3d9e
5dcb9c9255119b1c69dc0951f8d44d5e7b14812b
87232 F20101130_AABUUD kirschner_d_Page_15.jp2
e7d4b4cdd6f0b66e807ce5c83ce5a040
4cdef84c161da30609a84f1ec428598304976271
5675 F20101130_AABVIX kirschner_d_Page_17thm.jpg
b7ad8093fcc7e37831f0d100aaa5e736
0656b8f1fcd7f90303e4740f29a76e7fa10cd5c7
F20101130_AABUZC kirschner_d_Page_62.tif
07145e08fca48a1d82253362fb33eccb
211c7e92de02c14d63944e031035f53f642f7014
23712 F20101130_AABVBC kirschner_d_Page_33.pro
13f5e01dee63d6f97022ee9483180df3
15372c120a0de119514e6f33bd3d486e0d20363b
109431 F20101130_AABUUE kirschner_d_Page_16.jp2
b155cf8625747169806892e011899a61
9ce3c08ea25ed8050558188df202b9df4017b194
1803 F20101130_AABVIY kirschner_d_Page_03thm.jpg
8ba4f5c99dcbcf005f1f2e7c3c0f220c
135eefe6c585d3a62201d7fcfdea451758494732
1941 F20101130_AABVGA kirschner_d_Page_79.txt
912805c00fb3f08873c4070ee3b98e9b
0fdca0b1bf8c5c571cd104c9a540883129306058
F20101130_AABUZD kirschner_d_Page_63.tif
1c7310c39baa4dca11b10480132ed821
7d2746203184afc9a0cacae4e92ef39bf4880bd0
46882 F20101130_AABVBD kirschner_d_Page_34.pro
3ea86d9e2b3961297f1bad3493fb3c28
b1a9d9069f3422c0dd2892183bd903ffde74d797
83651 F20101130_AABUUF kirschner_d_Page_17.jp2
c8010746e8e4f6d4526794e41151854d
821a7850e87fdf99c9f560d17eaf166792922044
18250 F20101130_AABVIZ kirschner_d_Page_49.QC.jpg
95bad9363303ce1101f61429e1872b4c
f8312025f712cfc892823d3d63dfc34a482c5217
1341 F20101130_AABVGB kirschner_d_Page_80.txt
be00f76e83e72738c2aad51119e09554
56566f6d27604383616854ac4a929e293b434004
1053954 F20101130_AABUZE kirschner_d_Page_64.tif
d92978ff53ff94516f8a5c90cea8e286
0761f6083810c73bd8dae1ab9c5cc839e32b11e3
7623 F20101130_AABVBE kirschner_d_Page_35.pro
bb4c66dacf6a120f6ad49026e35946cf
12300576d498d79f17e793c5e38ea6f940fc27b4
112277 F20101130_AABUUG kirschner_d_Page_18.jp2
287e3495d6a3a74f04a5cb7f600e16bb
9f964b70bb5d9fbed70adc1201ef9aa508c41d2d
1365 F20101130_AABVGC kirschner_d_Page_81.txt
69174dfbb106fc73ed6a45a5f19f085e
123ac954ddf0675db7dc4bbcde1f89b308f6848a
F20101130_AABUZF kirschner_d_Page_65.tif
f1815de2d3838df571c8dc836d950c34
56cacbfbfa9027e8a97a3638b9b04e60a495409f
50474 F20101130_AABVBF kirschner_d_Page_36.pro
6453ff78b65625cec956a874d5f94862
b692e7c7410ee19d2f6f23ac039c267b0e6d7923
109352 F20101130_AABUUH kirschner_d_Page_19.jp2
4b7e7d641c165da7c27bbf57c11ff2d9
f46fcc4a0ddc2383597aaff289996d8ae2d8600f
19296 F20101130_AABVLA kirschner_d_Page_28.QC.jpg
080ba775e91d8003c8f58ff89fd053ce
06c54ef12815ddde5bd8641168d4ab23b9e68784
392 F20101130_AABVGD kirschner_d_Page_82.txt
2823dcaaa59f582b6880e15629d66c30
3b26ca673d05d4f02d5736dadeb29300484f3ca8
F20101130_AABUZG kirschner_d_Page_66.tif
a17fa6cd2de2424b724edfdc6a6de3d1
9f6edc4f289ce7207e1b4fc6991538ef866f11b4
12097 F20101130_AABVBG kirschner_d_Page_37.pro
440302e1fa45dddbd7ce6bb71984958a
b348840800b4a28e7e53a9acfe4a2a3932e7361d
107038 F20101130_AABUUI kirschner_d_Page_20.jp2
590e9674886b6a3464bc9cb6330adf42
5dcbc5c38793e496f842e94f5a2e15a193433878
3700 F20101130_AABVLB kirschner_d_Page_71thm.jpg
304b0ed8d0c2186d43e8d332a0eb6990
82a982be3d05905b7f4a6ac29866f82eff6aba0a
2420 F20101130_AABVGE kirschner_d_Page_01thm.jpg
a43eb9ab62dbb21ac31a7297f0b5796d
2087d6ab8e6af021deb25c09119b341e1f969539
51305 F20101130_AABVBH kirschner_d_Page_38.pro
5ea9961d59dbcd9bb2413b7b727ce705
fda67cd4e8c1aa2d2e0b91aac1507cee3414fd24
109069 F20101130_AABUUJ kirschner_d_Page_21.jp2
c494a2b947838cad5e8ee25aa986752c
6b046ee8ff7ff12be049d5d114d9827b633c2aee
1395 F20101130_AABVLC kirschner_d_Page_64thm.jpg
d2704da45083759c5cbd84bf83cccff2
687079292685d2e17d5d40852cf394ccb3e6c524
7292475 F20101130_AABVGF kirschner_d.pdf
f2669a020d0e704a28ae20f80c87c893
d7f4102781fd485739698954973d59176d3ac467
F20101130_AABUZH kirschner_d_Page_67.tif
40724a30943ea5336bfda366b99e7d70
b5a4741b1da4ef86c2cbdaa7f0f42ce080fb182d
45819 F20101130_AABVBI kirschner_d_Page_39.pro
f1a83f39d302888225a2e1cc2eefca7b
35a645a1e886779fe62d44b280c4e01e30f1bb57
110874 F20101130_AABUUK kirschner_d_Page_22.jp2
7e3a5bf5b2c8cc594127028a01416d9c
471ae06a281d7f1cebc3bfeb98701bbbb93d0d38
6204 F20101130_AABVLD kirschner_d_Page_42thm.jpg
d0bbb08aaca3770a08371b770299b1a7
3e2e3f3f1afcf317291af871431ceb170d55b288
15814 F20101130_AABVGG kirschner_d_Page_62.QC.jpg
fd4f8b0c0382e7c1c71128032650dbe7
a61da99ca0f347dba5d371e41f006c53c0b53c8f
F20101130_AABUZI kirschner_d_Page_68.tif
c5539975cbe785872387d5ec366dddb3
b04222f278f83c83bf9dddaa89dc361675853a1d
49188 F20101130_AABVBJ kirschner_d_Page_40.pro
74bed34ef87503052e32467a70b70b1e
afe51c82b0445248cb3a275e52071f32e81c719b
105115 F20101130_AABUUL kirschner_d_Page_23.jp2
7b6e406f4d5f1d53fe9940486c149dd9
4477a551976889a774aad8e2a03c5d37c05ea33a
125643 F20101130_AABVLE UFE0010297_00001.xml FULL
2a2671c6b927b0fbca294718d6269a93
be9c215930711ff1eed80f2852e338e574be08ac
13826 F20101130_AABVGH kirschner_d_Page_33.QC.jpg
28da9b3c232215369fc6c50d9f606f64
677c9695681afba51cd54ce236cbe97ab1729665
F20101130_AABUZJ kirschner_d_Page_69.tif
b419ee34ad2ff9dbd870707443b63e62
4f9c41c81c88a77844fe1345c39718db918f8062
43848 F20101130_AABVBK kirschner_d_Page_41.pro
48ebba8f585d001d50c872bf7b9b4964
0d2251cefa44434a393ea74268418e3d20593b73
70385 F20101130_AABUUM kirschner_d_Page_24.jp2
a35b3f9196d5c0e1ba84c484f85ea39f
ae32f2d4000892af35dd556c71ab906a66514675
4459 F20101130_AABVLF kirschner_d_Page_04thm.jpg
05070605e736d9e495176e86de45b05e
75037ada6d686d859b1d9b1e2005847ecd9cb2ae
13464 F20101130_AABVGI kirschner_d_Page_75.QC.jpg
7864b4c9b54bf0b54883e103ed99d9d6
e64a0acaed91d966850fd4c52bbf3d165edf5da5
F20101130_AABUZK kirschner_d_Page_70.tif
7897a9a501022e8beee78752118348c3
cfd8e10147129391982a5639f8e9a9e5f67f3043
55873 F20101130_AABVBL kirschner_d_Page_42.pro
0e465b34fb8550796e2493673b471f4e
a9cee7eee0c3124e87645fc4c40cc188c56495af
88304 F20101130_AABUUN kirschner_d_Page_25.jp2
65746a273702c9b7600c6b3b9dd09b27
686ea96492de0190088bd5a6725ecb09b7b0137d
8219 F20101130_AABVLG kirschner_d_Page_05.QC.jpg
c64168e3fbb318c907d961088049afb0
16cce01b1bbb99c9c0e57fafe517e55ab21f55f7
4224 F20101130_AABVGJ kirschner_d_Page_14.QC.jpg
7ef2ba289b132a8ca29d6041f8f8eb66
8d97a892eec3763a98cbb9414c17cc846d2d8723
F20101130_AABUZL kirschner_d_Page_71.tif
9b5def32eb7bf73c5708d2e72491d45a
ec607c80f9e80e232ebb6e9ed4632e216e4bb8f1
18411 F20101130_AABVBM kirschner_d_Page_43.pro
aa680213b381781a5a5872bbc96f5a97
8f0cac5688e40b1bccc0b820389dde602a154007
96742 F20101130_AABUUO kirschner_d_Page_26.jp2
6195c7e6206da8b8c65d79e68f4bb69b
4316bc5a5024dbf4e735fec5faad28f7f161bfbb
2727 F20101130_AABVLH kirschner_d_Page_05thm.jpg
fe40b75c3ddc9a9ec377535b8045a5e1
ac1277c07f4a97de330c7d96db82c2fa1d3c2671
10140 F20101130_AABVGK kirschner_d_Page_71.QC.jpg
c95f33f89498824d4cf9f19f647c0ca3
07e4ae1bfa36dc9b494d96749176d4b7ef6054c2
F20101130_AABUZM kirschner_d_Page_72.tif
e7883c09c504d295134d620232b316c4
9f0b0612a4333109cb830a6c1e0ead5fa168f3af
33120 F20101130_AABVBN kirschner_d_Page_44.pro
587aa8f45c1b683b12b533a04b350e10
ce132166a3eaf405864e49e7db1428b41a21f801
91794 F20101130_AABUUP kirschner_d_Page_27.jp2
ba3718ca3c93589c246a4d569645f019
e842ee8a83db600b3867d561c17b9c7916bc07c3
5988 F20101130_AABVLI kirschner_d_Page_09thm.jpg
c61f87618504ec5fbcebc91aa5cbc508
36a7bd0fb658be04d29f53473bc4c5d9e98c8f63
6633 F20101130_AABVGL kirschner_d_Page_40thm.jpg
bafa903e2574d201e87c16d920a273a7
36833c05f83befba9146972b3ed95de7082faf14
86217 F20101130_AABUUQ kirschner_d_Page_28.jp2
a5a2fc10fab56b53c0a6b8e9c8b50d93
054926c38a66fb7cf98631401d9f0002fff7252a
F20101130_AABUZN kirschner_d_Page_73.tif
5525e772f4266d989b92df0a8ab850b7
6041f71fef9b2f30f1015d8003e033907837eb02
22403 F20101130_AABVLJ kirschner_d_Page_12.QC.jpg
8eb6abc2f0a0f1ec258d78883e44c917
3f0af056c362f345237cf01f4710cb3647ee1deb
23570 F20101130_AABVGM kirschner_d_Page_36.QC.jpg
1a1ff8a8ebf00440f2bee4073f337e23
acb64d3927fda5f91011cb5979f60cc67549bb27
31043 F20101130_AABVBO kirschner_d_Page_45.pro
815357687520c369c268e386284e0961
07951da738ae0f30213c61f3ff1901f294aa9d4f
102721 F20101130_AABUUR kirschner_d_Page_29.jp2
f7360788fcb272ce80c0c52c96aac6b5
02aaeb77e1cbaa2437bcfca04b1a20975c41584d
F20101130_AABUZO kirschner_d_Page_74.tif
bd57c98c3974cdf8cbec6c736c941f72
ed39185890a269a0d0db8a571711e220981dc551
17446 F20101130_AABVLK kirschner_d_Page_13.QC.jpg
ce4116946b840f398c205b7f49ebada4
9b4acbc993b995e59dcc553183228cdf1d580386
3260 F20101130_AABVGN kirschner_d_Page_64.QC.jpg
0bbab1c400e7b16639a47984206fbef0
c5a6066568a10bb5c267c8ae53e22da70e2e0f10
21397 F20101130_AABVBP kirschner_d_Page_46.pro
dcb2d688fefade9b7c2bf4eca1806c05
3ba11c1fead54d9a3d4c2175064670748f076cc9
41798 F20101130_AABUUS kirschner_d_Page_30.jp2
8374a0deb3715575015692dd80121756
cddf2d10d1665b17c956ae96b022c206991f2496
F20101130_AABUZP kirschner_d_Page_75.tif
e12182955a4ed56c0173300b4861bc9b
be35f611186670296ec98969516d67fe97c34f5d
5438 F20101130_AABVLL kirschner_d_Page_13thm.jpg
fd9c64b5d0284b827749da195143ea0f
f1a4853c74af7767db24f8d3a1e3c22a4dbd4837
20028 F20101130_AABVGO kirschner_d_Page_48.QC.jpg
428fb5835d8e0ea897356985949ebd3e
9c3ebc97e7cc3aaad077541a7a1ede99a12bbefe
26939 F20101130_AABVBQ kirschner_d_Page_47.pro
365fe0b59e84aa8c083ac08510b259bb
40c99a4a2abe61ba88373584302ecdfbd88d8b51
107702 F20101130_AABUUT kirschner_d_Page_31.jp2
16c52ec772bc69a2c9228d6a535d6a29
c969adbde9db10b7251875ca52e5354fc8103355
F20101130_AABUZQ kirschner_d_Page_76.tif
aa6f922f26f79c6403efe9aaf7d1ed9d
b22e19f0ad8abd0ddd4abe57602c6bd19668cc99
23785 F20101130_AABVLM kirschner_d_Page_16.QC.jpg
279c06467adc6997441e17406d81dfdb
37b682e52b9ef2ef8a4b1c46fd7b38f7b9c0969d
21867 F20101130_AABVGP kirschner_d_Page_26.QC.jpg
b9dffdd6bb0be02286049b2d4e2f7567
1b762f3fa86236127fea192d0a65e51f95ca02d3
41890 F20101130_AABVBR kirschner_d_Page_48.pro
952f2d007331bff56f2386e394a352fb
98a525f6022335c8799f3094f7be13068b387e45
1051921 F20101130_AABUUU kirschner_d_Page_32.jp2
858731c93077d0aafe1cca99c173b09e
3bf8d1cd1a1f62bb90a1bc7c13e8e3e957e777fa
F20101130_AABUZR kirschner_d_Page_77.tif
f6a83e61048209c87a4a5d2d24e3f457
17f1fab33a6f0ed515feae4b1374555644a1c9d2
24680 F20101130_AABVLN kirschner_d_Page_18.QC.jpg
eb9a65810b481d07d578b052795df2f4
66d12b58e85b02dcbc863dbd8b04c3bd644f11da
5259 F20101130_AABVGQ kirschner_d_Page_51thm.jpg
355fd68bddc92844fe61e4f9c032030d
0d02ece730bed6a8a3b325006d054bb6d3442e88
43094 F20101130_AABVBS kirschner_d_Page_49.pro
558a3ffd987840f9ee5db8c867d8c60b
92975bee7f9a2e52b097b482feb466a8608438e7
578882 F20101130_AABUUV kirschner_d_Page_33.jp2
b64c5871c04a59d3e5a2fd665141f589
f4e52066be2ae62ac6328c1d47bad05d74f77b8c
1054428 F20101130_AABUZS kirschner_d_Page_78.tif
6438a7d7bf0c4d97a32db068edd2f8b0
4ee238cfea4fe662c49f6d5b80e81ba50c624721
6806 F20101130_AABVLO kirschner_d_Page_18thm.jpg
fddd613bb78b8eb8dab0dd2d6c87f7be
9a8f2ec1e5b171392e44d3b27dfdf9e213fb8c57
16532 F20101130_AABVGR kirschner_d_Page_24.QC.jpg
9cc0c63766ef2095fb8cb16edd31aaec
007b66aa2e7052b3e48f0e321a30312250805a6e
44155 F20101130_AABVBT kirschner_d_Page_50.pro
2a429e7461f35e5f6f9f41325a8de812
bb7a84f6a6995c9d071d56c3cd9dadd8e1ea8704
101839 F20101130_AABUUW kirschner_d_Page_34.jp2
560b31507b23c1684cd8f83b2628a2fb
47c70ed66563eb51fcd9ca0278a22123e6e84a53
F20101130_AABUZT kirschner_d_Page_79.tif
1f2d1dbff15f496150aaa0ff25a4e5f1
29a498b2742189ae1365c01b84c72774a40cc9f9
23931 F20101130_AABVLP kirschner_d_Page_19.QC.jpg
01bcbd01594612b1dab1d6b43056c942
a61b8715c312f593fadb76c837053c4d07cef933
16140 F20101130_AABVGS kirschner_d_Page_44.QC.jpg
d6498538492b0658d0fe30b49aa29dec
f356892bfa8a5edbf7557757e537b90902a5c86f
1051964 F20101130_AABUUX kirschner_d_Page_35.jp2
c8ca91ae9b174bc5d2483930844f4323
6c613521a12e7ab57aee7bc2e9513bf6b529cae3
F20101130_AABUZU kirschner_d_Page_80.tif
5492db50cc0e2c307798013b7a92c2d0
e6d1cf0cd33fe88ad10d27b25dac0a5fff0a345b
45545 F20101130_AABVBU kirschner_d_Page_51.pro
dbbad9a2e7adaeb46882576dcce5b8b0
2ab262dd38128f80994da179994b14b53bc72d6a
6613 F20101130_AABVLQ kirschner_d_Page_19thm.jpg
076a234437bf52a6b52d89f6a9c997d8
15f34e2f2778cab963f13f482147def7ccb013a3
107429 F20101130_AABUUY kirschner_d_Page_36.jp2
71b2cdcc9818aec79d190ef22cea9681
452c46eab0bc81e3b3ea573f3d6bc56face5e1b2
F20101130_AABUZV kirschner_d_Page_81.tif
fc18077a4d2348734c4e6e167cd1c1bd
3ee03fd3885eefc97d3f91220e9851019d30bf0f
40437 F20101130_AABVBV kirschner_d_Page_52.pro
a3a86e80efd045d5c027f58e630d26e0
ebc66800e693912c947dcade0879be1c2b3f5963
6533 F20101130_AABVLR kirschner_d_Page_21thm.jpg
03e960734e89cf6c19da17b6e9fe9f37
7caf7503687730ff8870f0189e8a185235c0951f
1610 F20101130_AABVGT kirschner_d_Page_14thm.jpg
7f1f889b97494ddd6af70fcb68271172
5abb937f9dad0472efceaed4836c009244344bb5
28782 F20101130_AABUUZ kirschner_d_Page_37.jp2
6eb1b2d1440ff0f33d8f2ecaabbcf909
8416484623d972f4944d7b1aa4dd288f21b1fbe3
F20101130_AABUZW kirschner_d_Page_82.tif
33d996400608a98cb4cac0827300256a
557f9e769f7f035efd51e1eecccc7efa6e403d67
44224 F20101130_AABVBW kirschner_d_Page_53.pro
c9637edeb57b3995751829523b97fafc
ffec0545d9632942d28e5b6388878660cbb0856b
67903 F20101130_AABUSA kirschner_d_Page_42.jpg
9a423af73a36de139e8a0b130f0ab658
9bb3f40d065f592bb7c331b8983903200c8a549d
19870 F20101130_AABVLS kirschner_d_Page_25.QC.jpg
4ece48fc2dd07b6d84fd4e1b924afe3a
eabf147b4c9162c091a37596feef8fe18bc75102
4787 F20101130_AABVGU kirschner_d_Page_61thm.jpg
947b7fdf1b0a887b10b6b8e614b2323e
50dc843ede100373eb88cddcf4c4f0d91752e295
8159 F20101130_AABUZX kirschner_d_Page_01.pro
166129f5fb6056c793ce78d2330a7a52
5b853d713b40292583b9a2574c62c24beb13f31a
37548 F20101130_AABVBX kirschner_d_Page_54.pro
ec160bd5f253c3f25f7247b5ee24b451
662a0f1e67f61b17393d2ed60b68f3cc7c6a3bf1
31631 F20101130_AABUSB kirschner_d_Page_43.jpg
6d1590b15f38a3c4594f9b70d5964422
95a7f71d30b9afa4865ab3782e26e77501a02c46
5969 F20101130_AABVLT kirschner_d_Page_27thm.jpg
ebc05a59d8aa8723d1230e3df9440714
2aff4888cad05da608858fbf0c6916964cc912d8
5037 F20101130_AABVGV kirschner_d_Page_59thm.jpg
8e385f5733c45df2bb324bc89bfc9318
7b8fa962fc78c802d811a56bf1a141ca923ec58b
F20101130_AABUXA kirschner_d_Page_08.tif
5b53ee07958636a8a9afe0f02cdd642f
d30d304966a5c4d487b42d33de26633ae0557682
1254 F20101130_AABUZY kirschner_d_Page_02.pro
b70d62eaed604a8c5c45f3c2a8b913a1
2715cf0b55e7c2fd36d7c7b4fcfb9fe45105ce6b
45876 F20101130_AABVBY kirschner_d_Page_55.pro
38bc167a9796c6ca599d81e073a5aa52
4f1c00f37fb2d820788de4963e30ee1b8a1937e7
50804 F20101130_AABUSC kirschner_d_Page_44.jpg
5fbe5ac3965f4f70eb4d5dec1b5c5af6
1ba6b9bdb7a299893910caa9c6813dad6ad4fc85
6321 F20101130_AABVLU kirschner_d_Page_30.QC.jpg
8d9ad0e8f29ce504254c680664b7a578
dfbd573776702671e44b2dce187ef7719dcc9ac7
16407 F20101130_AABVGW kirschner_d_Page_68.QC.jpg
e5a130aa2163c4ddd8b5ad7b4ebad844
ab11670a210ce741928ee8424e03e2e35eb2d083
F20101130_AABUXB kirschner_d_Page_09.tif
fc4bb9d4c104a0ff1a12fcdb8b7a8207
e97280f2bafaa24e24f0d42821a07066b86afc35
4140 F20101130_AABUZZ kirschner_d_Page_03.pro
a23169af9aca4c8f5e67b70e916fe9f5
98f0019103aa66e93d8541ecc4ea76fd396dbcd5
48586 F20101130_AABVBZ kirschner_d_Page_56.pro
cfc3d9b5c697bb49ceb2e466f03f2697
ac4d33eccb5f488836e1dcea0daa21561de79df7
45949 F20101130_AABUSD kirschner_d_Page_45.jpg
fce1bfb62da1ee46a8990585ef4b17da
1ff57f1ef6b7203ccb753c6e1c531b9a30e5aa96
21044 F20101130_AABVLV kirschner_d_Page_32.QC.jpg
663ec03bfc1ff227142f110f11c1a470
66372535795247abe655f3adc2b8e5b66286bc5f
2947 F20101130_AABVGX kirschner_d_Page_80thm.jpg
e2c901fb209912061752d58949c034cc
2cea2ba258b07de49ab85ad3ff481880c944a5c7
F20101130_AABUXC kirschner_d_Page_10.tif
8729b54322632473bea9b87839406adc
f5b1045155f252fbb03c8f4daef3106b56fd0553
42618 F20101130_AABUSE kirschner_d_Page_46.jpg
b478fc6f2269929e6a1f1b7e56d2f52a
f0985cf9dcc2c9670d56ab2e2030dd0d99a88ef6
6359 F20101130_AABVLW kirschner_d_Page_35thm.jpg
9cdc00e73582214aafaa04a18288dc09
080995aa5cda3b7f3660af261ddfd486adef6f6a
16524 F20101130_AABVGY kirschner_d_Page_74.QC.jpg
ff0df87419d444f92216f620fcefc9c0
cb2c19e4cac6b95e33c984a5ef22829d2f49a4b1
1761 F20101130_AABVEA kirschner_d_Page_27.txt
255eb8c1f1a2f425c8eed4b623686b2b
e1da3d191cfe3f0a1093d35910e9a824ea52cfc7
F20101130_AABUXD kirschner_d_Page_11.tif
c95994f8183e4da3615790700284dc6a
0e82ed98eda5864749369381dc0596950fe3905c
52472 F20101130_AABUSF kirschner_d_Page_47.jpg
e2b6693fe0a0796c3373055eaaeba778
663d16a61b8945e06874ab58435bdafce9a50d75
15268 F20101130_AABVLX kirschner_d_Page_45.QC.jpg
44dc958d8897b67eced6dce1e3a6fc8f
d84456ebe897a0b9d0533c3d6908342fbcc09c70
6866 F20101130_AABVGZ kirschner_d_Page_31thm.jpg
810d8c4ae0f6ea6ff53b46abfaeaa5dc
50c7b111a2753d58b6bce8971eadc849923d4ab3
1650 F20101130_AABVEB kirschner_d_Page_28.txt
a439a88acb2193f83ae0ca8348ea9a97
7457cecc0719135f188edf4b80361e7ee00d7443
F20101130_AABUXE kirschner_d_Page_12.tif
50dc01432ce754a010683bef8fa22f24
fe545f77a6ca3c43018e7578596636c97da32447
61899 F20101130_AABUSG kirschner_d_Page_48.jpg
7a9790c8e30074bcd8a017fcb123a942
01a64eba3ce2b6686113e14ae03a66641b1f8349
1883 F20101130_AABVEC kirschner_d_Page_29.txt
14768b9b34448642717949b1b408da36
68d8fa438c76b9dda5a85046fd88acc2c95dfe97
55423 F20101130_AABUSH kirschner_d_Page_49.jpg
7e87cb0acbd38403564fe3d09da3e3b4
ab85141aa5c53bfcb3947eb18da06e984ab18899
14782 F20101130_AABVLY kirschner_d_Page_46.QC.jpg
6856fddc9a2243ffbc420c5d222218c8
47489fc6cfa48ec885011aa28cc08be42d02e228
5640 F20101130_AABVJA kirschner_d_Page_39thm.jpg
cb350bf9afcef72f79a68870a77f78bc
e572301662b67b2704195573710c3fd5e705ca35
1377 F20101130_AABVED kirschner_d_Page_30.txt
997383624f77c3b87ba8d59372ccb639
64512161d136b79a5cf20d86efb5ccfe143d9e14
F20101130_AABUXF kirschner_d_Page_13.tif
935883978df6a88ca5ea68d78f79960a
aad3e88f94d9e1b5e28bba04f26f0cdec43e0ffc
61786 F20101130_AABUSI kirschner_d_Page_50.jpg
5087f9cd96c408813ce2c39c356f5504
8dc13363ae37584d3420b95d2ec54366feded231
5198 F20101130_AABVLZ kirschner_d_Page_47thm.jpg
c94086f1439ac3cbee9ebea6f0f0a4b9
02294122ad4fc0f8a8a805be9877f057cd9a43ba
5088 F20101130_AABVJB kirschner_d_Page_73thm.jpg
79de68b04d4e4d13fdb3359dc8370a5d
d0866967d88267f1567c9b9e2123dc389b41ef7a
1977 F20101130_AABVEE kirschner_d_Page_31.txt
bf5a6cd1c48c735db7864b762393af5f
2fb10d643630c9c41f296a5bc94cf73eede1b274
F20101130_AABUXG kirschner_d_Page_14.tif
f63c270b2bb8da5573161595007e2829
5af1fe6ceb01a59ab0bcbc562bc5e64589b628c2
54705 F20101130_AABUSJ kirschner_d_Page_51.jpg
eba8b5a422903ce94dea679744fb0070
3d7fe0535ad700be78b746e5296e429b57f822ab
3809 F20101130_AABVJC kirschner_d_Page_75thm.jpg
e2d692b5a8947d27205c9a8fae0358b7
bd162b0ef768434254c26895743cd80794e87479
953 F20101130_AABVEF kirschner_d_Page_32.txt
20e6880ffbf70db5f0150f88f38528d4
7dc792958b41addf3de873f9964d45ca1dce434d
F20101130_AABUXH kirschner_d_Page_15.tif
fbd95145606267a5ae49956874946503
55ec0ac9a8e50c60479b33dff998c89d79d83901
54715 F20101130_AABUSK kirschner_d_Page_52.jpg
895c34d0df74cac33110317a993b9399
9b56ea9d1006cceef12facb3872e890ab3a1e489
6934 F20101130_AABVJD kirschner_d_Page_22thm.jpg
d92b518d98d3929e7e912ca055d93de1
2ef4ccebcfb2613526f2cb18fa3f3fe897cbfc23
1030 F20101130_AABVEG kirschner_d_Page_33.txt
1131423c1eeaff55898c44261c5eb47d
19f10c13ec011a1ed951ae4a49ee8af2b13749e7
F20101130_AABUXI kirschner_d_Page_16.tif
1235f54c37c7f1e520bd54fcd7a1d6f6
ea1cb3ff55853057db4fbf06b6cc06e212336ccb
65048 F20101130_AABUSL kirschner_d_Page_53.jpg
576bc9dfcb60d20f92ff56d233bf2a16
08025d949ab4e2fd97dbd96305d1c954af45854f
17223 F20101130_AABVJE kirschner_d_Page_47.QC.jpg
5fbdbc5eb5fb4176571b923ca567f528
919344fe6ce40f12fcf03c9e8fbe5253ca263203
1857 F20101130_AABVEH kirschner_d_Page_34.txt
5199fb6ebb4e2dca5e5dacac7df3738d
ee8fe848048bb3998feb6fb9fffbb5ae95898cc5
F20101130_AABUXJ kirschner_d_Page_17.tif
130f8cc74ae98c77a1806c6a4e804c4f
1a95a2369942bab6b37ff1c96989fad6c33d7717
58140 F20101130_AABUSM kirschner_d_Page_54.jpg
f257da08ea00815ece6b397fff6e6728
16034a0a9bb56ee7a577ea521799c16e77a01f36
7340 F20101130_AABVJF kirschner_d_Page_07.QC.jpg
05dea83f4f6eb51aee38b0c30a619e04
1f3a89036e783c8a2a94b5982ad35b69e2112c20
328 F20101130_AABVEI kirschner_d_Page_35.txt
f719b2b93ee5aba618ec50f73f7d6f99
9e01b5ca75871bbff55df571dbef74f8acc285f3
F20101130_AABUXK kirschner_d_Page_18.tif
b4340e9b067a0f7e6a898bd40d92a959
bd6d045ec4b3d81ec96a51a4b6de58140c4970b3
64939 F20101130_AABUSN kirschner_d_Page_55.jpg
b01a5c35b09d83189abf903a68eeedd6
41f4602f87a1b71d119850f2bf58abc45f5451cf
10152 F20101130_AABVJG kirschner_d_Page_79.QC.jpg
6df551d980677adafee1111bee532708
2500528d5072951eaec59ce7e99ef66e01250800
1995 F20101130_AABVEJ kirschner_d_Page_36.txt
fe64481397f6569b17e4c96e79dd0023
a4f8b092cd1fae93ffc906409832a32a0bceebf0
F20101130_AABUXL kirschner_d_Page_19.tif
cf58551206c31cb8ab1876801e7660a5
36df8b4210dc9b98268556e2c97c8824542d1f86
71764 F20101130_AABUSO kirschner_d_Page_56.jpg
31e63f9746108755018d725612c9b6d9
80562953161091795c445dc7e8c19f6363590e5f
18446 F20101130_AABVJH kirschner_d_Page_39.QC.jpg
ad9ba8df2a7cd00ff4e06c60aa77c63b
038d747497d9515e9089912e1367cf8c174b4941
524 F20101130_AABVEK kirschner_d_Page_37.txt
bd545c989fd322780d1c80d4966b2473
708005369faa647538880f3fd9dde6539b0b91f0
F20101130_AABUXM kirschner_d_Page_20.tif
1a0f8b8b2f7699c006dbc8661064dd39
34bd1fca4ee0ce6cd07433eb89ac6652d3e99c49
59898 F20101130_AABUSP kirschner_d_Page_57.jpg
5e041ae0f3df56781592e78ca1be4e56
8898589f689978efc6c71cac78b6b974111deb12
10410 F20101130_AABVJI kirschner_d_Page_63.QC.jpg
a8214f6f9e0058986c44075fe797f91c
efcab43ee1ba4d7979c16871c525e309c8c652e3
2932 F20101130_AABVEL kirschner_d_Page_38.txt
3c1938b4a92a5648dbe8025c5a31201b
9ddabad2ed133b51908f39053480cedb1340a8c1
F20101130_AABUXN kirschner_d_Page_21.tif
b36feccf5da79ec87a0ccff48be5b6e8
574cbe77d54ff27a0b0b23764792f631936f437f
77385 F20101130_AABUSQ kirschner_d_Page_58.jpg
1452dcfbb336f316bee6b11c4fb36c71
98e12aa0280cc891631ba62f27b73519581604ea
21762 F20101130_AABVJJ kirschner_d_Page_09.QC.jpg
aba62f5549840fb43ac12f7b16dd1110
d87255c2b4e367d1bf0f83cc48055303ff8a4fae
2047 F20101130_AABVEM kirschner_d_Page_39.txt
9f0061f1db2f28c05a15a9be426785b1
be9c1cf556cc42f109965698b2500d745839dcf6
F20101130_AABUXO kirschner_d_Page_22.tif
4b11c658fcc2b653fbbcf9b16cecc86e
8cd428dfddf8af76916a173df7a04ef58ef9db7d
57614 F20101130_AABUSR kirschner_d_Page_59.jpg
f85649e714ac5d1c36d1084b22f76f32
c3b95ac9456ff8a4bd0c0b490c6db7e17df95b92
23598 F20101130_AABVJK kirschner_d_Page_21.QC.jpg
27e864676b4836202be6a9ea5786e5f7
2e64625130f7dfda77e72fe61a6b99d1d9aa4514
1950 F20101130_AABVEN kirschner_d_Page_40.txt
f18eaa2ddb1ca40daad4ec4b041b750b
706d71e2ecf43b86f5f6fedfeb99387cffb0be34
F20101130_AABUXP kirschner_d_Page_23.tif
b43a71a42e5c0718c09b2670cf8207cb
be5d10182f503d94ee63290efa83a956a4592f18
30339 F20101130_AABUSS kirschner_d_Page_60.jpg
eef07106f2d425d6bd2246d931673cb1
6ae2072e5b0ea058ead62cd620897290ea612e62
19490 F20101130_AABVJL kirschner_d_Page_57.QC.jpg
9604201b86596deb592daeecbbef3097
89bd2d1858ebacb9200f390a2c50e1fc4adc4604
1776 F20101130_AABVEO kirschner_d_Page_41.txt
0c60fa784cd0c889d7b954a5fa0e40d6
d12e0425316d2a1dfeb66b8ddc11deda315499ee
F20101130_AABUXQ kirschner_d_Page_24.tif
8afcb1c73a29c3a9dc3d4d7053faa754
ffc660e36fa5b4922dc3b48ca66061d2344aabc3
54021 F20101130_AABUST kirschner_d_Page_61.jpg
a6bd0dd149a5466a1ffb64dc44b56503
0e322b04d2ad10f3e85eb395c6f819fae68d391b
6208 F20101130_AABVJM kirschner_d_Page_26thm.jpg
333fbe1f3b708dfe0e87ed7a28a46612
7a31bf1938d7a3c8c4a4100e3b6c764a2e5e3039
2482 F20101130_AABVEP kirschner_d_Page_42.txt
ef5f41629b5e715b9eaceaae9135524b
304cf5a34892450394cddf55334613d44af19a1e
F20101130_AABUXR kirschner_d_Page_25.tif
9c354975bc05778b1d8b533e49948af0
f87b02c8419476dcc39bde81b4a71d008bb8d9fd
50246 F20101130_AABUSU kirschner_d_Page_62.jpg
4e4f5449cf4c2825708ef9a4eb22bf75
4e101e42a280082bff5e46d92106498114127fe3
15361 F20101130_AABVJN kirschner_d_Page_69.QC.jpg
51a2bdea349027f96a4a7ebd628209b0
48fde466b68a8d0dff7ba5d677f39c9e038dec34
731 F20101130_AABVEQ kirschner_d_Page_43.txt
3b24934a3dc0ec10740235ae34bdff57
6623ea61dc581dcab50986e22350dfce9c0d5439
F20101130_AABUXS kirschner_d_Page_26.tif
d58322f7b98c51724b046a84046ded65
1e06d1b4c0c85b6ff609a15087de7744275ab379
34089 F20101130_AABUSV kirschner_d_Page_63.jpg
d8ffcb1fb4642d9c16f6559292c20268
e0f595adbbf885c50dbbad73ef25e7721067668a
3464 F20101130_AABVJO kirschner_d_Page_06thm.jpg
aedeff78f9fd7a20420e0f8da4bc482c
95f8fe83c28ac4e30871557c716c673d4200cd8d
F20101130_AABUXT kirschner_d_Page_27.tif
ea2a239759386d7171076cdd28015c97
bd576459a9f9f7ea5736912b1794ed2e0220142c
10328 F20101130_AABUSW kirschner_d_Page_64.jpg
5394ca31c16e74edc852302a47f2762a
1062d5d1ab976a697953aa47c225e4bb4af58ec7
11339 F20101130_AABVJP kirschner_d_Page_06.QC.jpg
a6b8ad02f83f4fc33a3d4d8eec104454
7ddd07b14b998b7ba9d12075e639bd7460471127
F20101130_AABUXU kirschner_d_Page_28.tif
9724f56ba3d384a67441f78200b4bcfe
21903cec4323052c97165bc1d68c64fef0f89093
53461 F20101130_AABUSX kirschner_d_Page_65.jpg
99a7c35532e991070e46496ac09529d2
783158e4d9fffbb7c72ac5fd7f001234857a5d41
1505 F20101130_AABVER kirschner_d_Page_44.txt
784a817be54bda2319c763b323ce7d39
d800b27e1d1dea7b3d9de2d42be435fec568593a
5593 F20101130_AABVJQ kirschner_d_Page_52thm.jpg
b7ad415a722856dd79d13eb481d5a82a
6ddda5fdeb365f2588c66c9b281e8dfeece19101
F20101130_AABUXV kirschner_d_Page_29.tif
d6ab49989dbea6768d68ceefa4b90b55
c1a84924efa84fa94f40866985f49b3327435be1
52537 F20101130_AABUSY kirschner_d_Page_66.jpg
b45265fd8ad116e874e93263a058417b
129dddf7fcf4780edf6d14bcdb8657a11d0c5e08
1564 F20101130_AABVES kirschner_d_Page_45.txt
3a5b88f74be3ed4c0762e7b7ab78ccb7
fb4547565c765ab435bb5f667391845973bf8a6a
6475 F20101130_AABVJR kirschner_d_Page_29thm.jpg
29ba9b372b4306e426548e28424a4ee5
df0e0351be277ef059ce932979a3628dbf81b6b9
F20101130_AABUXW kirschner_d_Page_30.tif
5b373a6f56b0d972fa06af91e41cbc9e
92f3b3f29a7747c54c285cbf3dee077a0ebb4b3e
53660 F20101130_AABUSZ kirschner_d_Page_67.jpg
b79e12741338b6e42b6172f8f9827b03
bb71917c2509c7af92a57cf17117255ecb685181
933 F20101130_AABVET kirschner_d_Page_46.txt
b10fe68cacf92ea3873f971bae3801eb
7fb75a26202139ac3741d7ded7da4eefb71d973e
2142 F20101130_AABVJS kirschner_d_Page_30thm.jpg
fec053189658cf2893a712c924bb1af3
d4da08b96c41d5ff2465493e3e9ff0ca5e618c1a
F20101130_AABUXX kirschner_d_Page_31.tif
dc9be4c082301c30c4ef320599377445
3249d84cef4b827c45111e5e033a3b6a5eeca2bc
1101 F20101130_AABVEU kirschner_d_Page_47.txt
26efc5de2e0938732b573c5b286def55
87fba7e8c1865e1d360d9be2cb5f9f50c93e6bed
5683 F20101130_AABVJT kirschner_d_Page_28thm.jpg
ab0201b704c0740ca05221bad6626cbd
2951584e8da0ae5790792103ccfaaa848d537915
F20101130_AABUXY kirschner_d_Page_32.tif
97c7d32719cf85f9df29307ceaeb6703
25310cf574f66c67c004f0b7bbe80a720da8c777
1756 F20101130_AABVEV kirschner_d_Page_48.txt
d33a0b83c84d015ed811fe716ee740c0
adbb89f662f7ac2b2d138a959668436cf92a2b5e
53835 F20101130_AABUVA kirschner_d_Page_38.jp2
f432e81d2c4875151c474e8f7e1d67ca
f10bd72f5746c690a5351dbae2438c8c1d8502b3
5574 F20101130_AABVJU kirschner_d_Page_54thm.jpg
4c667d877c19e2eac296607b1bf48a41
1ec495214b6e01dd30d85d5c9585daad3c31a283
F20101130_AABUXZ kirschner_d_Page_33.tif
16c3af077edc1e6bfc5ec8dc5e20472a
82b9bfcc3aba06a4f6add26b58287d93131611e7
2106 F20101130_AABVEW kirschner_d_Page_49.txt
66579c04e0a3f0976bab59c8cda1ca2a
aecd2bc3504d94e207b534b880c06e64e89ca582
88193 F20101130_AABUVB kirschner_d_Page_39.jp2
8ab434cf8fece8fe1198f62bfa3c7646
f8dd1cccab7c4dfe632b9451fd3b23c3d5284360
1484 F20101130_AABVJV kirschner_d_Page_72thm.jpg
08a061782a93ac61d72e2faa24ed24f1
26afabb6f0327e766c80aec2c2adb10e3e801390
1893 F20101130_AABVEX kirschner_d_Page_50.txt
7cdcfdcbcac68edba0fe3fab468978df
026441b9f1c8fbd9c393842999425cd207717987
104281 F20101130_AABUVC kirschner_d_Page_40.jp2
76a422832e5bf7619380b60d20576da3
eaf2e3cd264450b91b05e3f4401103b3191ce3fe
40714 F20101130_AABVCA kirschner_d_Page_57.pro
41cc9420e82904c5360de3d67893db2d
60a7b8ef9bcf71c0e49c22de2ead56f52c8f205b
2158 F20101130_AABVEY kirschner_d_Page_51.txt
d31c971448a181af173537897c1feacc
0051208c9ba1f503d137b8d4ef6b9141b3ef7fa0
6466 F20101130_AABVJW kirschner_d_Page_82.QC.jpg
d1dfdd6a29fa603b837052d7f0c07a26
ab9e14cf13af243952d4d6861c7d99006dd49530
26300 F20101130_AABUQG kirschner_d_Page_38.jpg
f7a943f3f931e6916226b00f821445e0
66e63d23c9a1e53f039f20ef27203846b22da0cb
53088 F20101130_AABVCB kirschner_d_Page_58.pro
03aad10917f3aff5bfbca7bdb00579ba
ba8af17e1e0b0afb5524c12ef4936a6d7fbf2940
1788 F20101130_AABVEZ kirschner_d_Page_52.txt
5d05741bffd5183761ab580d82c73e13
c2633d7c3d3c4bf6ad048f5a05095811aee0292d
97687 F20101130_AABUVD kirschner_d_Page_41.jp2
afc92f886ec535f8d96bfdec0463df20
429b4548bccbd1176f11fc3ca8c30cc5b486e0ad
7491 F20101130_AABVJX kirschner_d_Page_01.QC.jpg
e41112dd32fd583290ec477d6eaf1cf5
f9fa712fed3e1f518910a34683330fd36b7f84c3
32210 F20101130_AABUQH kirschner_d_Page_06.pro
f14f85fb372afd2a37f2a0374f7d567f
8fffcc6e1a4d2767087eba7d5dfa4416ba7b0d26
36442 F20101130_AABVCC kirschner_d_Page_59.pro
e34ecd525836c950f885124407c0a15f
f05dba24954d283bfca9e9f37506d573f182cbab
101694 F20101130_AABUVE kirschner_d_Page_42.jp2
a3f7abc20dc6b86ee30bf0c31c6868aa
ce3da047d2a59ad5fa675eaa2ec66fb1d9b002e2
6536 F20101130_AABVJY kirschner_d_Page_34thm.jpg
ec8b65f3b77e7e3b8ed44ac25246a8bd
5bd90698b92b3a1d58883c36a0d3dff5966e7dc3
19170 F20101130_AABVHA kirschner_d_Page_15.QC.jpg
e73b61c39ea70f12f47e5d574643f507
c0bdba89a1f63c8c0ad582f465e330c5fc32048b
10898 F20101130_AABUQI kirschner_d_Page_78.QC.jpg
69acea7838759be99fa38e94d782040b
276972f21718056ce04fecfc5ccb0141e487eb79
2777 F20101130_AABVCD kirschner_d_Page_60.pro
09e65eb57ff2b54010d821a94ce7835e
bc70b628504b583b0073ef175d467fafb342ec9e
42804 F20101130_AABUVF kirschner_d_Page_43.jp2
df02d32e8e8b0447c48ac7681b19046a
9e456ec2234b08bbfb4ddf305a84a05ebeaa4c7f
6084 F20101130_AABVJZ kirschner_d_Page_55thm.jpg
2b0a5bd0bf3df432a151e2b6039dc138
a8bd44e8dd4f3e95ce9862f21360ccb02864f685
24745 F20101130_AABVHB kirschner_d_Page_22.QC.jpg
e4516d648e0fa6b9f6e0a97d30aeb2da
7a55283dca73f5549fb14102dd368e98d989f5c7
97165 F20101130_AABUQJ UFE0010297_00001.mets
ea0ef166211fe2a85ca74e0e6a9395fa
93110423d11e467c54d83ce852f588563e25b6f5
7556 F20101130_AABVCE kirschner_d_Page_61.pro
845a480e7482d39b7a628276e5bd329b
761d14807fd746405ef71ead0c0f78f6aea089a9
72751 F20101130_AABUVG kirschner_d_Page_44.jp2
964fe0692bc249ee4a02d998b427fecf
125e7d6013e2fc98228a27c56919ace4ee77c802
5590 F20101130_AABVHC kirschner_d_Page_15thm.jpg
1fa13dfa0b5448229039bfef3e68dd68
0f8ce31b0a00ee21e893b2b71b9682c7be3d65a8
2338 F20101130_AABVCF kirschner_d_Page_62.pro
286aae47f00d0c04d8b019451b93a37f
df96b06cef73acd43ab231ad2113ac093516f9e1
66523 F20101130_AABUVH kirschner_d_Page_45.jp2
3405b0b982f718c5b7384cca328b4048
fef0f249dfcf1c8ee0eedcfe1889c7a6c6868a5a
5962 F20101130_AABVMA kirschner_d_Page_48thm.jpg
7ebe11adf5ed84466105a95a4d4d3db8
b09c25e559cd15c736d743c3db672cf4a8786685
6620 F20101130_AABVHD kirschner_d_Page_11thm.jpg
6e85cc9e19e80dc0fa2559dd87e2b1db
e29bd38fa202095693cadbafd03a641f67b7ab83
3896 F20101130_AABVCG kirschner_d_Page_63.pro
f26de6e83af4a339b5244140a61c928f
9b722ad319d0ae0dd98128ce225e50e810ba180b
562096 F20101130_AABUVI kirschner_d_Page_46.jp2
6aa3191744b851db5eedebd23cd060a3
c34d7794c8749e37899132e12be16b00a2ea57ea
20428 F20101130_AABVMB kirschner_d_Page_50.QC.jpg
b4fd47eb1260164c297193464121ad1e
e2f66cefaa39bc09ad013bed4d2950489617dd24
3537 F20101130_AABVHE kirschner_d_Page_72.QC.jpg
0ca868463b10f67871eb09b864bc9aeb
35daa55e7364f4f50903abf4d896bb536281b481
23026 F20101130_AABUQM kirschner_d_Page_01.jpg
9cc3d4032beb2c2edd102b135940655c
036335cc83d4ea88579cee0f8dc45f2fe72304c2
1161 F20101130_AABVCH kirschner_d_Page_64.pro
4f184c4266f7b1a99f042c337b4b2687
38a6f98abd4a9417aeb7e7bb8a9507716462e511
695264 F20101130_AABUVJ kirschner_d_Page_47.jp2
ffe04c7cea455a3d42c1eee9646df3a6
5e731bfd9d429dbb5ec53722ca9135414218cbff
17916 F20101130_AABVMC kirschner_d_Page_51.QC.jpg
1165618cf226269b57f5a6ccbd71e4a9
36341efab0719e79802ad9b4b64b158d523a8aa1
9868 F20101130_AABVHF kirschner_d_Page_60.QC.jpg
d3a787fa85da929a34272a8bd22c3a0a
4b815469202b1fdee309a0a1dd481a9eed62112e
10447 F20101130_AABUQN kirschner_d_Page_02.jpg
935fbdbc99825d4652012266c446c0fb
0727edba530b510ad203a1ef2b3b3162479440f6
549 F20101130_AABVCI kirschner_d_Page_65.pro
9787152f56a84be1f790499327e34167
b92213473e59d1eaf178bc72722cf83d937ebb25
90052 F20101130_AABUVK kirschner_d_Page_48.jp2
cf6e96886d7f8c14ffbdffaa55f12a70
18377bf0271137ebaa52e94f73082d093f61e5eb
21431 F20101130_AABVMD kirschner_d_Page_53.QC.jpg
3dd1c349cbf2c1f3d386a66b276c4273
c0d4564a1dd4744f546a56f88c8d2f59f35a52b7
16173 F20101130_AABVHG kirschner_d_Page_61.QC.jpg
3cf74e6f8d23e17fc1ff55a3d8776a00
4e98b9e6e52c64062d073b643e210974f88bffc6
14710 F20101130_AABUQO kirschner_d_Page_03.jpg
b1a461b6b98cf88600eed2176778e6a4
a49bee0113d69752e25b8ef4176ce4016c23d83d
334 F20101130_AABVCJ kirschner_d_Page_66.pro
d6db5d501e963d03d4dfd173842f4ebe
34772e9e8f735950dc34e3a5a565bdec879856ba
81182 F20101130_AABUVL kirschner_d_Page_49.jp2
dac19c976d4fb25a97b94aec27fc75ae
7c46edb1722bb9f1963bc4a046b0b150bf54302d
5586 F20101130_AABVME kirschner_d_Page_57thm.jpg
1a029a3eef7a59b7c5a53abf21d97fea
b72807071c21f8c7b0ff937cfbfaa082d85d9bae
17268 F20101130_AABVHH kirschner_d_Page_67.QC.jpg
232aa84bcc1746a90ef9ae0397d213e5
054daa21de7a361e3aaebf0accaeb087960ce192
58154 F20101130_AABUQP kirschner_d_Page_04.jpg
e6183526c94ecf00915728931a10ada1
b172bf8bbaedc1ce3e92d86b3a91e5b72cf02829
495 F20101130_AABVCK kirschner_d_Page_67.pro
a052bb1f7d4d14e17c88adf09311cec6
9cc0e9c136f4c3df6606f4fc2cb7943fddb4f8d8
90753 F20101130_AABUVM kirschner_d_Page_50.jp2
2c67a0c21d9fddb8c298a444a94f5e6d
2aae7c90a3e1144999c63f8ada947b09e8640cf2
6107 F20101130_AABVMF kirschner_d_Page_58thm.jpg
688931566eea5960f75707c31b588e04
81b6bfb93b1202b3ae71d8683faf05a78e942345
5927 F20101130_AABVHI kirschner_d_Page_10thm.jpg
1bb58f888eb9c2977ce167696b59a6cd
31040ea0a5da9c3e71abd6e3134d34b4f571cbae
27332 F20101130_AABUQQ kirschner_d_Page_05.jpg
93b491bc6fbbe94791c6f37a3fd70abe
c38a6dc4260dded43a2ab04fdf61c8a0701af48d
F20101130_AABVCL kirschner_d_Page_68.pro
6ea0706f80cb210674e07f2bbeb9d0ea
720349c50658a89d55f1da872fa49416740b606f
80680 F20101130_AABUVN kirschner_d_Page_51.jp2
e21fef6e0593ccb3f69caf7ba133d804
8ed155bde922af0443447d15f29be961075f6515
17444 F20101130_AABVMG kirschner_d_Page_59.QC.jpg
259a89d66b1655d82c3cbc966e04e5fe
97e076888517d1d035e75899700400f11e254d4e
6424 F20101130_AABVHJ kirschner_d_Page_41thm.jpg
b2ba535012153af77901bc9879d6822c
bf866f5185fd607d17d1064c331dac97dc86a20d
38240 F20101130_AABUQR kirschner_d_Page_06.jpg
0d8f306468c24659662109f4508a0824
9aac28a2ed6c91c616ae096d6086f48ed1d64f0c
F20101130_AABVCM kirschner_d_Page_69.pro
12e4e0866e0338e1697c1b7aaa00cc0f
2e33d5ee52a23217264ffe13763c7eadc5ed7b32
82757 F20101130_AABUVO kirschner_d_Page_52.jp2
ff4dd108de4a6ac818f1290c4b677047
78e07588015a2b625763853795399c729375008a
5702 F20101130_AABVMH kirschner_d_Page_68thm.jpg
c59605a94337ade6e46820503ad2ccc3
c223aaf3749bf677bbf5195b54f617375d30f04e
5399 F20101130_AABVHK kirschner_d_Page_49thm.jpg
e11017e45bf78d20e15a279ecf7cf7ef
68fb050b66d3c1540f45d7177864f1421464f394
25556 F20101130_AABUQS kirschner_d_Page_07.jpg
8d39e6d7a3400681d59246cffbf85225
444e63a768b23f57283726d7335bf49ad9e8070f
786 F20101130_AABVCN kirschner_d_Page_70.pro
7d237b2df32efc310cde3135f2306693
ab08a30654a9bb68bfc06799815fbb6fd553d069
97502 F20101130_AABUVP kirschner_d_Page_53.jp2
be8eb768330bcb98b0a7f5db6489ad5b
9ff05e3476af71ee28016be85916b47ca5554f4f
19710 F20101130_AABVMI kirschner_d_Page_73.QC.jpg
0dfbdd86f8c154255102ab04fa5b20c2
9022660305e20ceac49e13bf5079716833451818
23101 F20101130_AABVHL kirschner_d_Page_23.QC.jpg
12f7732d9ebfcf155b187205fdbbd1df
f21f5a3326a27e7f401cd47a31d84c7a0dc5d73d
60880 F20101130_AABUQT kirschner_d_Page_08.jpg
e98cb886865ad4cba6f1a4f5bcbb6818
ca6e7972ba240a6fdef1bf50fa959c9bbe35dc9a
428 F20101130_AABVCO kirschner_d_Page_71.pro
ab083312f333083142e26d72558ae3cf
5915dbfa182c2949033a00e4c3a32c817df0fb79
84978 F20101130_AABUVQ kirschner_d_Page_54.jp2
88af5640aaaa452f006bee1af0ed3095
464fe041d68c090de9b85b78c4089b5a0351c44b
4617 F20101130_AABVMJ kirschner_d_Page_74thm.jpg
e89667f2a33f14767c18e91139e7ea30
d273d8f89943aecb0c0d4b112d9981cfdcac135e
16153 F20101130_AABVHM kirschner_d_Page_04.QC.jpg
c6b5fc3b2a5505b4dc1138ee7a15299f
2007d13ba0ac80a8075dc55183984b0b39b5697c
67954 F20101130_AABUQU kirschner_d_Page_09.jpg
f6ccd5adfdb1e49146c4fd4721873e43
d32c049cb4974d1a82b0dd796563654d2c42ecea
92477 F20101130_AABUVR kirschner_d_Page_55.jp2
981dd7659786d68ca3c1f097b1fb642b
54cea6feaf1aa2a2184c873026092f4ed6f36c6b
1409 F20101130_AABVMK kirschner_d_Page_77thm.jpg
1401d39d92015cda39fe8f8172929ee5
de7b758e0ed4971a756bef82b6940652a5268d0e
5357 F20101130_AABVHN kirschner_d_Page_08thm.jpg
1cfbc8a571ddd630ce429908129aac18
f8a7b0b137613ed2627ed23f51f836c6ad7d260d
63670 F20101130_AABUQV kirschner_d_Page_10.jpg
56f8aa16c7449bdd9ef924805439a64b
043f4f5ec8079d18cc3310a0d003b92fcb9f3826
1577 F20101130_AABVCP kirschner_d_Page_72.pro
1fc0fb5cacd677e4402e4f1fede66bac
97aea49f97212fe57ae78833464d50c4baba3cd3
106919 F20101130_AABUVS kirschner_d_Page_56.jp2
277b5167d8be76e80bf3908d2682e04a
dea4a2a3b3ab5f9b6861bbec17673226983fea3f
3151 F20101130_AABVML kirschner_d_Page_79thm.jpg
34f44f403872f1f1ec69825fc7f6bd3e
d98c06760d964fbe6d0ba25859dc40617adad4e4
10080 F20101130_AABVHO kirschner_d_Page_80.QC.jpg
203046f9c093d3992f1cc9e98be0879c
77bd55ec84ac8a8ae15d9111a29cf998e49537ad
70678 F20101130_AABUQW kirschner_d_Page_11.jpg
6e22a85988c0e75e3c418d3d6642e606
ea3ea12dfe86ba1655f11e7aaee7bd42d3998c6e
79105 F20101130_AABVCQ kirschner_d_Page_73.pro
2192e92c53a900d3a498abef9ae9cde2
3544dcf888c10188fc13694b352780248a3fc1a3
87998 F20101130_AABUVT kirschner_d_Page_57.jp2
ad2fca312c81fb7e96424768585c81ab
0208357b377877cba9bc268706f67d1e67605185
19109 F20101130_AABVMM kirschner_d_Page_81.QC.jpg
1f0e0106063a5e96f715780ac7ad6e68
b2335e5d9488284f73d9f8bee9393b8a7d266c78
2730 F20101130_AABVHP kirschner_d_Page_78thm.jpg
7300fb09b27eb55c5a799925c909694a
8d680dc421f5e9948337dfb12cdda75ee7df1031
69662 F20101130_AABUQX kirschner_d_Page_12.jpg
715dc21595b707d70e6c22131e5f4fae
2bcadccdde6ad4a4a0adc4d9b13b220984ed50fb
45579 F20101130_AABVCR kirschner_d_Page_74.pro
d58c9d828708b51f45fd8e4ebfe2ea0c
6c8ec097b39d310972f170ef1a4cca13289d90c4
115681 F20101130_AABUVU kirschner_d_Page_58.jp2
cc2fda2364b0de0e828e6a9def5037e7
641a640d193a96fe35d893d99b694d60a13b9dc7
21748 F20101130_AABVHQ kirschner_d_Page_58.QC.jpg
8815a57f47c794cc8908a5c775597f3f
b2ab5a229453b227a369ecd366a7012cc555d583
53479 F20101130_AABUQY kirschner_d_Page_13.jpg
178f5723d54f71b4a413ec3d861401ee
ee85e98fd76d578b9b99f2afe795a0db68cb107f
13799 F20101130_AABVCS kirschner_d_Page_75.pro
be860a4e81041a3207e1a428f3a095b2
71e67cd40e4225f4113e974c5b695d5655a74bc3
83039 F20101130_AABUVV kirschner_d_Page_59.jp2
34f11143f8eeda385619aed8c6dff9c3
2bba6f9106fa9887959e5e8add481961c6e93f5c
19311 F20101130_AABVHR kirschner_d_Page_17.QC.jpg
9b5c87f1ebf82b4c343509fd043394b4
70d87267c478912076427d1b3f9f657612457c06
19493 F20101130_AABVCT kirschner_d_Page_76.pro
0f1ee83368bdecff121a76934e9bead0
9d4175004c5ec1739790ebf999dcde7d3ba97cd8
749183 F20101130_AABUVW kirschner_d_Page_60.jp2
71506995e80fed3f569da19c70fcecba
ea142978f8a00a2f2d38340ad3ad9bf10f7baae0
12476 F20101130_AABUQZ kirschner_d_Page_14.jpg
b57643bc4d9f10b1a00c7c6bce9c6402
86cac42aeba355955679583542fe170f142c005e
3360 F20101130_AABVHS kirschner_d_Page_60thm.jpg
598a8491cd89263aad1b93606ed3aca2
fa80b07104062ee75a86c6f4e2c4b47f96fda4dd
1239 F20101130_AABVCU kirschner_d_Page_77.pro
821ea0580a6f3f1ce1bfd201b6ba8c18
615b40f137ceac97eecb1031d46e126a9e848f00
1051947 F20101130_AABUVX kirschner_d_Page_61.jp2
e21c113f5320d4e78fa5ba8ccbc3c6ff
785c96f91f33fab6fdb9fe46d6c7b2660b6c4360
17284 F20101130_AABVHT kirschner_d_Page_65.QC.jpg
a96a933d5bbb8adb5803cf476886bb77
19ce6a1ab7bfc177e1926697c7a939937f488c8d
109346 F20101130_AABVCV kirschner_d_Page_78.pro
2046da74b6754e777d61c42c3338ea73
59d5dc597480daf77222af24fb91a46d1b941575
50735 F20101130_AABUTA kirschner_d_Page_68.jpg
ebb946cb05132d4aee71cb55136ad709
de3d85ba276a3e266c1016dafd36cbb7adf6e1b1
1051880 F20101130_AABUVY kirschner_d_Page_62.jp2
87972d2e2433b14dc2d3e876e91ce02c
26cb0db23ea65690e5a7785335031800aba46c09
35707 F20101130_AABVCW kirschner_d_Page_79.pro
40d51a0acbf18450ecbcde600479e7bd
558c21c4dd1b79771a442ea23ccd11486222f58c
904593 F20101130_AABUVZ kirschner_d_Page_63.jp2
7cfd161a0d3b1aa3bbbf8f1c2116a81c
b2cb8b5db9ff90bcfbb99c694327b3b2a70eedaf
4769 F20101130_AABVHU kirschner_d_Page_24thm.jpg
b4b30e9084aa4d3b8dd5c398dc59896a
e1054451c2e34dd0f709534476143bb1adce2783
23650 F20101130_AABVCX kirschner_d_Page_80.pro
3dff4adca1f01ff9ababa9723a3fda94
a92b5bc43f254506b03cc3277ce450973862333e
46860 F20101130_AABUTB kirschner_d_Page_69.jpg
0f4bb03194aec948426dd6101c556638
46a6b63896bfc3006ea2cc0081a0ea31f48e5ecd
6675 F20101130_AABVHV kirschner_d_Page_20thm.jpg
b25da43a9cc11bc55d1b487ed6a67c52
2651c8935ac9f05faac870950b50bd2701b1f5bf
21500 F20101130_AABVCY kirschner_d_Page_81.pro
c83367087970ff909848d0a85a3d9a89
27bc908403db2c4672adc25b39e1c04ff3934389
54766 F20101130_AABUTC kirschner_d_Page_70.jpg
f2c3062156c2d188435c5e86434fa7fe
c04d6b5c97b8fc04cfb9f1132553caa4fa5b2e90
F20101130_AABUYA kirschner_d_Page_34.tif
0fd8dca102da1f949e0ccd2f35c59908
65ec420523c30e6fde05328345eb4620ec1d3288
62483 F20101130_AABVAA kirschner_d_Page_04.pro
e5647974d7e1864571398c6af6f07853
a53f3936c14d3dc056e5921f53c25f75b32d3b1c
22789 F20101130_AABVHW kirschner_d_Page_29.QC.jpg
6ada9d5f074de66706b292d26457357d
d2659ddb36ecba612f46fde507ab4ad6f7abf5ec
8590 F20101130_AABVCZ kirschner_d_Page_82.pro
59c75ea9ef474035579283c2c4d1b0c1
f11dde14911551180326cd9acb457b8850171f3c
31501 F20101130_AABUTD kirschner_d_Page_71.jpg
b2bae004efb628127b01b828af9957b4
bc14217d44f8a6c2667fc500c9539a626a417cb7
F20101130_AABUYB kirschner_d_Page_35.tif
485e95d97063a413d72848a93d129624
bb283a776c751f74b05e98a64f764b0b4a9a5553
17655 F20101130_AABVAB kirschner_d_Page_05.pro
702c7f29aa593725121b708da77d9f94
c949a31e0765e6fc833381afeeb85e4f8ca74a61
23419 F20101130_AABVHX kirschner_d_Page_56.QC.jpg
1de6b9918dedbcf3b71624efe5da7eef
b9cdacc376ba554c3ccc0b69d11e8a812efc025a
10960 F20101130_AABUTE kirschner_d_Page_72.jpg
c1499d6d5355af741ea898f11c90a384
41af638199b6696126bf7bd281e2d471b06d95b7
F20101130_AABUYC kirschner_d_Page_36.tif
3ec3fd5c4b33f6c3fc7a6e8e37fa4c9a
3b92729f4596b6abfecf75480bfcfd00690a095a
18354 F20101130_AABVAC kirschner_d_Page_07.pro
648ae1e137376387fb5bcef95c0e6253
b3a0415bb46205e1bae9e890ed290b7e17fb440f
6632 F20101130_AABVHY kirschner_d_Page_23thm.jpg
323d8927119c790ac312f9f41a646b3f
b64144e3d0b93c8462bc5e1d01c273958a9c887c
76700 F20101130_AABUTF kirschner_d_Page_73.jpg
573bdef1110dd1c41592c256e1eb50ed
72b00b8c314e3aedbc78ee1e0612989879758145
1766 F20101130_AABVFA kirschner_d_Page_53.txt
c7c4476cc7d2defcbbfe5f7ff8e1dcca
5663b98afcf99b2aad415ee63cc0cb574a3d053c
F20101130_AABUYD kirschner_d_Page_37.tif
c5731366ed70b86ee0e57ab43fd98e3a
35ab44118a31be287bbc584e2bca7e694d96b8b4
41036 F20101130_AABVAD kirschner_d_Page_08.pro
2c8e11b52b3557b9ce6bde7f058c6045
ba5569096b1f3df16cb75c5898fb1273f6ce4a9f
2513 F20101130_AABVHZ kirschner_d_Page_37thm.jpg
6016d7802b6816a9fbd124f4253d9516
bdb3ae47760c414d6307b95b8c2663859dbcb665
54760 F20101130_AABUTG kirschner_d_Page_74.jpg
cf7906450d6df95cc652f88e98428435
aa23062df05c882cfc651e4965d577ba343e4944
1588 F20101130_AABVFB kirschner_d_Page_54.txt
b105a1e749966f1e141241f5cde1a73d
e829535ec86dd6ddb3876c946db5d0a8d80d26d8
F20101130_AABUYE kirschner_d_Page_38.tif
ee7bcbd734c81cf13e93bf0322ebae4f
9af45963ad1b5019a9809a5ef95a033cf3baf001
48771 F20101130_AABVAE kirschner_d_Page_09.pro
3a787984f4112946bf8c3c1405bd7431
59c2917cd3c303a31668b48e0c60f75cb3878f7c
42677 F20101130_AABUTH kirschner_d_Page_75.jpg
2a0d77ae2ed4971193784f9d7c061a97
9b7ca46f69c2c2955041bfc3f0c1652b6eb74e92
1992 F20101130_AABVFC kirschner_d_Page_55.txt
ae6078b8057e7d22bf1c047f159f49bc
a741b7b9f865e2a0a38f468bb95d709bd98089b1
F20101130_AABUYF kirschner_d_Page_39.tif
0cc2dccbd7c9b85b4c559813477e079a
f3904e74009072440bdf1979d873a977c61459b5
41280 F20101130_AABVAF kirschner_d_Page_10.pro
424370652c59d27baf9561e29467f2f1
d2c8e000dd91fd0905f71becaf58f2a4c629b837
6868 F20101130_AABVKA kirschner_d_Page_36thm.jpg
5ee4abd40f6eb48ccda336f70ade8499
779d415539649ccc66ac2bb65a9ea21cbc8f7059
37375 F20101130_AABUTI kirschner_d_Page_76.jpg
dfc89046a630b61935dbd6b49d7f0948
4545bba3c0892d0a8bacd026c3a41bb77471bede
F20101130_AABVFD kirschner_d_Page_56.txt
dbcb948eabf8fd8a294752a059f80887
c5c42c766410ef67b962bbd7c2c6e9d86995b565
48352 F20101130_AABVAG kirschner_d_Page_11.pro
5a76b5f97a2273a6b8f8b12baf3f676a
ef8c2606d7fe905aaf1d12576aaf6ece800b6353
3767 F20101130_AABVKB kirschner_d_Page_76thm.jpg
3a80981f9504fe59ba26bdaf9a345c58
f4c3c44dd02acb2718f0d1d0617333e79e10ded8
10431 F20101130_AABUTJ kirschner_d_Page_77.jpg
3b97d66403347b023306eba5a88f060a
7cf2e63c4ab82fe20dd87b10dc12658022647082
1620 F20101130_AABVFE kirschner_d_Page_57.txt
0921aa9f1c8bddd877081d394b53f70a
77de90c90b866e15f1288c5d16ca85a3a1795589
F20101130_AABUYG kirschner_d_Page_40.tif
605e298fa60927c8760b60993b6cdc66
5f65d5cc7c68b6035bb47191ac9bea78383cb528
47211 F20101130_AABVAH kirschner_d_Page_12.pro
7d969e92e88bfb2821a0cbd245e59145
bab70c25aac5de47ca9b277aa3452f54cd579b08
10430 F20101130_AABVKC kirschner_d_Page_43.QC.jpg
6a57344bbee1ed602684d0a4ad391b11
d37bb4647c6d9f9abb75a568dc71ca35580e26b2
33556 F20101130_AABUTK kirschner_d_Page_78.jpg
ba018f373a5b40b11eb2e627b40f8940
ef264810a374b78f822c5a2e6dffe0dfdb336d2a
2220 F20101130_AABVFF kirschner_d_Page_58.txt
669a01fab2c4fd765ce8d6945e0f32b2
2848ba781cea33acea64cd099113b5ab05845069
F20101130_AABUYH kirschner_d_Page_41.tif
79fd44c115b557cf3d0d777be79d0247
fb17674f7a275061d528b851c7fb46176486627c
35427 F20101130_AABVAI kirschner_d_Page_13.pro
cda6a66b5da46b582aa7e8cde16b041a
2d9b485ef9279b01a7417118c8040638ee810b8e
6300 F20101130_AABVKD kirschner_d_Page_32thm.jpg
abc7dbc22a5819f7581533f6a17786cb
b3ce16e969515e95de90a51dcd9df973fefdb9e7
26399 F20101130_AABUTL kirschner_d_Page_79.jpg
2cdec4622b5ee42d16d2cb52765fe00b
5219fea11824be1b2a7e33f5e9063bc635587d10
1531 F20101130_AABVFG kirschner_d_Page_59.txt
4256ed70ed749e233d781e4e87ea8011
266a7454c7937793c002064055148660a245ff55
F20101130_AABUYI kirschner_d_Page_42.tif
43c29fae52f7ec5a8a1dce79f89ffcfd
5a75e168c2c0e9f7158d253e626e3389d7edb92a
3381 F20101130_AABVAJ kirschner_d_Page_14.pro
6c25ea8d350ea05373ad9e292155295b
8308d050fb9634617aa29211ba946d6e2a603688
5951 F20101130_AABVKE kirschner_d_Page_67thm.jpg
30b7b9614ca771ea521e0d2c9b62ba41
403b1fb70897bc9e80a28a3fef82972b7a88f921
32107 F20101130_AABUTM kirschner_d_Page_80.jpg
edbdf679f1bf93c339f9a38c37e7b574
c5a57b5e5b0bc7977676d2bde920ea23a420a12d
290 F20101130_AABVFH kirschner_d_Page_60.txt
96b13f7247e5236133b0a27b0903624f
758a81a12b21a206dbbb2ac8031f42175bb8156d
F20101130_AABUYJ kirschner_d_Page_43.tif
dfc15969e7abc30192008ee306c5b72d
7044e2a0b1e1492f8585bbc6fe3f2f38e9d12f68
38816 F20101130_AABVAK kirschner_d_Page_15.pro
0b32c5ce430197aab52eddb08840521d
ad53006fc9044e4841be46b9503b7e4312228b0b
6011 F20101130_AABVKF kirschner_d_Page_65thm.jpg
b93e8ee407d932b095e77f8bd9b599ba
9caccc447c778cf81ea5296ede2e74d3ba4c8bc0
64174 F20101130_AABUTN kirschner_d_Page_81.jpg
538a37f25bffeb84c0a0a385e14d7554
378f012cfb4eb1bc527f1b86ac90d8b446c33dfe
482 F20101130_AABVFI kirschner_d_Page_61.txt
7885becaccb829ea6a11d33343ca77d0
11ccf09089cad83713374094ff14407fde7d8815
F20101130_AABUYK kirschner_d_Page_44.tif
72537c6d844426ac4e894ab3e20429f7
d74b56ca00aafd658d9f7948cb8f3b4db18fcbb5
51168 F20101130_AABVAL kirschner_d_Page_16.pro
dad5ebaf553ce1e2ae5812d94df40101
7733b3b3250c8f2c289b303adfb686ea24d17aa2
20731 F20101130_AABVKG kirschner_d_Page_10.QC.jpg
ab73ccb1257b17c6a9c8386440e79e53
017d837336a055578b826749285f69ac5fb72f66
20296 F20101130_AABUTO kirschner_d_Page_82.jpg
0c084695c0482c84f1cb73348a18999f
93ea73401f0e59feb0f1db0020b63198c6be91eb
260 F20101130_AABVFJ kirschner_d_Page_62.txt
ab7ee50af30da556afad6cab5068dcd6
c0cc3ab6782d355db24748b7d94b2225c8cbab44
F20101130_AABUYL kirschner_d_Page_45.tif
4753cbda2d94814c4ca911a36236e55e
fd9bcdcfb59a7498ec1a38f0020d74034d68b079
40310 F20101130_AABVAM kirschner_d_Page_17.pro
229f04a7266f674a7802a0e1bbf46d84
7307a94f1f265891d644a0d96a52ae92ba2cfcef
5782 F20101130_AABVKH kirschner_d_Page_25thm.jpg
4bc5aa6bbe4b817a11479c6b5788191a
9884e7c716e5e4bbf3b5c3f80ce644f66210b762
24703 F20101130_AABUTP kirschner_d_Page_01.jp2
a27dec72880e71748fd9b186b56adc24
d2a6a46a3e525694779ff091fda2f7eb1743965c
F20101130_AABVFK kirschner_d_Page_63.txt
dfd24d5a7f4fe3e28f249df1c82336d5
c3cd70564db20820f31a4432d43ff58728289617
F20101130_AABUYM kirschner_d_Page_46.tif
9e1740a705c3e81294978e2a4c3d3276
7bbb0a02c064903015c617219ec6e945dbf7fe73
4424 F20101130_AABVKI kirschner_d_Page_33thm.jpg
10f4d51118a72f393fed6bfacc47a351
189650979dd34d2d9b5c912550bb8c59dd16df69
5891 F20101130_AABUTQ kirschner_d_Page_02.jp2
8d841a7a5074089d4276a4aedc5dde13
fb4fa5ee40416cfb1f0cf15ecc02d6fd08dc0eef
81 F20101130_AABVFL kirschner_d_Page_64.txt
8fdba33f6fe3ee812bf817b4e6445c48
c07fc4cc3235a8729041a7e2e802da6547dd78c5
F20101130_AABUYN kirschner_d_Page_47.tif
7a005fa035033c7087ab13c48d1c4b39
0959aa7153dd71b67c0af7873b30553e7ef334c1
52529 F20101130_AABVAN kirschner_d_Page_18.pro
45ff21fc31b3f1330021a576c73e2b62
7d1471c3e5b3c0cd7e39a27b1b51aa9e2f4f9aeb
2287 F20101130_AABVKJ kirschner_d_Page_82thm.jpg
2be468ce6b8e61ec2d606ad9609f637a
5da92f0b4a21c3f09060aae24c50dfe2d374ea12
13158 F20101130_AABUTR kirschner_d_Page_03.jp2
9dcd2ecd7e75a65b451cadd4dd4cccf1
2117cb42168e36768a415e944351416a1881bfc8
86 F20101130_AABVFM kirschner_d_Page_65.txt
2638f7f7509a1f405a99f5bd2b2900d7
74a9d9cf1da8dfc9fad83e4cdfaf298e64e09c93
F20101130_AABUYO kirschner_d_Page_48.tif
b13f29ec96e39875b93baa1f98fe367e
b6d084b067a9e654715bea1dca8fb3a9468777fa
50158 F20101130_AABVAO kirschner_d_Page_19.pro
fecdfa1834163107d9a9a8b784710b7a
1bbd9044f002c51a880472a235144b27684b0b26
20694 F20101130_AABVKK kirschner_d_Page_27.QC.jpg
4c3470cbdd63e5bb0355075e2fdbb4a9
45e8693264a970403cdbc312b4f16898d00fd7eb
1051984 F20101130_AABUTS kirschner_d_Page_04.jp2
301c11954ab825b5c74f44dea43905f3
573b6c355c2faf867926d9fd933187661f3990d1
13 F20101130_AABVFN kirschner_d_Page_66.txt
47b55ed0608369c58e3462428a5009ff
e27eab310b5e42efc23c62d4880f47a985528441
F20101130_AABUYP kirschner_d_Page_49.tif
ce1aad7ba33a93dbc1ac6c530a5e35f9
7ce35a7d9f4ccf2e31744e1e73cf286de1c6c827
49934 F20101130_AABVAP kirschner_d_Page_20.pro
f6bab9ded1a752a39b0bf866a5e912ef
ecb4d4e6c1d748106b5e32e50d5eaaacfe3c045e
2692 F20101130_AABVKL kirschner_d_Page_38thm.jpg
2820f067f4397c071223f43902c2a952
51c30b9c421fb8c65a9a67232b7da39562cdd452
600252 F20101130_AABUTT kirschner_d_Page_05.jp2
f4b1097c3f2403d6fa72546822d4ec28
d304e7a7b745d8433bcc032b97f1b7ddcbae1995
33 F20101130_AABVFO kirschner_d_Page_67.txt
11b2c61ad510bc46dc73f5d6325ec28a
57a9b6379a0a073de3966c9d36c6b0b707510b2f
F20101130_AABUYQ kirschner_d_Page_50.tif
3c498d2fd762f84a964c7ace43d41dd5
9d3c680ac129cdb3338245a0a213a707f93b1974
50686 F20101130_AABVAQ kirschner_d_Page_21.pro
6d9cd6091b2202b6ddeb430286aa227e
71c4757a4ff7d7b656b97804b8e41e58fd4095f2
5792 F20101130_AABVKM kirschner_d_Page_66thm.jpg
1ddc8f450dc3288fe5a82f92005db90c
c06971a3507bffaa271cb49b4aa3435e5cde41cd
932340 F20101130_AABUTU kirschner_d_Page_06.jp2
cd57f635563eec4200782226bd27e8a9
9645b84163d5af0aa6c4a9fd49172f1b40520797
F20101130_AABVFP kirschner_d_Page_68.txt
f215f46ad69781b87504099e1ff620d6
44dbd983b15e997f9b0753aa4df9849ea97adbca
F20101130_AABUYR kirschner_d_Page_51.tif
c5447a628e4924169c05064ba1d99b03
386e97b9da250d0d8e4c2cc4865c4d33f6ff39d0
50859 F20101130_AABVAR kirschner_d_Page_22.pro
7e1b718492255c2c512473a49a3e97cc
9674be0cc4c23521dd45d27b246ea2c3872e6e39
16944 F20101130_AABVKN kirschner_d_Page_66.QC.jpg
51b81f9e37af75f4f5b428e96949575b
6e0fbb367b85218e189da09e959843bbf093fed1
F20101130_AABVFQ kirschner_d_Page_69.txt
0e6d544d47aa3a00c614fe163f3b0f64
6906361dca48227cc5aebdb4da8d04d190155f61
F20101130_AABUYS kirschner_d_Page_52.tif
2a06b943aee00a9dbf975845f5f1c635
c72f27419443371fe754f34f09503a44c733608e
49341 F20101130_AABVAS kirschner_d_Page_23.pro
2d6d047be2c94466263f7ce8fe2ee0a3
e89c7c0bd5fba69a0bb3f3a77d874a8b7304deea
531533 F20101130_AABUTV kirschner_d_Page_07.jp2
3f9368e8b7b02d8a445eec317b83d965
713a275f1668613683b76d8e7377bde88197ec97
2476 F20101130_AABVKO kirschner_d_Page_07thm.jpg
b6338e36ef06b100133a61260a039ec5
a7d48f390e8b49ef1bb540fce025a6eea0f59133
150 F20101130_AABVFR kirschner_d_Page_70.txt
94dd75fd315072a50b9063cf5699efea
73874615b907074a683030a7902ad9e3abe580f1
F20101130_AABUYT kirschner_d_Page_53.tif
efb1da0ab028c53e22c670138063d6c0
1ae6b0f32e5417b6916d40c330378a7ce875fc4e
31792 F20101130_AABVAT kirschner_d_Page_24.pro
515ee13e75ce5e6a7d852cabbcb27a65
054df67013b8e00270d439c49b21f288dfbd7bd2
86888 F20101130_AABUTW kirschner_d_Page_08.jp2
8c3c6238576c8545f54840197c12aacc
7a9e01e6738e08890bb887942c330b66fbc9b324
17315 F20101130_AABVKP kirschner_d_Page_70.QC.jpg
0808b5f0bbecfaf5ade9df5ed2160101
d95c66a235e4fb05fad09cfb07683bd8bdab8a60
F20101130_AABUYU kirschner_d_Page_54.tif
743c49f41cfbe61d0a99fdf116643682
0a5119ee1a65961cc308f402e2b9d0d73494696e
41143 F20101130_AABVAU kirschner_d_Page_25.pro
fb99d374fe1702b276728e3d09e07b72
0591df7a0c12d6088f65b1f28732be33cc66f08e
100998 F20101130_AABUTX kirschner_d_Page_09.jp2
c0860fa1915eb75f66ee32074970493d
37a7f50643444884becc7ccf53908a67b55614a6
5026 F20101130_AABVKQ kirschner_d_Page_44thm.jpg
9eaba1aec81d7f89fb6a2790ceccde7a
2de958fd402f308a610105713f64f1c210483e22
106 F20101130_AABVFS kirschner_d_Page_71.txt
acf3b2cbe976f6926ad5f6cba20d8466
982db3cd8c0bc7fe1ff83c5cb1596d757e317e18
F20101130_AABUYV kirschner_d_Page_55.tif
9b4b0c75426962024c2838804d57b970
665f289d5e41c4900a90effb4b473d913fb07a45
45357 F20101130_AABVAV kirschner_d_Page_26.pro
e205facf27fcb13fdd40eff5c666eaa5
86c8609fe54d42872585e9e2ed3a6e604226ad3d
91722 F20101130_AABUTY kirschner_d_Page_10.jp2
5900048ae235a950696251fdb2a7a09c
7712714264a0cd7869397f760daf96689d6505fb
8554 F20101130_AABVKR kirschner_d_Page_38.QC.jpg
1f814cf19c6cc321aa1964dd88ab22c4
9c5892c9d1edf50620ce90c4572558dd1ddac876
105 F20101130_AABVFT kirschner_d_Page_72.txt
3518c5e6c78245d0d3aa4afadb28c97f
3c546ac2dba29f8dffa2e8dd1a97a4f04a71f87a
F20101130_AABUYW kirschner_d_Page_56.tif
d4e8884ec26b54a31b5ca49e2b9558dc
413ce46aa5de5ab167b893ffc9e3ff5986611ed5
42375 F20101130_AABVAW kirschner_d_Page_27.pro
c55b4e4028f1b7ba8c77ac2df5c5885a
148a2891351fed4f33afd13b92524532f2c7b488
60327 F20101130_AABURA kirschner_d_Page_15.jpg
c58a5dbe46ff0ecfb6c35dc8c5a049f8
785164517ae7efd2fd074545a297b45dba1aafb3
105801 F20101130_AABUTZ kirschner_d_Page_11.jp2
7ace048fc58665f745be261c5945dcb9
e0c7a5d58caec5ea50658b999f630d575a98e455
21978 F20101130_AABVKS kirschner_d_Page_41.QC.jpg
0c587d82f963277b172a4c3c2c3d93c6
486e1a8b49698849b3d3e22e6b284b6aed70ebca
3641 F20101130_AABVFU kirschner_d_Page_73.txt
5522de64e02b43d6d7d0eb8c33debb4b
16f8c8516d1afb288b11b6f0dad14bcdd191ca96
F20101130_AABUYX kirschner_d_Page_57.tif
98e7295a64cedb3d3f3966d04b3f88f9
182a054a1e4ad6bca615246fefbfa3041840db5d
38852 F20101130_AABVAX kirschner_d_Page_28.pro
3df0623526ae65878039a680adbb62ed
3c338852377acfeeacd3899d83aafe932d88d5ef
73209 F20101130_AABURB kirschner_d_Page_16.jpg
adf2fc11d9558b9a8634025111b90ef1
8bd4982eacf70def053d6365492b5152da9e84d2
23567 F20101130_AABVKT kirschner_d_Page_20.QC.jpg
c8d2fd0ee26c464afcf54a99f5b8bb6f
beb53f54c0e75609588ef0ea2047ee779a69f896
1983 F20101130_AABVFV kirschner_d_Page_74.txt
cbd5cb65e7be7bc73a7c10f3333f2a2f
16d0f2b1ce015e570b6b628a879f2d4769eafe01
5745 F20101130_AABUWA kirschner_d_Page_64.jp2
0a865718989d688d88374c7ca73af715
4ee0ce38c38d13a34c242d11de9909f0788c18d5
F20101130_AABUYY kirschner_d_Page_58.tif
7df4204bae290c0ef2adf5f511ea6c9c
e1dddee8868afde8d125a0f95184a8a323d878e4
47601 F20101130_AABVAY kirschner_d_Page_29.pro
c477505e1082f3e61d0ec8beaa271007
85f4121289547985a647ddb466f5934302624654
57925 F20101130_AABURC kirschner_d_Page_17.jpg
fc1329b28ee91ec7a2c88e1dd353d1d1
9dfb7c146d05a521799e866cd1d31b59eb508bf8
1371 F20101130_AABVKU kirschner_d_Page_02thm.jpg
2f9d3b8c4d1e2b88bf88b137d709d0ab
9593f6d139c6cbb05ec10d197b9b043bb2dfa35c
769 F20101130_AABVFW kirschner_d_Page_75.txt
807c15b6760b6c46a2dfff58a53f41c2
1b4e65f9c585b309f15af3986a0eccfd35fad963
1051927 F20101130_AABUWB kirschner_d_Page_65.jp2
f2b5e3235acf5f4190ab1ff1a48fdaa4
63d04a1ae9ddd33d3bc60e6a69a3037236e8181a
F20101130_AABUYZ kirschner_d_Page_59.tif
30b47a537ecfcf64e6202b36f997ae44
d32fa3bbd7bee936303f1404c571b1d69dc4704c
24881 F20101130_AABVAZ kirschner_d_Page_30.pro
71cdc04c4975292ae4ccd05ddd969f3d
18d52f1a4a7c5516e2a31bec6816bb3bcf9f941f
75076 F20101130_AABURD kirschner_d_Page_18.jpg
3a77ea088e8062b73b139279f9ad8c3d
c8bf5d84a5c9d4eed5f548062e7f83ba092b321a
23163 F20101130_AABVKV kirschner_d_Page_40.QC.jpg
6fffc5852339f53e3a452434bf16df32
303442b513916e50c0a8113a331b62ed18b29be9
864 F20101130_AABVFX kirschner_d_Page_76.txt
e52482f2918c2912786faf8543258e94
bf0a15d5197cd65c26445dd904ad874e8f8bdce8
1051967 F20101130_AABUWC kirschner_d_Page_66.jp2
3f778f35e9426ca145598b1f5e90e759
d5657ddc18328851ee827a1157c6b52b08bba2a4
73127 F20101130_AABURE kirschner_d_Page_19.jpg
2d59fb80e6f43043745773e8cb32c1ee
e5aff88c898ef8d4226333b52f89dc959946aab3
4475 F20101130_AABVKW kirschner_d_Page_45thm.jpg
aedc1698f26bdc6b54c426416b108928
23ebe49a2cfb233e22ac094806736b0a8bad2c52
85 F20101130_AABVFY kirschner_d_Page_77.txt
b0690a1f18daa2b73a5f1e0b340b2960
b55f01c883bb2755111ea4e0f50c95aa41d3eb95
1051945 F20101130_AABUWD kirschner_d_Page_67.jp2
4530b576ec45890ded9973fcb0a9c2f2
d62dba5f39b499e748b41b4011008b3f4f98fd53
72188 F20101130_AABURF kirschner_d_Page_20.jpg
2c83fdcd108e53d310d8e118dafbbf38
133eb8cac4f84797d6f89a377400388e1bd1babc
468 F20101130_AABVDA kirschner_d_Page_01.txt
3497829cdc0c6c96ed4b76db40c82864
48b10311296ebcf717ba4357b6a30176bee82258
6206 F20101130_AABVFZ kirschner_d_Page_78.txt
76bbf9b108826854252d53a5274fc873
80134e6614d74597e66c56497741261f7b303c39
72570 F20101130_AABURG kirschner_d_Page_21.jpg
413dfd243b9f57c00ec90b23e6668505
61923b2c2e3d42232428d07c7fe7d4dbdffbce17
114 F20101130_AABVDB kirschner_d_Page_02.txt
b8dcdb3f079b17ed9dbce3b0c389df3e
55909e5263dd123cf4d9f4bb4dae11dbc52e8efc
3339 F20101130_AABVKX kirschner_d_Page_77.QC.jpg
cae40599cf55e16e80d972d279d7eca7
e40690e0287eab35160ef14bc540c2bec5544746
1051910 F20101130_AABUWE kirschner_d_Page_68.jp2
60793879ea206509964126c266218a50
e4105d9a41eb806586b2549428df6822a3158dd6
74724 F20101130_AABURH kirschner_d_Page_22.jpg
29e16a8d77fa7027eac946a829fe801e
4ca21e12ad9ff6cc65560496ac582ebadb386abc
204 F20101130_AABVDC kirschner_d_Page_03.txt
20f7032cbd402e95b903877b67d70ae4
f8a08c9a0d042b5cec3c8ba348a74b456e4b53f3
17884 F20101130_AABVKY kirschner_d_Page_35.QC.jpg
80554861301f02ae2336c8bd2a386b58
b38df208143ce5dae763ba5a6a045259419196c2
5899 F20101130_AABVIA kirschner_d_Page_70thm.jpg
1bfac5c9e04c7247a3fa23f370baa199
4c8f8fc388ff4998c41694aebca1df13023df381
1051973 F20101130_AABUWF kirschner_d_Page_69.jp2
5bbf0192aad10169ca61d374407f56f4
e104998294011194793761f5cfcf06cb26904da5
70291 F20101130_AABURI kirschner_d_Page_23.jpg
e1bc6da154a589c2046734099df1810c
be1110644fdfe2f2dc760912e7319b477668690f
2494 F20101130_AABVDD kirschner_d_Page_04.txt
7de62a233cd1d65cfea47f56179b1dbb
2a021978bc5ac4d2ccfd3ed2a8ab4d43d81f4244
5319 F20101130_AABVKZ kirschner_d_Page_69thm.jpg
4db48f739217421c6d9a7c58a909f6c4
13ae83b322e255d9142330d7cd980a12f9e81bc0
4999 F20101130_AABVIB kirschner_d_Page_03.QC.jpg
6c0c44e20613d2bc613e3d39ea0745ba
58df67ec0835fbcc5417ce6e3f65ee079170093a
1051968 F20101130_AABUWG kirschner_d_Page_70.jp2
da7d409bacae8e0a88559baf7f7c6688
19e7430205739466e5b5f9851a75882a0fd85d86
49425 F20101130_AABURJ kirschner_d_Page_24.jpg
b4d580c57f0df87b7cc4909846cebed1
e705fd69f0c713b0c2040404ee601d5a9348b632
679 F20101130_AABVDE kirschner_d_Page_05.txt
3063d459108890fc66f56fbc38acbbe2
93ad0b6acec8f9c8494e4e4d497909dacdd728c2
6430 F20101130_AABVIC kirschner_d_Page_12thm.jpg
7f7540d8165e18a89451451bb67eef45
d27bfeec2315e8507625b71815ed0f0dc23a39fd
1051940 F20101130_AABUWH kirschner_d_Page_71.jp2
8f56403a7c3e64f61db3fe0ed90bbd04
21d0b24aa038c1488861929ad07cb1223c651ec1
60706 F20101130_AABURK kirschner_d_Page_25.jpg
58a21883bd880a509349ba4d3472b78d
dc0c735505dd619a2c2d9ba4376cc207a6c9c5f2
1322 F20101130_AABVDF kirschner_d_Page_06.txt
80540c6223ad92cdd1960591931145b2
c767b07392097d5f7132e4ee6132edd25d3c3b88
6927 F20101130_AABVID kirschner_d_Page_16thm.jpg
9dc69bb6873da15bb8c7a1b6fab7e594
cc2ab07a7df0223a7c4f88be1304b7fb028f52b3
7119 F20101130_AABUWI kirschner_d_Page_72.jp2
40abe48919421004fb4fa186864ea514
f9fb39af607e1106d225a425d31ebf4a6538702f
66162 F20101130_AABURL kirschner_d_Page_26.jpg
b034c2c327f3b8906bdffd84b73becb1
e3d50a4a9dc6519efb4b737ae198cc6b8d012d7e
770 F20101130_AABVDG kirschner_d_Page_07.txt
1bb4f9592d5db07fb27158ea80b97ac7
cdf81160743978179c61c24d76f26fab44a38623
6654 F20101130_AABVIE kirschner_d_Page_56thm.jpg
fb8260a25e1fb501f69bc583fe0c1fcd
63f4c5099dcba259f859de2ecff9c1573abd1ee8
1051932 F20101130_AABUWJ kirschner_d_Page_73.jp2
999904fc5ebee5b7d9061fe35a57f921
a3b9c2d22c7f5e5a1c08b26c6d214db6b05469dc
62633 F20101130_AABURM kirschner_d_Page_27.jpg
26bf31b174da011576ef0152fabe3063
0217ac1e2d4663459b7ed66e3196856b590308ee
1815 F20101130_AABVDH kirschner_d_Page_08.txt
35013e97a09e6d0111d6194786956aa4
160ebd6081ce1330fe950be4c6586b74744959ad
21443 F20101130_AABVIF kirschner_d_Page_55.QC.jpg
99504d2e0feb2ab65db9353f4fafda19
859451dedd6c4e2e7b5a02585124432ca25e04bd
831586 F20101130_AABUWK kirschner_d_Page_74.jp2
d6d451673782f5c22283daf336415639
5c45ff11f91a479524b5d1da2488f7cb8588e6f1
58221 F20101130_AABURN kirschner_d_Page_28.jpg
b67c8213396c7bfa5d42d56110ae688f
fd1318541a3a9bc62caaca3d6f70150ac7ad7ec5
1936 F20101130_AABVDI kirschner_d_Page_09.txt
35b64c385f0f61755795db2e147fd34b
6bfda83cc5ffc237871028b8a89d402c72776aad
7916 F20101130_AABVIG kirschner_d_Page_37.QC.jpg
caf83eb990a0fb1060f4d5038273be1e
55093f0b77bf0df6ae561a05bb7bf96924d9bfb9
609457 F20101130_AABUWL kirschner_d_Page_75.jp2
f00f3d2aaf4e3b52419af645752cfb5d
d332e54a5304ff077f405ef000eeb2d4a53c6dee
69733 F20101130_AABURO kirschner_d_Page_29.jpg
997ad008bba53e4176e54472205100fd
52bea50220d67f665993f148cb5bcd1bef817d71
1743 F20101130_AABVDJ kirschner_d_Page_10.txt
6cb4d6b8752804bfa4252a573c7c0ded
a25f39399024aece98ef74e604c68ab1dbc263a9
3494 F20101130_AABVIH kirschner_d_Page_63thm.jpg
c7b89291a9bccbb9bf5aeb8094692418
aa792b62d136b11675f779cef758afbc48ba764a
548600 F20101130_AABUWM kirschner_d_Page_76.jp2
c833573387ebc504a786fc5403b7682a
9bcb284fa4e4833120a56680cc231ae559e3f641
20859 F20101130_AABURP kirschner_d_Page_30.jpg
ee73faa98f51c2e792718d6ca517b276
262d7f4053ee1e0a17c9fb6e5a33e76e4f298f9d
1935 F20101130_AABVDK kirschner_d_Page_11.txt
7874b2280aca46f211dc1a362132dff6
6077df139d92b9c5098a4fc23f16568d71d8ff67
4523 F20101130_AABVII kirschner_d_Page_46thm.jpg
46e41745ec317ccef7753df361e53a25
57f6a549d201e439509be34192cfae6dfd4f8b2e
6217 F20101130_AABUWN kirschner_d_Page_77.jp2
eb1a751e2bb20b29d1ffd171650e0628
4117ac71abdd966a70a0a8bd663fea8d7bf97082
72573 F20101130_AABURQ kirschner_d_Page_31.jpg
c6d1303ccb0448bdab0e5bdd367f2bcc
29cde4f7d6907a7be265b14027f44d6364c30ba3
1868 F20101130_AABVDL kirschner_d_Page_12.txt
47683cffd5b9d0a0f1a38c43c0744c9c
f910baf76fe8bf292aad4ccba03b368ff758a544
5715 F20101130_AABVIJ kirschner_d_Page_81thm.jpg
1ddd631fa0d97c27dc59ff6cb55a846d
d09fbb2ce76b109f5aa077b6d33033fb81109577
67807 F20101130_AABUWO kirschner_d_Page_78.jp2
054962d0c370570097e88708ee12c25d
a78d6040766eb2a49542cdd82bfe73479a1dc1a1
65561 F20101130_AABURR kirschner_d_Page_32.jpg
f8f9b48a66950227905f9cf7290eb5e5
458229a726909a61b383db6d4b835980488ec891
1516 F20101130_AABVDM kirschner_d_Page_13.txt
79f950cfe6d01c23e94202004086a696
8f66717c1f95bd10abac4bf958a01624540bb656
18948 F20101130_AABVIK kirschner_d_Page_54.QC.jpg
5f22f11ff8dccbf6508f04609d42e620
69d3752a627def716cac695a91863554ea984b4e
53865 F20101130_AABUWP kirschner_d_Page_79.jp2
79daadb2eb2a126d41d8be13538810ac
ab1a00bb9aa522d13016ec3c9d5b70e755c8eca8
44350 F20101130_AABURS kirschner_d_Page_33.jpg
5630ad897834022ccc70197cf3daffcc
d33c67d1c105f07fc78c0e6a6faac6f8c290079e
178 F20101130_AABVDN kirschner_d_Page_14.txt
9362acb960ae307e761e2afed9f73763
b61f76bae89bcb9a4cdc9758d072de82b8c8c2e8
21256 F20101130_AABVIL kirschner_d_Page_42.QC.jpg
5c19bd45dfd3cb570c5c7da004e1c94a
931a57c7aa857dc54633f446f75e5a6e421ee28c
60874 F20101130_AABUWQ kirschner_d_Page_80.jp2
bbf4f1412eeea73157f3a37b69a1b20c
d7be571e48b6e1d5c2b5346f0c1b504366896195
67861 F20101130_AABURT kirschner_d_Page_34.jpg
914b5bf182e1230e6f8462e89dbcc14c
dbd455b09a3d3c3d99e30bc6ae1bb2f4b4f9b147
1647 F20101130_AABVDO kirschner_d_Page_15.txt
cdd3eb424581c0ba04128152dded6987
7d4ea3087549d8b3d129ad7a78411c402661f14b
3281 F20101130_AABVIM kirschner_d_Page_02.QC.jpg
a3220e46d87810a90b369d9641d906ac
42a3552f94c762bb2e91564b59746605fc1c1845
54699 F20101130_AABURU kirschner_d_Page_35.jpg
1f8e1dcf380427fd86cc95681618b510
56dfbccc99001a52e7b5f0868404f7e9cfdf9174
2021 F20101130_AABVDP kirschner_d_Page_16.txt
8b1b59aafa5545021e2574d80e930643
404fea0a4ac750f01ec6c2bf1bc29225f11cb183
1051798 F20101130_AABUWR kirschner_d_Page_81.jp2
f25e92816824961430af63f7cf09dd81
9443c970349e75c44be6460c5334a106fffb9fca
19424 F20101130_AABVIN kirschner_d_Page_08.QC.jpg
fff81575cec9015e6c4dd82be12f1b44
a0aff09adb85cba851a84baf68563f9211c4a78d
72169 F20101130_AABURV kirschner_d_Page_36.jpg
8f737218eb66c291e411bda2ba3661b7
7a3806ee45e2d019330c437cb8add687169c346f
23075 F20101130_AABUWS kirschner_d_Page_82.jp2
af833ed6ce2e211755a059daeb502e8f
60ef4668862e20ddaed7386ae93da1943b62ee83
12369 F20101130_AABVIO kirschner_d_Page_76.QC.jpg
93e0e279e31becc538d69e1c1cc6eceb
6526dcdff1e4f6290007bb2134a71db04d628338
23470 F20101130_AABURW kirschner_d_Page_37.jpg
91ebcd6e0bac486fd20c8f6904f0c4d1
059fdad377f3510370e750c619ab4157c09aa184
1888 F20101130_AABVDQ kirschner_d_Page_17.txt
caab9dbe63a5d218adb8f11868ee1318
a3b4c7b32941293dfc748359996b0a65f4680942
F20101130_AABUWT kirschner_d_Page_01.tif
5fe9dbda90aba8873f8b5d83493f51f8
5e3fba4955dd7025a21fc3586f06955338fe1fbf
4789 F20101130_AABVIP kirschner_d_Page_62thm.jpg
23d976173ba86d453d25808d90d6a76a
7c17de9e19be041c9d6758a46a0fa55749b90a18
61096 F20101130_AABURX kirschner_d_Page_39.jpg
48df40e15cf328d0a8e85889522f774e
24880ffd027252d77f27b4e9138d51cf6ac0d10f
2068 F20101130_AABVDR kirschner_d_Page_18.txt
8a5dc5d2743d00b40fd89efc56953937
0bd483c53d017a93415814b77f6a1e0592eef052
F20101130_AABUWU kirschner_d_Page_02.tif
ea7fe1f74dea85eb71f57b2aec09cd68
c6586f83aeb308f71587d58274658fb0544885eb
6252 F20101130_AABVIQ kirschner_d_Page_53thm.jpg
3a8ce89b3f939c42e6586442c1f81b8b
0cd6e788af3cb3d96a0d51d198c4b1f3a2987262
70639 F20101130_AABURY kirschner_d_Page_40.jpg
9a34caf46ece4b8fa4d297a4de4f45f6
113cad0c01b89bfeb737189fb60713ddf8d91fd0
1985 F20101130_AABVDS kirschner_d_Page_19.txt
6047ba59506961af4e9f49a225723122
2bb734c6cecbc220c39cb9eb54ed6460e42ac286
F20101130_AABUWV kirschner_d_Page_03.tif
6a510129c1c4b579f5c44aef7676bde6
6456e6bd49a3c161e4e11511ccac15d4366a5249
22703 F20101130_AABVIR kirschner_d_Page_11.QC.jpg
77df987a491db49468c6b3ff943280f5
599ad6ea7057e14885b4fa1e9d7dc141d23faa05
65616 F20101130_AABURZ kirschner_d_Page_41.jpg
ec8701d1f16b3447e58248cf3066f9a9
0902ad59a8b9c9b200e194d14a0fc5eaec5ae4eb
1968 F20101130_AABVDT kirschner_d_Page_20.txt
4597719f2a4472aa6b0849c5940a5260
05c238cec51001252508dd971ae1534e381b8063
F20101130_AABUWW kirschner_d_Page_04.tif
3d41afff9c6c6fbba3daeac9dbfb9638
c04e2669cfb907434f3c9275f70fcbaf9d69703e



PAGE 1

DEVELOPMENT OF A RURAL FREEWAY LEVEL OF SERVICE MODEL BASED UPON TRAVELER PERCEPTION By DAVID S. KIRSCHNER A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ENGINEERING UNIVERSITY OF FLORIDA 2005

PAGE 2

Copyright 2005 By David S. Kirschner

PAGE 3

ACKNOWLEDGEMENTS I would like to thank my committee chair, Dr. Scott Washburn, and my committee members Dr. Lily Elefteriadou and Mr. Bill Sampson. iii

PAGE 4

TABLE OF CONTENTS ACKNOWLEDGEMENTS ...............................................................................................iii LIST OF TABLES .............................................................................................................vi LIST OF FIGURES ..........................................................................................................vii ABSTRACT .....................................................................................................................viii CHAPTER 1 INTRODUCTION ..........................................................................................................1 Background .....................................................................................................................1 Problem Statement ..........................................................................................................2 Research Objective and Tasks ........................................................................................4 Chapter Organization ......................................................................................................4 2 LITERATURE REVIEW ...............................................................................................6 HCM Freeway LOS Methodology .................................................................................6 Studies Investigating Traveler Perception of LOS .........................................................8 3 RESEARCH APPROACH ...........................................................................................16 Alternative Survey Methods .........................................................................................16 Video Data Collection ..................................................................................................19 Survey Sessions ............................................................................................................30 4 ANALYSIS AND RESULTS .......................................................................................35 Analysis Method ...........................................................................................................35 Statistical Analysis ........................................................................................................39 5 CONCLUSIONS AND RECOMMENDATIONS .......................................................45 Data Collection and Video Clip Creation .....................................................................45 Statistical Analysis ........................................................................................................46 iv

PAGE 5

Study Limitations and Recommendations for Further Research ..................................47 REFERENCES .................................................................................................................49 APPENDIX A LOCATIONS OF DATA COLLECTION SITES .......................................................51 B VIDEO CLIP SCREENSHOTS ...................................................................................55 C RURAL FREEWAY TRIP QUALITY SURVEY FORM ..........................................63 D SAMPLE LOOP DETECTOR DATA ........................................................................68 BIOGRAPHICAL SKETCH............................................................................................73 v

PAGE 6

LIST OF TABLES 1. HCM Level of Service Thresholds .................................................................................8 2. Data Collection Sites and Traffic Data .........................................................................21 3. Data Collection Times, Locations, and Directions .......................................................24 4. Traffic Data for 13 Video Clips ....................................................................................29 5. Clip Sites, Dates, and Times .........................................................................................30 6. Dates and Locations of Survey Sessions ......................................................................33 7. Density Model Estimation Results ................................................................................40 8. Comparison of Estimated and HCM LOS Thresholds .................................................41 9. Traffic Characteristics Model Estimation Results ........................................................42 10. Level of Service Model Estimation Results ................................................................43 11. Realism of Video Survey Responses ..........................................................................46 vi

PAGE 7

LIST OF FIGURES 1. Camera Setup-Front View, Side View, Speedometer ...................................................23 2. In-Vehicle Equipment Setup .........................................................................................24 3. Setup of a Survey Session .............................................................................................26 4. Sample Video Screenshot .............................................................................................26 5. Illustration of an Ordered Probability Model ................................................................37 6. Illustration of an Ordered Probability Model with an Increase in .............................38 vii

PAGE 8

Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Engineering DEVELOPMENT OF A RURAL FREEWAY LEVEL OF SERVICE MODEL BASED UPON TRAVELER PERCEPTION By David S. Kirschner May 2005 Chair: Scott Washburn Major Department: Civil and Coastal Engineering The concept of Level of Service (LOS) is meant to reflect the trip quality a traveler will experience on a roadway or other transportation facility. Despite this, there have been relatively few studies that have tried to measure the association of prescribed level of service assessment methods with traveler perceptions. The objective of this study is to provide insight into how road users perceive their trip quality on rural freeways, and to examine how the existing service measure (density) relates to these travelers perceived trip quality. Study participants were shown a series of video clips of rural freeway travel from a drivers perspective, then filled out survey forms indicating their opinion of the trip quality provided by the conditions in the video clip, and ranked these video clips on a scale from Excellent to Very Poor. In addition, the survey participants were asked to give background information about themselves and their driving habits in case these factors also turned out to be significant in influencing perceived trip quality. These video clips were matched with inductance loop detector data that were collected simultaneously at the data viii

PAGE 9

collection sites, in order to see how well the existing service measure (density) corresponded to the participants rankings. The data from the surveys were analyzed using an ordered probability model to determine which factors influenced the participants decisions and how. Three models were created. The first model used only density as a predictive factor. The second took into account only roadway and traffic characteristics, and the third examined all the significant factors that could be gathered from the survey. The density only model showed that density is indeed a strong indicator of travelers perceptions of trip quality. A set of LOS thresholds was also calculated using the survey participants responses. While the survey thresholds and the HCM thresholds had similar values for facility failure, the intermediate thresholds estimated from the survey participants responses were noticeably lower than the HCM thresholds. This suggests that travelers tolerance of congestion is lower on rural freeways than the HCM indicates. The other models showed the significance of other factors in the perception of trip quality in addition to density, such as socio-economic information and personal driving habits. This study provided some preliminary insight into travelers perception of trip quality, but further study is needed. It is suggested that more research be conducted regarding the effects of different factors on the perception of trip quality, such as a more diverse population sampling. Eventually, the results from this type of video-based study should also be compared to results obtained from a comparable in-field driving experiment. This study indicates the need for a further exploration into the differences between urban and rural freeways, and possibly a different set of thresholds for rural freeways. ix

PAGE 10

CHAPTER 1 INTRODUCTION Background Transportation engineers are responsible for targeting roadway infrastructure improvements where they will have the most beneficial effect. Since the capital available for these improvements is limited, engineers must carefully select the projects they choose to fund so that investments will have the best cost-benefit ratio. In a large part these decisions are guided by the procedures and methodologies found in the Highway Capacity Manual (HCM) [1]. The HCM is considered to be the definitive reference guide for traffic operations and analysis in the United States. The procedures in the HCM are used to estimate the operational performance of a variety of transportation facilities (e.g., signalized intersections, two-lane highways) and the corresponding level of service (LOS). The assignment of a LOS is based on designated performance measures and corresponding threshold values for individual facilities. The HCM is published by the Transportation Research Board (TRB) and its development and maintenance is the responsibility of the Highway Capacity and Quality of Service (HCQS) committee of the TRB. The current edition of the HCM was published in 2000. The concept of LOS is a foundation of the HCM. The LOS of a facility is used in the HCM as a qualitative indicator of the operating conditions being experienced by travelers of that facility, under specific roadway, traffic, and control conditions. The HCM describes LOS as A qualitative measure describing operational conditions within 1

PAGE 11

2 a traffic stream, based on service measures such as speed and travel time, freedom to maneuver, traffic interruptions, comfort, and convenience. LOS is divided into six categories, A through F in the 2000 HCM. LOS A indicates excellent service and LOS F indicates extremely poor service. An analysis yielding LOS A would indicate that the facility is performing extremely well, with low volumes and little congestion. If an analysis shows a facility to be performing at LOS C, it is in the middle range of congestion. If a facility is at LOS E, it is still permitting traffic flow but is experiencing significant delays with conditions approaching capacity. At LOS F, a facility is experiencing oversaturated conditions and the demand has exceeded the capacity of the facility. Problem Statement The performance measures that are used to calculate LOS for a facility are referred to as service measures. The currently designated service measure(s) for each facility is (are) based on the collective experience and judgment of the members of the HCQS committee. The same is true with the selection of the threshold values for the various LOS designations. There is currently no quantitative procedure to define which values are used as LOS thresholds. The LOS determination process, therefore, is based on the perspective of transportation professionals. The selection of service measures by the HCQS committee is, however, guided by two principles: 1) the service measure for each facility should represent speed and travel time, freedom to maneuver, traffic interruptions, and comfort and convenience in a manner most appropriate to characterizing quality of service for the particular facility being analyzed, and 2) the service measure chosen for a facility should be sensitive to traffic flow such that the

PAGE 12

3 service measure accurately describes the degree of congestion experienced by users of the facility [2]. The 1985 HCM described LOS as A qualitative measure that characterizes operational conditions within a traffic stream and their perception by motorists and passengers. The descriptions of individual levels of service characterize these conditions in terms of factors such as speed and travel time, freedom to maneuver, traffic interruptions, and comfort and convenience [3]. This statement indicates that the selection of performance measures and thresholds for the determination of level of service should be consistent with how operating conditions are perceived by the traveling public. Until recently, road users perceptions of quality of service were rarely compared to the LOS assigned to a facility by the HCM, despite the above definition emphasizing the importance of reflecting road users perceived quality of service. There have been suggestions from within the HCQS committee that a new approach needs to be explored when selecting a service measure for a facility. Instead of the measure and corresponding thresholds that transportation professionals (the HCQS committee) believe represent the quality of service as perceived by travelers, the publics opinion should be taken into account so as to determine what measure or measures they associate with quality of service on a transportation facility. Under the current methodology, the HCQS committee believed that the service measures were highly correlated with public perception, but this was not known for sure [4]. Since billions of dollars of transportation investment decisions are made every year based upon the outcome of HCM level of service analyses, it is essential that the transportation

PAGE 13

4 engineers assessments of the impact of these investments be consistent with traveler perception of the investment impacts. Research Objective and Tasks The objective of this study was to develop a model for assessing the LOS of a roadway facility that takes into account the road users perceived quality of service. Specifically, this study was focused on rural freeways. The following tasks were carried out in supporting the above research objective. Determine appropriate rural freeway sites to perform field data collection Collect video of roadway and traffic conditions from these sites Collect traffic data from count stations at these sites Produce video clips to be shown to survey participants Develop a survey instrument Recruit survey participants Conduct survey sessions Perform an analysis of survey responses Develop a level of service model Chapter Organization Chapter 2 contains an overview of the current HCM freeway analysis methodology as well as an overview of relevant literature. Chapter 3 describes the research approach for this study, including the field data collection, survey instrument development, survey response data collection, and the statistical analysis method used to analyze the data. Chapter 4 contains the analysis results. Chapter 5 contains the

PAGE 14

5 conclusions and recommendations. Additionally, several appendices with supporting data and information are included.

PAGE 15

CHAPTER 2 LITERATURE REVIEW The Highway Capacity Manual [1] states that the level of service of a roadway section should accurately reflect the perceptions of travelers, yet the current methodology does not directly take these perceptions into account. There have been some recent studies performed seeking travelers opinions about what factors and qualities are important to them in assessing the quality of their trip. A literature review was conducted to identify these studies and note their findings with regard to the travelers perception of LOS. HCM Freeway LOS Methodology A freeway is a section of divided roadway with controlled access and two or more lanes in one direction. Within this definition there are significant differences between urban and rural freeways. Rural freeways have greater distances between interchanges than urban freeways, higher speed limits than urban freeways, and a higher percentage of social and recreational trips than urban freeways. Urban freeways have a higher percentage of work and shopping trips than rural freeways. Despite these differences, urban and rural freeways both use density as their service measure with the same thresholds for LOS. Traveler expectations and perceptions of quality of service are different for rural and urban freeways. While urban freeways experience the full range of LOS conditions from A to F, rural freeways rarely drop below LOS C. Rural freeway travelers have come 6

PAGE 16

7 to expect these higher levels of service, therefore while urban freeway travelers are concerned with their overall travel time and the reliability of this travel time, rural freeway drivers take travel time for granted. Urban freeway drivers expect their ability to change lanes to be restricted, while a restricted ability to change lanes negatively impacts a rural freeway users perceived quality of service [5]. The original HCM had a basic three-point scale to define level of capacity. In 1963 the Level of Service concept was introduced and replaced the previous scale. In 1965 the six-point LOS scale (from A to F) was introduced. In 1985 this six-point scale was redefined to use traffic density (vehicles per unit length of roadway) as the service measure for defining LOS on freeway sections. This is the method that is still used today. Although the concept of LOS is meant to reflect the operational conditions as perceived by motorists, no freeway LOS methodology in the history of the HCM has been based on driver perception studies. Therefore, there can be no way to make sure that the LOS thresholds freeways (as well as any other type of facility) accurately reflect users perception of the quality of service they receive. Under the existing LOS methodology, rural and urban freeways have the same service measure density, as well as the same thresholds for each rank on the LOS scale. These thresholds for all freeway sections are shown in the table below. Do these thresholds accurately reflect the quality of service perceived by travelers on all freeways, urban and rural? In particular, the studies by Hostovsky [5] and Washburn [4] indicate that rural freeway travelers may judge the quality of their trip based on different qualities and criteria. A potential outcome of this study is a set of LOS thresholds unique to rural freeways. This idea of differing service measures for different

PAGE 17

8 categories of a specific facility type is not a new one. Currently there are two service Table 1. HCM Level of Service Thresholds Level of Service Density (pc/mi/ln) A 0-11 B 11-18 C 18-26 D 26-35 E 35-45 F > 45 measures used for assessing the LOS of a two-lane highway. These classes share a common service measure, but the thresholds are different (one of the classes also uses an additional service measure). In addition, the HCM procedure for analyzing arterial streets uses the same service measure for all arterial streets but includes four sets of thresholds for four different classifications of arterials [1]. Studies Investigating Traveler Perception of LOS A study by Pcheux et al. [6] noted that the Transportation Research Boards Committee on Highway Capacity and Quality of Service recognized a need to improve the HCM methodology of assessing LOS. Specifically, concerns were raised that the LOS of a roadway section did not correspond to road users perceptions. The authors felt that for LOS to accurately reflect travelers perception of quality of service they would first have to find out what performance measures were significant to travelers. The study method involved test participants driving along a pre-selected 40-minute route, encompassing mostly arterial streets, accompanied by an interviewer and a traffic engineer. The participant would discuss what factors they personally found

PAGE 18

9 important to the quality of their trip. Participants identified over 40 factors that were important. These included such factors as intersection efficiencyif the intersection was being utilized by opposing traffic while travelers were waiting, and the aesthetic qualities of the intersection. Both of these topics are not covered by the HCM. The study concluded that more research was needed to focus on traveler perception. A study by Hostovsky et al. [5] used focus group participants to identify factors important to trip quality on rural freeways and then compared those findings with those from a focus group study using regular urban commuters and commercial truck drivers. The participants in the rural freeway focus group identified three factors that were most important to trip qualitylow density, regular (predictable) travel time, and maintaining a steady travel speed. Other topics discussed were the safety issues inherent to the isolated locations of rural freeways, aesthetics, speed differential between cars, the presence of heavy vehicles, and the need for better traveler information. When compared to the results of a focus group study involving urban commuters, it was found that urban commuters placed high importance on the overall speed of their trip, where rural freeway travelers felt that the ability to choose their speed was a positive. This reflects the fact that urban drivers rarely have the opportunity to choose their speed in the traffic stream, so a faster speed is usually preferable over a slower one. Urban commuters were also not as concerned with the ability to change lanes and move about the facility at will. Most of the urban drivers were happy if they could stay in one lane and maintain a desired speed for their trip. The rural drivers were pleased if the density of the freeway section was low enough to allow movement between lanes and passing at will. This study was significant due to the fact that it recognized the

PAGE 19

10 differences in how travelers rate their trip quality on an urban versus a rural freeway. The HCM uses the same methodology for freeways in both types of areas. A study by Nakamura et al. [7] evaluated traffic flow conditions along an expressway in Japan from a drivers viewpoint. The study intended to quantitatively analyze the relationship between traffic flow conditions, drivers perceptions, and drivers behaviors. The field data portion of this study was intended to collect data on drivers behavior and perception under various flow conditions. Drivers had a video camera mounted in their own vehicle and were asked to drive a section along an expressway. After each trip the subject was asked to complete a survey about the traffic flow conditions. Twenty-two subject vehicles were used and 105 surveys were collected. The behavioral data collected was number of lane changes, travel time by lane, and percent time spent following. This study found that the most important factor influencing drivers satisfaction with their trip was the traffic flow rate. Other factors affecting trip quality were found to be number of lane changes, and the percent time spent following. Additionally, choosing the LOS based on the drivers level of satisfaction was attempted and then compared to the conventional LOS methodology. The results of this comparison suggested that the traffic conditions on Japanese expressways are not satisfying drivers. The realistic meaning of this result was that if facilities were designed to the drivers satisfaction level rather than the conventional LOS it would require an enormous investment. Several studies have been identified using road-user surveys and video selections to evaluate LOS methodology. The first study, by Sutaria and Haynes [8], used a road user survey to evaluate the LOS methodology for signalized intersections. Over 300

PAGE 20

11 drivers were shown video clips taken both from a drivers perspective and from an overhead camera at an intersection. The film segments were specifically chosen to represent a specific LOS and were intended to be shown to drivers for one or two signal cycles. The final compilation shown to drivers included both types of view and the clips were put in a random order. Their road user survey consisted of two partsa group attitude survey and a film survey. The group attitude survey used a questionnaire, to be answered before the film portion of the survey. The questionnaire included demographic information such as gender, age, and education, as well as questions about the participants driving experience and the type of roadways the participants usually drove on. They were then asked to give the relative importance of factors including delay, number of stops, traffic congestion, heavy vehicle density, and ability to change lanes as these factors applied to the quality of service at an intersection. After the initial questionnaire the participants were shown the video clips, consisting of a drivers view of a vehicle approaching, waiting, and passing through an intersection. After each of these clips the participants rated the quality of service they felt the intersection provided. At the end of the video portion the participants were again asked to rate the factors important to quality of service at an intersection to see if their initial opinion had changed. In all, 310 drivers participated in this survey. The results from the survey showed delay to be the most important factor both before and after the film portion of the survey. This study provided the first results that took into account the perceptions of travelers and changed the performance measure used by the HCM to evaluate LOS in the 1985 edition. This study also recommended further

PAGE 21

12 similar studies, and for further studies to simultaneously collect video and traffic flow data to allow for accurate measurements of what is depicted on the films. A study conducted by Pcheux et al. [9] addressed the issue of developing a study method to assess the perceived LOS at signalized intersections. The first objective of all the study methods was to determine how well the current LOS methodology reflects the opinion of road users. The second objective was to determine the factors affecting users perceptions at signalized intersections. The participants in this study represented a wide range of ages, education levels, and incomes. The participants were first shown a series of approaches to signalized intersections from a drivers perspective. After being shown a sequence of these clips, the participants were asked to fill out a survey including their attitudes about certain driving situations as well as their socio-economic information. After filling out these surveys the participants were asked to discuss the factors that influenced their perception of quality of service as a group. The study results showed that on average, the participants delay estimates were fairly accurate, however individual delay perceptions varied significantly. Fifteen factors were identified that contributed significantly to quality of service. Finally, the study found that participants tended to perceive service quality on three or four distinct levels as opposed to the six HCM levels of service. Another study using video clips and road user surveys was performed by Choocharukul et al. [10] with the intention of evaluating the current HCM methodology of assessing LOS. This study intended to provide a multivariate statistical analysis of the factors that were important to road users perception of trip quality and to compare those factors to the current performance measures for LOS.

PAGE 22

13 The data for this study were collected at various urban freeways. Cameras mounted on overpasses were used and were focused on sections that included inductance loop detectors. The cameras were focused so that only one direction of travel could be observed. The data from the loop detectors were collected and synchronized with the time of the video clips so the researchers would know the actual traffic flow conditions during the time the video clips were recorded. Two sets of video clips were chosen, each containing twelve clips. Two video clips were used for each HCM LOS designation, AF, with one clip on the high end of an LOS designation and the other clip at the low end. These designations were determined by the loop detector information. There were two groups of survey participants in this survey, one consisting of students, transportation professionals, and environmental management professionals, and the other consisting of commercial truck drivers and clerical and support staff. The participants were provided with written descriptions of the six HCM LOS designations (directly from the HCM). They were then asked to view the twelve video clips and rank each of them with the LOS they thought was appropriate for the conditions. The participants were also surveyed for demographic information such as age and education levels, as well as information about their driving habits. This study used an ordered probit statistical model to assess how users perceive the LOS of the roadway sections. The results of the survey and analysis revealed that perceived levels of service do not closely follow the HCM. Almost all the participants in this study had a lower tolerance for LOS A than the HCM, with the average cut-off for LOS A among the study participants shown to be 7 passenger cars per mile per lane (pc/mi/ln) as opposed to the HCM cutoff of 11 pc/mi/ln. The HCM threshold for LOS F

PAGE 23

14 also does not correspond with the findings of this study, with the participants selecting an average of 82 pc/mi/ln as the upper bound of LOS F as opposed to the HCM LOS F of 45 pc/mi/ln. The study also found that factors other than density are likely to influence road users perception of quality of service. The results from both groups indicated that a freeway with 4 lanes (instead of 6 or 8), an increase in traffic density, and an increase in the standard deviation of vehicle speeds all contributed strongly to a worse perception of LOS. It should be noted that the use of an overhead view of traffic could likely affect survey participants perceptions in a different way than that of an-vehicle view of traffic and roadway conditions. Background Study A study was performed at the University of Florida by Washburn et al. [4] with the objective of discovering what factors are important to drivers when evaluating the quality of their trip on a rural freeway. Several methods were considered for this study (focus groups, video/simulation viewing and review, interviews, etc.) with the final choice being an in-field survey-based approach. Two hundred and thirty-three travelers were surveyed at rest stops and service plazas along rural freeways in Florida. These locations were chosen due to their access to travelers in the process of a rural freeway trip. It was believed that this in-field survey approach would provide more reliable data, than mail-back surveys for example, as the drivers experiences would still be fresh in their minds when filling out the surveys. Drivers were asked to rank the factors that contributed to the quality of their trip on a scale from 1 to 7. The most important factor, ranked in the top three 64.3% of the time, was the ability to consistently maintain the desired travel speed. The factor with the

PAGE 24

15 next highest ranking was the ability to change lanes freely and pass other vehicles. This was ranked in the top three 33.3% of the time. The third most important factor was the ability to maintain a speed no less than the posted speed limit. This factor was ranked in the top three 33.0% of the time. This preliminary study showed that though density is important to rural freeway travelers, it is not the most important factor in determining trip quality. It also showed that drivers consider many other factors when determining trip quality. Conclusions The studies detailed in this chapter have shown that, while some research has been done on travelers perception of quality of service, there is a need for more study. The current HCM methodologies for evaluating LOS may be insufficient for determining the perceived quality of service from the travelers point of view. From the studies summarized in this chapter, we can see that it is possible to understand and approximate a travelers perception of quality of service using the factors that are found to be important to them. This type of research may ultimately assist decision makers when planning for new roadways and roadway improvements.

PAGE 25

CHAPTER 3 RESEARCH APPROACH This chapter will describe the methods used in collecting the sample data for this study as well as the methods used to refine the data for use in public surveys .Detailed within the chapter are the choice of a survey method, the selection of data collection sites, the creation of the survey form, and the process for conducting a road user survey. Alternative Survey Methods Common methods of data collection include the following: focus groups, field surveys, in-vehicle surveys driven by a researcher, in-vehicle surveys driven by the research participant, driving simulators, and video surveys. Focus Groups This consists of recruiting test participants in order to arrange a roundtable-type discussion about rural freeway travel. Participants would discuss their rural freeway trip experiences and relate which aspects of rural freeway travel are most important to them when evaluating the quality of their trip. The advantage of a focus group is the relative ease of the survey, there is no video data collection, field work, or liability on the part of the researchers. The disadvantage is that participants may influence each others responses and one particularly vocal participant could swing the rest of the group towards his or her opinion. Another disadvantage is the lack of a control element for the researchers there is no one experience on which the participants are basing their opinions, so the researchers can not look at the data or video record to interpret the responses. 16

PAGE 26

17 Additionally, the potential lack of quantitative feedback upon which to build an analytical model limits researchers in their ability to predict the responses of other travelers faced with similar roadway and traffic conditions. Field Surveys Researchers distribute survey forms at locations frequented by rural freeway travelers, such as rest stops or service plazas. The participants give their opinions on rural freeway travel in a survey form, rating which factors are most important when they judge their trip quality. One advantage to this method is that participants surveyed have recently driven on a rural freeway and have this experience fresh in their mind. Another advantage is that it is relatively easy to recruit participants for this sort of survey; there is always a ready supply of people in this type of location. The disadvantages are similar to the focus group. In-Vehicle Surveys (driven by research personnel) Participants are recruited and driven along a section of rural freeway, then surveyed about their perception of the trip quality. Advantages to this method include all participants would have the same experience to draw upon for their responses, and there would be no need to attempt to simulate the driving experience as participants would be experiencing the conditions firsthand. The disadvantages to this method include the liability to the researchers should the vehicle be involved in an accident, and the time and effort involved in conducting a survey of this manner. The controllability and repeatability of the conditions are also disadvantages because it is not possible to ensure the same conditions will be experienced by multiple survey participants.

PAGE 27

18 In-Vehicle Surveys (driven by research participants) Participants are recruited to drive along a section of rural freeway and provide the researchers with feedback on their trip once they return. Once again this method is advantageous in that it would provide participants with a firsthand look at the conditions involved. The disadvantages to this are similar to the previous method in that there is significant liability attached to a method like this, and this method would be even more time-consuming than the previous one. This method also suffers from the same lack of control and repeatability as any in-car survey. Driving Simulator Participants are put behind the wheel of a real vehicle, but the driving environment is simulated with the use of computer animation and video display monitors. They would then participate in the virtual driving of a rural freeway segment. This would give participants a closer likeness of actual freeway travel without the liability of having them drive a real section themselves. Disadvantages include cost (simulator time is expensive) and the well-documented motion sickness problem for participants (which increases recruitment time and costs). Video Surveys This method involves participants viewing pre-recorded video scenes from actual rural freeway sites. The clips could be from one of two perspectives: o Overhead View A camera placed over the test section of rural freeway records the traffic flow for survey participants to review at a later time. While this method does not give a simulation of actually driving the

PAGE 28

19 freeway section, it does give the participant a broader overview of the traffic stream. o Drivers Perspective A vehicle is equipped with a video camera to record the rural freeway trip from the drivers perspective. This method would better simulate actual rural freeway travel than an overhead view. After considering all advantages and disadvantages, the method chosen was a video survey from the drivers perspective. This method would allow larger groups of people to be surveyed simultaneously while giving a reasonably accurate depiction of rural freeway travel This method allows for complete control and repeatability of the conditions experienced by the participants, as well as eliminating the liability issues inherent in an in-vehicle survey. Video Data Collection The data collection method was developed after selecting the form the survey would take. It included five specific tasks Site Selection, Equipment and Setup, Video Data Collection, Video Clip Creation, and Loop Data Collection. Site Selection The sites at which the video clips were captured were all within Florida. Reasons for this include the proximity to the University of Florida and the access to the Florida Department of Transportations (FDOT) network of more than 7,500 traffic monitoring stations. The FDOT maintains a network of inductance loop detectors (ILD) along Floridas Intrastate Highway System. There are two types of ILD stations permanent

PAGE 29

20 and portable. The permanent stations were used in this study since they are continuously recording data 24 hours a day, 365 days a year. The portable stations require a data recorder to be hooked up at the location of the ILD station in a roadside cabinet. The permanent stations are telemetered such that data can be downloaded and archived on a daily basis. The data archived from the permanent stations are compiled every year by the FDOT and published on the Florida Traffic Information (FTI) CD. Also included on this CD for each state highway are the number of lanes in each direction, the ILD station type (permanent or portable), a description of the site, the Average Annual Daily Traffic (AADT), the percentage of trucks on the highway, and the peak hour in each direction, Multiple sites were examined around the state. There were several factors leading to the final site selections. South Florida was excluded due to the limited number of rural freeway segments and the long driving times required to get there and back. Sites west of Tallahassee were also not considered due to driving distance. These conditions hinged upon the availability of suitable sites in north-central Florida. The final sites selected represented a mix of four-lane and six-lane freeways, level and rolling terrain, truck percentages, and a wide range of volume. This information was obtained from the Florida Traffic Information CD published by the FDOT [11]. All sites selected were permanent count stations instead of portable stations. Additional data were used from a similar study conducted months before at the University of Florida. A list of the data collection sites and their associated traffic data follows in Table 2. Maps of the locations of the data collection sites can be found in Appendix A.

PAGE 30

Table 2. Data Collection Sites and Traffic Data Volume Site Site Type Description Direction 1 Direction 2 Two-way AADT K Factor D Factor Truck % 189920 Telemetered SR-93/I-75, 3.5 mi south of Turnpike, Sumter Co. 20472 N 21250 S 41722 10.1 59.84 21.66 360317 Telemetered I-75, 0.35 mi north of Williams Rd overpass, Marion Co. 37630 N 37844 S 75474 11.14 55.41 21.76 140190 Telemetered SR-93/I-75, 0.6 mi. south of SR 54, Pasco County 37443 N 37203 S 74646 8.76 53.67 11.71 730292 Telemetered SR-9/I-95, 1.4 mi south of Palm Coast Pkwy, Flagler Co. 29276 N 29980 S 59256 9.91 54.92 17.82 269904 Telemetered SR-93/I-75, 3 mi. north of Marion Co. line, Alachua Co. 31304 N 31023 S 62327 11.96 55.97 19.09 970428 Telemetered SR-91/Fl. Turnpike, 797 ft. south of CR561, Lake Co. 17655 N 18088 S 35743 11.09 55.42 12.34 21

PAGE 31

22 Equipment and Setup The objective of video data collection was to depict travel along a section of rural freeway from a drivers perspective. In order to give the survey participants a more complete representation of the conditions on the freeway section, it was decided that two more video images would be captured the vehicles speedometer and the driver side rear-view mirror. This brought the total number of cameras needed to three. The images would be combined during the creation of the video clips. The vehicle used for the video data collection was a minivan. As mentioned above, three cameras were placed in the vehicle in order to capture different aspects of the rural freeway trip. The first camera captured the view through the front windshield, including a view of the interior rear-view mirror. It was placed on a mount secured to the right side of the drivers seat. The second camera captured a view of the speedometer. It was mounted on a suction mount affixed to the steering column. It was found during a preliminary test that the instrument cluster needed to be shaded to reduce glare, so the image would not appear washed-out. The final camera captured a view from the vehicles driver-side rear-view mirror. This camera was mounted on a pole secured between the drivers seat and the door. Figure 1 shows these cameras as they were mounted in the vehicle. The video images were captured by three portable VCRs placed inside the vehicle. A microphone was also connected to one of the VCRs allowing the researcher to announce when they crossed a detector loop and any other potentially important information. This would allow the researcher to match the captured video clip to the loop data collected in a later step. All these devices were powered by three 12-volt deep-cycle batteries. A schematic of the equipment and connections is found in Figure 2.

PAGE 32

23 Figure 1. Camera Setup-Front View, Side View, Speedometer Video Data Collection The method used to capture the video clips remained constant throughout the collection of data. The researcher would activate the three VCRs and start recording. Then the researcher merged onto the freeway. The cameras captured conditions between the exit ramps that came before and after the ILD station. The researcher would speak into the microphone when the detector loop was crossed, giving the exact time and site number so the clip could be matched with the relevant loop data. Up to four runs were made at each location giving a number of clips to choose from when creating the clips for the survey. The data collection for this project was performed during November of 2003 and March 6-9, 2004. A summary of each video data collection session is shown in Table 3.

PAGE 33

24 Figure 2. In-Vehicle Equipment Setup Table 3. Data Collection Times, Locations, and Directions Date Site Freeway Direction Time 11/4/2003 730292 I-95 NB 12:55 11/5/2003 269904 I-75 SB 11:01 11/21/2003 140190 I-75 NB 6:48 11/21/2003 140190 I-75 SB 7:04 11/21/2003 140190 I-75 SB 7:14 11/21/2003 140190 I-75 SB 2:31 3/7/2004 730292 I-95 SB 2:35 3/7/2004 730292 I-95 NB 2:47 3/8/2004 189920 I-75 SB 12:36 3/8/2004 360317 I-75 SB 11:50 3/8/2004 360317 I-75 NB 12:04 3/8/2004 360317 I-75 SB 2:07 3/8/2004 970428 Turnpike SB 1:49 After reviewing the video data gathered on the first day of data collection, it was deemed unusable. The mounting for the camera allowed too much vibration in the picture and the video would not work for a public survey. Although the second round of video

PAGE 34

25 data collection was scheduled for March 6-8, the runs made on the first day needed to be redone due to problems with the camera placement, necessitating a fourth day of data collection on March 9, 2004. Video Clip Creation The survey participants were shown a single video display that contained the video scenes of the front windshield and interior rear-view mirror, the drivers side rear-view mirror, and the speedometer. The display used was a video projector and a wall-mounted screen, located between 5 and 20 feet away from the participants depending on the specific survey location. The setup of one of the survey sessions is depicted below in Figure 3. The majority of the screen was taken up by the view through the front windshield. Since the front windshield view captured a portion of the dashboard as well as the view from the front of the vehicle, the other two images could be overlaid on this area. A screenshot from one of the video clips used in the survey is shown in Figure 4. Screenshots from all 13 clips can be found in Appendix B. The clips were assembled using a video-editing program (Adobe Premiere) [12]. They first had to be captured from the VHS tapes using an ADS digital encoder [13]. After they were stored on the computer hard drive they were combined using Adobe Premiere into clips from 1.5 to 2.5 minutes in length. The length of an individual clip was chosen based on events in the video that the researchers wanted to include or exclude, as well as with a survey participants attention span in mind. The clips shown to viewers were chosen based on conditions they represented that were unique or different from other clips. This selection process is explained further in the section entitled Video Clip Selection.

PAGE 35

26 Figure 3. Setup of a Survey Session Figure 4. Sample Video Screenshot Inductance Loop Detector Data Collection It was desired to calculate the LOS of these sites according the HCM methodology in order to assess how strongly correlated it was with the responses

PAGE 36

27 provided by the survey participants. In order to determine the HCM LOS of the rural freeway segments, data were collected from the inductance loop detector (ILD) stations at each test site. The data collected came in three filesspeed, volume, and vehicle classification. FDOT personnel programmed the detectors at the sites selected for the study to record data in five-minute intervals (the hardware minimum interval) rather than the usual one-hour interval. This shorter interval allowed for traffic data that more accurately reflected the conditions depicted in the video clips. It should be noted that even with a five-minute data collection interval the conditions shown in the video clips could potentially vary from the average provided by the ILD data. These ILD data were used to categorize the collected video data and provided a starting point for selecting a range of conditions to be represented in the survey. The ILD data were provided in the form seen in Appendix D. When available, there were three data files for each site speed, count, and class. In the speed file, counts are provided for each speed range. The midpoints of the speed ranges are shown at the top of the table. In the class file, descriptions such as CL01 are given to the columns. These refer to the specific class of vehicle counted in that group and are explained by the figure provided. From the data provided it was possible to calculate descriptive statistics for the traffic flow at each site, such as the percentage of heavy vehicles in the traffic stream, the total 5-minute volume, the average speed, and the density. Video Clip Selection There were thirteen video clips chosen for the final survey. The final number of clips chosen was a result of five pilot test sessions, striking a balance between coverage of alternatives and attention/focus span of participants. These preliminary tests had

PAGE 37

28 shown that many participants lost interest after two minutes and had already started writing their opinions down. The final video clips were chosen to represent a variety of conditions in categories including lane configuration, traffic density, terrain, truck percentage, the presence of a median or guardrail, and shoulder configuration. The relevant data for each video clip included in the final survey is included below in Table 4 and Table 5.

PAGE 38

Table 4. Traffic Data for 13 Video Clips 29 Clip # Road Dir Lanes Clip Length Volume 1 Truck % 1 Density ILD Truck % Inside Speed Middle Speed Outside Speed LD Avg Speed Inner Lane Middle Lane Outer Lane ILD 5min Volume Terrain Speed 1 I-75 S 2 2:10 low none 8.00 0.13 77.1 72.2 74.30 42 57 99 flat 75 2 I-75 S 3 1:52 med-high high 63 75 66 204 flat 70-75 3 I-75 N 2 2:00 med-high med 13.79 0.20 76.4 69.4 72.20 67 99 166 flat 60-70 4 I-95 N 2 1:35 very high 26.45 56.5 55.4 56.00 102 145 247 flat 40-55 5 I-75 S 3 1:40 low-med med 6.30 0.17 77.6 74.0 66.4 72.30 26 51 37 114 rolling 70-75 6 I-95 S 2 1:59 med 10.21 76.9 74.7 75.80 63 66 129 flat 70 7 I-75 S 2 2:00 med-high high 26.31 0.34 71.6 65.5 68.90 167 135 302 flat 67-72 8 I-75 N 3 2:01 med-high low 61 98 80 239 flat 67-72 9 I-75 S 2 2:00 high high 26.11 0.32 71.3 66.1 69.20 179 122 301 flat 55-65 10 I-95 N 2 1:43 med 10.86 78.4 69.4 72.90 80 52 132 flat 75 11 I-75 S 3 1:26 med med 48 82 48 178 flat 70-75 12 I-75 S 2 1:27 med-high med 17.04 0.15 73.6 68.2 71.10 110 92 202 flat 60-65 13 Turnpike S 2 2:03 med high rolling 75-80 1 These levels (low, med, high) indicate subjective judgments that were used to choose between clips.

PAGE 39

30 Table 5. Clip Sites, Dates, and Times Clip # Clip Site Time Date Closest City 1 189920 run 1 189920 12:36 3/8/2004 Wildwood 2 360317 run 1 360317 11:50 3/8/2004 Ocala 3 Tampa 0648 140190 6:48 11/21/2003 Tampa 4 730292 run 4 730292 14:47 3/7/2004 Daytona Beach 5 Micanopy 1101 269904 11:01 11/5/2003 Micanopy 6 730292 run 3 730292 14:35 3/7/2004 Daytona Beach 7 Tampa 0714 140190 7:14 11/21/2003 Tampa 8 360317 run 2 360317 12:04 3/8/2004 Ocala 9 Tampa 0704 140190 7:04 11/21/2003 Tampa 10 Daytona 1255 730292 12:55 11/4/2003 Daytona Beach 11 360317 run 3 360317 12:07 3/8/2004 Ocala 12 Tampa 1431 140190 14:31 11/21/2003 Tampa 13 970428 run 1 970428 13:49 3/8/2004 Winter Garden Survey Sessions Development of Survey Form and Participant Instructions The survey form for this study had to serve two purposesrecord the participants opinions about the rural freeway video clips and their reasons for these opinions, and record characteristics about the participants that might influence their ratings. Thus the form is divided into two sections. The first section of the survey form is for personal information about the traveler taking the survey. Examples of this information include education level, income, and number of years possessing a drivers license. This section also records information about the participants rural freeway travel habits. It asks for information such as the amount of rural freeway trips taken per month and the average length of the participants rural freeway trips. Finally it asks for some driving habits, such as any changes in the

PAGE 40

31 participants driving style when driving alone versus with a passenger. It also asks the participant to rate their usual driving style, from Conservative to Aggressive. The second section of the survey is for recording the participants opinions and rankings of the video clips. It is divided into two sections for each of the thirteen clips. The first section asks the participant to rank the quality of the trip depicted in the video clip on a scale from Very Poor to Excellent with 6 total ranking levels. A total of six ranking levels was chosen so that there would be general correspondence with the six levels of the HCM (A-F). Participants were asked to use the word ranking rather than a numerical ranking (e.g., 1-6) to minimize the possibility that those familiar with the HCM might try to equate the numerical rankings with the HCM LOS rankings. The second section asks the participant to record why they ranked the video clip as they did, listing all factors that significantly contributed to their ranking. The participants were to then number these according to their relative significance to each other. Finally the form includes questions about the survey itself. These include the participants opinion on the video clips as a representation of rural freeway travel and if the participant would have changed their rankings based on the purpose of the trip (e.g., business, recreational, or social). A one page written survey instruction sheet was developed because there was a significant amount of information that needed to be communicated to the participants in order for them to complete the survey form in a manner which would be useful as study data. The participants could refer back to it if there were any questions about the survey process. The instructions given to each survey participant are provided in Appendix C.

PAGE 41

32 Conducting the Survey Sessions Survey participants were recruited from various sources. They include the following: Undergraduate students in the University of Florida civil engineering program, recruited from the introductory transportation engineering course, Graduate students in the University of Florida civil engineering program, recruited from the transportation degree program, Employees of the University of Florida Technology Transfer Center, Employees of the Florida Department of Transportation, and Alachua county residents (Random participants recruited for a fee by the Florida Survey Research Center) The undergraduate students were recruited from the Principles of Highway Engineering and Traffic Analysis course during the Fall 2004 semester. The graduate students were those enrolled in a transportation engineering degree program during the Fall 2004 semester. The University of Florida Technology Transfer Center is an organization that provides training and technical assistance to Floridas transportation and public works professionals. Their survey session was conducted at their off-site headquarters in Gainesville, FL, with participants ranging from high-school educated support staff to professionals with graduate degrees. The FDOT survey session was conducted at the central office in Tallahassee, FL. This session also included participants of varying backgrounds and demographics. The public sample was comprised of Alachua county residents, recruited by the University of Florida Survey Research Center. The survey center was instructed to recruit individuals with varying socio-demographic

PAGE 42

33 characteristics and also make sure that the participants had experience driving on rural freeways. Additionally, they did not recruit college students as there was already a sufficient number in this group. In total there were 126 surveys filled out for this study. The locations, dates, and groups of participants taking the survey during each session are given in Table 6. Table 6. Dates and Locations of Survey Sessions Survey Session Date City Location Participants # of Surveys 1 8/4/04 Gainesville UF Technology Transfer Center T 2 employees 16 2 11/16/04 Tallahassee Florida DOT Central Office DOT employees 11 3 12/2/04 Gainesville University of Florida undergraduate students 14 4 12/2/04 Gainesville University of Florida undergraduate students 9 5 12/4/04 Gainesville UF Hilton Conference Center public 1 13 6 12/4/04 Gainesville UF Hilton Conference Center public 1 15 7 12/4/04 Gainesville UF Hilton Conference Center public 1 11 8 12/9/04 Gainesville University of Florida undergraduate students 20 9 1/22/05 Gainesville University of Florida public 1 9 10 1/27/05 Gainesville University of Florida graduate students 8 Total Number of Surveys 126 1 Participants were recruited through the University of Florida Survey Research Center Because of the video format of the survey, multiple surveys could be filled out at a time, the main limitations being the ability of the participants to comfortably view the video screen and the length of time for which the participants could be expected to focus on this task. The screen was placed as close as possible to eye level so participants looking at the screen saw it as they would a cars windshield. Before viewing the clips the participants were given the instruction sheet and time to read it. These written instructions were also verbally reviewed by the session moderator, as well as some supplemental information. The participants were also told that they could ask

PAGE 43

34 interpretation questions in-between the viewing of the video clips. It was decided to create two example clips, each 20 seconds long, to show the upper and lower ends of the range of possible traffic flows. The first was a nearly empty four-lane freeway and the second was stop-and-go traffic along a four-lane freeway. The participants were then shown each of the 13 video clips and instructed to watch each clip entirely before writing their responses. Since it was not intended for the order of the clips to have any effect on the participants rankings, the order was shifted for each survey session. After each clip was finished, the participants were given time to record their rankings.

PAGE 44

CHAPTER 4 ANALYSIS AND RESULTS This chapter contains information about the methodology used to analyze the survey data, as well as the results of these analyses. Analysis Method To determine how or if the participants responses correspond to the six LOS rankings, a statistical analysis was needed to predict the probability of selecting discrete rankings (1-6 as included in the survey). While one of several multinomial discrete-choice modeling methods would suffice to predict a discrete outcome, most do not take into account the ordered nature of the responses in this survey (1 is better than 2, which is better than 3, etc.). Using a standard multinomial discrete model, such as a multinomial logit model, would still yield consistent parameter estimates, but with a loss of efficiency [14]. In order to account for the discrete and ordered responses in this survey, an ordered probability model was chosen as the statistical analysis approach. An ordered probability model is derived by defining an unobserved variable, z, that is the basis for modeling the ordinal ranking of data (in this case the six clip rankings) [15]. This variable is specified as a linear function for each observation n such that z n = X n + n (1) 35

PAGE 45

36 where X n is a vector of variables determining the discrete ordering for observation n, is a vector of estimable parameters, and n is a random disturbance. In this analysis, y is defined as each participants evaluation of each of the 13 video clips. Since there are 126 participants and 13 clips, there are a total of 1638 observations. Using this equation, the observed clip ranking, y n for each observation is written as y n = 1 if z n 1 y n = 2 if 1 < z n 2 y n = 3 if 2 < z n 3 (2) y n = 4 if 3 < z n 4 y n = 5 if 4 < z n 5 y n = 6 if z n 5 where the values are the thresholds that define y n The values are estimated jointly with the model parameters (). The estimation problem then becomes one of determining the probability that a participant will select a particular ranking for each clip. In using the ordered probit model, it is assumed that the error term, n is normally distributed with a mean of 0 and a variance of 1. The resulting ordered probit model has the following probabilities corresponding to each clip ranking: P(y n = 1) = (-X n ) P(y n = 2) = ( 1 X n ) (-X n ) P(y n = 3) = ( 2 X n ) ( 1 X n ) (3)

PAGE 46

37 P(y n = 4) = ( 3 X n ) ( 2 X n ) P(y n = 5) = ( 4 X n ) ( 3 X n ) P(y n = 6) = 1 ( 4 X n ) It can be shown that threshold 1 can be set equal to 0 without loss of generality [15]. In the above equations, (.) represents the cumulative normal distribution: uwdweu22121)( (4) This model can be estimated using maximum likelihood procedures. The thresholds 1 and 1 define the upper and lower thresholds for outcome i. This is illustrated in Figure 5 Figure 5. Illustration of an Ordered Probability Model A positive increase in the term implies that an increase in x will increase the probability that the highest category response will be returned (in this case, y = 6). An increase in the

PAGE 47

38 term also implies that the probability of returning the lowest response (y = 1) is decreased. This is illustrated in Figure 6. Figure 6. Illustration of an Ordered Probability Model with an Increase in A unique issue was present in this data set that complicated the analysis procedure. Each of the 126 participants viewed 13 clips and thus generated 13 observations. The issue is that there are unobserved characteristics that are unique to each participant that will be reflected in all 13 of their rankings. If this is not accounted for in the model, the model will be estimated as though each of the 1638 observations came from a unique participant. This approach would result in lower standard errors in the models estimated parameters, leading to inflated t-statistics and exaggerated degrees of significance. The solution to this problem is found in a standard random effects approach. The first equation is rewritten as

PAGE 48

39 z ic = X ic + ic + i (5) where i denotes each participant (i = 1,,126), the c denotes each video clip (c = 1,,13), i is the individual random effect term and all other terms are as previously defined. The random effect term i is assumed to be normally distributed with mean 0 and variance 2 When this random effects model is estimated, an estimate of is also calculated, the significance of which determines the significance of the random effects model relative to the standard ordered probit model [16]. Statistical Analysis The results of the surveys were put into spreadsheet form, with unique cases for each clip viewing. Each participants rankings were kept together within the spreadsheet for analysis purposes. The data were analyzed using LIMDEP [17] with a random effects approach as detailed in the previous section. The first analysis was performed to explore how the quality of service perceptions of the participants in this survey correlated with the HCM LOS thresholds. The density for each of the video clips was calculated from the loop detector data (and the video data, in cases where the loop detector data was incomplete). A statistical analysis was performed using density as the only independent variable to find out where the thresholds of the survey participants fell relative to the six clip rankings. The results are given below in Table 7. The very high level of significance indicated by the t-statistic (coefficient divided by standard error) calculated for density in the above model offers some evidence that

PAGE 49

40 this performance measure correlates well with perceived LOS. The reference t-statistic for these analyses is 1.282, representing a 90% confidence level in a one-tailed t-test. The positive coefficient calculated for density indicates that, as density increases, the likelihood of a traveler perceiving a worse LOS increases. The random effects term, is also highly significant, meaning that the choice of a random effects model for this data set was correct. Had this term not been significant, a normal ordered probability model would have been sufficient. One test for the goodness-of-fit of a model is calculating that models 2 value. The 2 value of a model is between 0 and 1. A 2 value of 1.0 indicates a perfect model fit. The 2 value of a model is calculated as follows: )0()(12LLKLL (6) where K represents the number of variables in the model, LL() represents the log likelihood at convergence, and LL(0) represents the initial log likelihood [15]. Table 7. Density Model Estimation Results Variable Coefficient Standard Error t-statistic Constant -0.138 0.076 -1.82 Traffic Characteristics Density (pc/mi/ln) 0.096 0.003 34.37 Threshold Values 1 0.918 0.038 23.89 2 1.922 0.048 39.92 3 2.863 0.053 53.88 4 4.112 0.066 62.47 Standard Deviation of Random Effects 0.455 0.050 9.12 Initial Log Likelihood -2710.16 Log Likelihood at Convergence -2314.60 2 0.15

PAGE 50

41 Using the participants responses it was possible to calculate a set of thresholds for the participants assigned LOS rankings. Using the calculated values in Table 7, the threshold values can be calculated as ( k 0 )/ 1 In this equation, k designates the five threshold values, 1 = 0, and the other threshold values are given in A comparison between the calculated threshold values from this survey and the HCM LOS thresholds is given in Table 8. Table 7 Table 8. Comparison of Estimated and HCM LOS Thresholds LOS Estimated Thresholds (pc/mi/ln) HCM thresholds (pc/mi/ln) A 0-2 0-11 B >2-11 >11-18 C >11-21 >18-26 D >21-31 >26-35 E >31-44 >35-45 F >44 >45 These thresholds are generally lower than the HCM thresholds for corresponding rankings, indicating the participants in this survey had a lower tolerance for high-density traffic conditions than could be inferred from the HCM LOS thresholds. The second analysis that was performed was intended to take into account all the traffic and roadway characteristics influencing the participants perception of trip quality. The results of this table are given below in Table 9. The traffic characteristics examined produced effects according to expectations. The calculated difference in speed between the inner lane and the outer lane was in the model as speed differential. As this value increased, participants were more likely to assign a worse LOS to a given set of conditions. A higher average speed resulted in a more favorable LOS ranking. Motorists in this survey found three lanes in one direction

PAGE 51

42 to be a preferred configuration over two lanes and were more likely to assign a favorable LOS ranking to those roadways with three lanes in one direction. Table 9. Traffic Characteristics Model Estimation Results Variable Coefficient Standard Error t-statistic Constant 6.296 0.597 10.55 Traffic Characteristics Speed Differential (mi/h) 0.163 0.027 6.08 Average Speed (mi/h) -0.096 0.009 -10.97 3 Lanes (1 Yes, 0 No) -1.848 0.210 -8.82 Truck % 0.005 0.004 1.04 Density (pc/mi/ln) 0.061 0.006 10.59 Threshold Values 1 0.949 0.064 14.88 2 2.192 0.077 28.48 3 3.258 0.092 35.60 4 4.630 0.106 43.80 Standard Deviation of Random Effects 0.522 0.060 8.76 Initial Log Likelihood -2710.16 Log Likelihood at Convergence -1472.53 2 0.45 An increase in the truck percentage resulted in a higher possibility of a worse LOS ranking. While the t-statistic for the truck percentage was below 1.282, it was decided to leave this variable in the model because it was felt that this was a very important variable from a policy standpoint. As expected, the participants preferred not to have a high percentage of trucks in the traffic stream. Finally, density was very significant in this model as it was in the first. A higher density led to an increased possibility of a worse LOS ranking. The random-effects term was again significant in this analysis, justifying the use of a random-effects model.

PAGE 52

43 The third analysis that was performed was aimed at discovering which factors are important to travelers when judging their trip quality. This model was estimated including demographic data as well as roadway and traffic flow characteristics. The values given in Table 10 should be interpreted such that a positive parameter estimate means that an increase in that variable will lead to a better perceived quality of service, and a negative parameter estimate means that an increase in that variable will lead to a worse perceived quality of service. Table 10. Level of Service Model Estimation Results Variable Coefficient Standard Error t-statistic Constant 6.156 0.622 9.90 Demographic and Background Information Age > 35 (1 Yes, 0 No) -0.358 0.121 -2.96 Income (thousands of $) -0.003 0.002 -1.89 Average Number of Rural Freeway Trips per Month 0.025 0.017 1.49 Average One-Way Trip Distance > 100 miles? (1 Yes, 0 No) 0.395 0.127 3.11 Less Aggressive Driver with Passengers? (1 Yes, 0 No) 0.267 0.186 1.43 Traffic Characteristics Speed Differential (mi/h) 0.162 0.028 5.85 Average Speed (mi/h) -0.095 0.009 -10.58 3 Lanes (1 Yes, 0 No) -1.836 0.217 -8.47 Truck % 0.005 0.005 1.03 Density (pc/mi/ln) 0.062 0.006 10.56 Threshold Values 1 0.939 0.065 14.56 2 2.181 0.078 27.93 3 3.247 0.093 34.90 4 4.613 0.107 43.21 Standard Deviation of Random Effects 0.435 0.059 7.42 Initial Log Likelihood -2710.16 Log Likelihood at Convergence -1447.34 2 0.46

PAGE 53

44 In Table 10, a positive coefficient value indicates that as the variable increases, there is an increased likelihood of a worse perception of LOS. Likewise, a negative coefficient value indicates that as the variable increases, there is an increased likelihood of a better perception of LOS. The results indicate that, while density is important to travelers, it is not the only factor influencing perceived quality of service. The survey results showed significant effects of demographic and background information on drivers LOS rankings. Table 10 indicates that participants with over 35 are more likely to assign a given set of conditions a better LOS, as are those with higher incomes. Travelers who drive on rural freeways more frequently are more likely to perceive a worse LOS, as are those whose average rural freeway trip is over 100 miles in one-way length. Those participants who indicated that they tend to drive less aggressively with passengers in the car as opposed to driving alone were more likely to assign a worse LOS to a given set of conditions. A possible explanation is that these drivers are more aggressive than the average motorist. Participants were asked if they considered themselves to be an aggressive driver, and the results of that model did not display significance. Perhaps motorists were more reluctant to admit they drive aggressively, but this tendency manifests itself in their responses to this question. The results estimated using the traffic and roadway characteristic variables showed similar significance and magnitude to the model estimated only using these variables. The random effects term was once again significant.

PAGE 54

CHAPTER 5 CONCLUSIONS AND RECOMMENDATIONS Since 1963, the Level of Service concept has been integral to the Highway Capacity Manual methodology for assessing the performance of transportation facilities. There is, however, still relatively little known about how the HCM methodologies for assigning LOS correspond to road users perceptions of their quality of service. The purpose of this study was to investigate what factors influenced road users perceptions of quality of service, and how that perception compares to HCM calculated LOS. Data Collection and Video Clip Creation The data collection process used for this study proved successful in gathering the necessary video data. After deciding on the best camera positions and mounting techniques, all cameras recorded clear, steady views of their intended targets. The equipment in the vehicle performed exactly as intended, capturing the necessary information while keeping all three VCR timers consistent so the video data could be synchronized at a later time. The sites chosen generally provided a good variety of traffic conditions, but some clips from a pilot study were also used to provide additional roadway and traffic conditions that were not captured in the data collection effort for this project. These clips were re-edited using the same process as the clips filmed for this study so there would be consistency in the screen views. 45

PAGE 55

46 The loop detector data did not work out as well for some of the sites as was initially hoped. Due to malfunctioning detectors or construction at the selected sites, some of the desired data were not available. The final form of the video clips and the presentation to survey participants worked very well, exactly as intended. The last question on the survey form (as seen in Appendix A) asked participants to rate how well the video clips simulated the driving experience for the conditions depicted on the screen. The majority of participants found the survey to be a very good representation of the actual driving experience, with 95% of the participants rating the survey as a good or better representation of the actual driving experience. The responses to this question are tabulated in As shown in this table, the average response from participants was approximately a 2 out of 6, corresponding to very good. Table 11Table 11. Realism of Video Survey Responses Ranking Excellent Very Good Good Fair Poor Very Poor 1 2 3 4 5 6 Frequency 21 64 36 5 1 0 Percent of Total Responses (%) 17 50 28 4 1 0 Average Rank 2.2 Statistical Analysis The analysis process chosen for this survey was an ordered probability model, specifically the ordered probit model. The structure of the standard ordered probit model formulation does not account for each participant providing 13 responses, so a random-effects formulation was used. This modeling choice was justified, with the standard deviation of random effects showing significance in all statistical analyses.

PAGE 56

47 The first model developed was one incorporating only density as an independent variable. This produced results that were as expected, that density is very significant to travelers when they are judging the quality of service provided by a rural freeway. A complimentary outcome of this analysis was that density thresholds for each LOS were estimated according to the survey participants responses. For LOS A-E, the survey participants showed a lower tolerance for high-density traffic conditions, hence their estimated thresholds were lower. The HCM thresholds and the estimated thresholds showed similar values for LOS F. The second model was estimated to include the influence of other roadway and traffic characteristics. The results of this model showed that while density is significant to user perception of LOS, there are other significant factors influencing this perception, such as average speed of the traffic stream and the speed differential between lanes. The final model included all factors from the survey that were found to be significant, including demographic factors as well as roadway and traffic characteristics. The results of this model indicated that the background and characteristics of the individual road user can influence their perception of LOS. While this result was expected, it is still significant due to the implications for a potential future modification to the HCM LOS methodology. Study Limitations and Recommendations for Further Research Since the scope of this study was limited to North Central Florida, additional testing with participants from a variety of other geographic regions would be needed to adopt any findings on a national level. An expanded sample, both geographically and in roadway conditions, would provide much more comprehensive coverage of the roadway

PAGE 57

48 and traffic condition combinations. The video survey format has inherent limitations as well. In a future study, it would be desirable to allow road users to drive in a traffic stream with known characteristics (density, truck percentage, etc.), then express their opinion regarding the LOS of the roadway section. This was not considered for this study due to cost and liability. The results of this survey could be compared to the results of the video survey to assess the accuracy of the video survey. If the video survey is shown to be an accurate method of simulating traffic conditions, it can be used in future studies and will be more effective than in-field surveys. Finally, although participants were told to imagine the conditions in the video scenes as if they were occurring throughout the duration of a trip, it is not known whether actually experiencing these conditions for an equivalent time to an entire trip would change the outcome. It is hoped that the findings of this study will lead to further developments in this area. The study does show that density is significant in determining a road users perception of trip quality. It is also known that there are significant factors influencing LOS other than density and these should be explored more completely. Ultimately, a better understanding of travelers perceptions of quality of service will lead to a better use of the available resources to improve the roadway network where it is really needed, and to more accurate planning and accommodating for future demands.

PAGE 58

REFERENCES 1. Transportation Research Board (2000). Highway Capacity Manual. TRB, National Research Council. Washington, D.C 2. Harwood, D., Flannery, A., McLeod, D., Vandehey, M. (July 2001). The Case for Retaining the Level of Service Concept in the Highway Capacity Manual. Presented at the 2001 Transportation Research Board Committee A3A10 Highway Capacity and Quality of Service Midyear Meeting, Truckee, California. 3. Transportation Research Board (1985). Special Report 209: Highway Capacity Manual. TRB, National Research Council. Washington, D.C 4. Washburn, S., Ramlackhan, K., McLeod, D. (2004). Quality of Service Perceptions by Rural Freeway Travelers: Exploratory Analysis. Transportation Research Record: Journal of the Transportation Research Board, No. 1883. Washington, D.C., pp. 132-139. 5. Hostovsky, C., Wakefield, S, Hall, F. (2004). Freeway users Perception of Quality of Service: A Comparison of Three Groups. In Transportation Research Record: Journal of the Transportation Research Board, No.1883. TRB, National Research Council. Washington, D.C., pp. 150-157. 6. Pcheux, K., Flannery, A., Wochinger, K., Rephlo, J., Lappin, J. (2004). Automobile Drivers Perceptions of Service Quality on Urban Streets. Transportation Research Record: Journal of the Transportation Research Board, No. 1883. TRB, National Research Council. Washington D.C. pp. 167-175. 7. Nakamura, H., Suzuki, K., Ryu, S. (2000). Analysis of the Interrelationship Among Traffic Flow Conditions, Driving Behavior, and Degree of Drivers Satisfaction on Rural Motorways. Transportation Research Circular E-C018: Proceedings of the Fourth International Symposium on Highway Capacity. National Research Council. Washington, D.C., pp. 42-52 8. Sutaria, T.C., and Haynes, J.J. (1977). Level of Service at Signalized Intersections. Transportation Research Record: Journal of the Transportation Research Board, No. 644. TRB, National Research Council. Washington, D.C., pp. 107-113. 49

PAGE 59

50 9. Pcheux, K., Pietrucha, M., Jovanis, P. (2000). User Perception of Level of Service at Signalized Intersections: Methodological Issues. Transportation Research Circular E-C018: Proceedings of the Fourth International Symposium on Highway Capacity, National Research Council. Washington, D.C., pp. 322-335. 10. Choocharukul, K., Sinha, K., Mannering, F. (2004). User Perceptions and Engineering Definitions of Highway Level of Service: an Exploratory Statistical Comparison. Transportation Research Part A, 38. pp. 677. 11. Florida Traffic Information 2003. (2003). Florida Department of Transportation, Tallahassee, FL, CD-ROM. 12. Users Guide for Adobe Premiere Pro Software. (n.d.). Last Accessed November 17, 2003, from http://www.adobe.com/products/premiere 13. Users Guide for ADS Pyro A/V Link. Last Accessed March 15, 2005, from http://www.adstech.com/products/API-555/intro/api555_intro.asp?pid=API-555 14. Amemiya, T. (1985). Advanced Econometrics. Harvard University Press. Cambridge, MA. 15 Washington, S., Karlaftis, M., Mannering, F., 2003. Statistical and Econometric Methods for Transportation Data Analysis. Chapman & Hall/CRC. Boca Raton, FL. 16 Greene, W., 2003. Econometric Analysis. Prentice Hall. Upper Saddle River, NJ. 17. Users Guide for LIMDEP 8.0. 2004. http://www.limdep.com Econometric Software, Inc. Last Accessed March 23, 2005.

PAGE 60

APPENDIX A LOCATIONS OF DATA COLLECTION SITES 51

PAGE 61

52

PAGE 62

53

PAGE 63

54

PAGE 64

APPENDIX B VIDEO CLIP SCREENSHOTS

PAGE 65

56

PAGE 66

57

PAGE 67

58

PAGE 68

59

PAGE 69

60

PAGE 70

61

PAGE 71

62

PAGE 72

APPENDIX C RURAL FREEWAY TRIP QUALITY SURVEY FORM

PAGE 73

64

PAGE 74

65

PAGE 75

66

PAGE 76

67

PAGE 77

APPENDIX D SAMPLE LOOP DETECTOR DATA

PAGE 78

69 0 Tag County Site Lane Year Month Day Hour Min Int 15 23 28 33 38 43 48 53 58 63 68 73 78 83 91 Total Vol. Avg. Spd 1 5 min vol1 veh/hr/ln1 Density 1 SPD 18 9920 1 04 03 08 00 05 005 0 0 0 0 0 0 0 0 2 1 3 4 3 1 1 15 72.2 SPD 18 9920 2 04 03 08 00 05 005 0 0 0 0 0 0 0 0 0 0 0 2 5 1 0 8 77.4 23 138 1.86 SPD 18 9920 3 04 03 08 00 05 005 0 0 0 1 0 0 0 0 0 0 0 2 4 1 1 9 73.9 SPD 18 9920 4 04 03 08 00 05 005 0 0 0 0 0 0 0 0 1 2 6 10 5 1 0 25 71.8 34 204 2.82 SPD 18 9920 1 04 03 08 00 10 005 0 0 0 0 0 0 0 0 1 2 3 3 7 2 0 18 73.3 SPD 18 9920 2 04 03 08 00 10 005 0 0 0 0 0 0 0 0 0 0 0 3 6 1 2 12 79.3 30 180 2.38 SPD 18 9920 3 04 03 08 00 10 005 0 0 0 0 0 0 0 0 0 0 0 5 7 2 1 15 77.9 SPD 18 9920 4 04 03 08 00 10 005 0 0 0 0 0 0 0 1 0 1 5 10 4 1 0 22 71.9 37 222 2.99 SPD 18 9920 1 04 03 08 00 15 005 0 0 0 0 0 0 0 0 0 2 2 7 3 0 2 16 74.3 SPD 18 9920 2 04 03 08 00 15 005 0 0 0 0 0 0 0 0 0 0 0 1 2 2 3 8 83.5 24 144 1.86 SPD 18 9920 3 04 03 08 00 15 005 0 0 0 0 0 0 0 0 0 0 1 7 5 1 0 14 75.1 SPD 18 9920 4 04 03 08 00 15 005 0 0 0 0 1 0 0 0 2 5 8 7 4 1 0 28 68.5 42 252 3.56 SPD 18 9920 1 04 03 08 00 20 005 0 0 0 0 0 1 0 0 0 5 3 6 4 0 1 20 70.2 SPD 18 9920 2 04 03 08 00 20 005 0 0 0 0 0 0 0 0 0 0 4 0 3 0 1 8 74.6 28 168 2.35 SPD 18 9920 3 04 03 08 00 20 005 0 0 0 0 0 0 0 0 0 2 3 3 6 1 2 17 75.4 SPD 18 9920 4 04 03 08 00 20 005 0 0 0 0 0 0 0 0 0 4 4 8 12 0 0 28 73.0 45 270 3.65 SPD 18 9920 1 04 03 08 00 25 005 0 0 0 0 0 0 0 0 0 0 5 5 3 2 1 16 74.8 SPD 18 9920 2 04 03 08 00 25 005 0 0 0 0 0 0 0 0 0 0 0 2 3 2 0 7 78.0 23 138 1.82 SPD 18 9920 3 04 03 08 00 25 005 0 0 0 0 0 0 0 0 0 0 0 0 2 2 0 4 80.5 SPD 18 9920 4 04 03 08 00 25 005 0 0 0 0 0 0 0 0 0 1 7 2 5 0 0 15 71.7 19 114 1.55 SPD 18 9920 1 04 03 08 00 30 005 0 0 0 0 0 0 2 0 0 1 2 9 4 0 0 18 70.2 SPD 18 9920 2 04 03 08 00 30 005 0 0 0 0 0 0 0 0 0 0 0 3 2 1 0 6 76.3 24 144 2.01 SPD 18 9920 3 04 03 08 00 30 005 0 0 0 0 0 0 0 0 0 0 1 2 1 1 2 7 79.6 SPD 18 9920 4 04 03 08 00 30 005 0 0 0 0 0 0 0 0 0 2 5 8 4 2 1 22 73.6 29 174 2.32 SPD 18 9920 1 04 03 08 00 35 005 0 0 0 0 0 0 0 0 3 3 3 6 3 5 0 23 71.9 SPD 18 9920 2 04 03 08 00 35 005 0 0 0 0 0 0 0 0 0 0 5 1 0 0 6 73.8 29 174 2.41 SPD 18 9920 3 04 03 08 00 35 005 0 0 0 0 0 0 0 0 0 0 0 3 2 4 0 9 78.6 SPD 18 9920 4 04 03 08 00 35 005 0 0 0 0 0 0 0 0 0 1 7 11 8 0 0 27 72.8 36 216 2.91 SPD 18 9920 1 04 03 08 00 40 005 0 0 0 0 0 0 0 0 0 2 6 7 2 1 0 18 71.3 SPD 18 9920 2 04 03 08 00 40 005 0 0 0 0 0 0 0 0 0 0 0 1 2 1 0 4 78.0 22 132 1.82 SPD 18 9920 3 04 03 08 00 40 005 0 0 0 0 0 0 0 0 0 0 0 1 4 1 2 8 81.3 SPD 18 9920 4 04 03 08 00 40 005 0 0 0 0 0 0 0 0 2 2 3 7 6 1 1 22 72.7 30 180 2.40 1 These categories were calculated from the given loop detector data and added to the speed data spreadsheets.

PAGE 79

70 Tag County Site Yr. Mo. Day Hour Min Int Lane # Lane # Lane # Lane # Total NB Total SB Total Volume CNT 18 9920 04 03 08 00 05 005 1 15 2 8 3 9 4 25 23 34 57 CNT 18 9920 04 03 08 00 10 005 1 18 2 12 3 15 4 22 30 37 67 CNT 18 9920 04 03 08 00 15 005 1 16 2 8 3 14 4 28 24 42 66 CNT 18 9920 04 03 08 00 20 005 1 20 2 8 3 17 4 28 28 45 73 CNT 18 9920 04 03 08 00 25 005 1 16 2 7 3 4 4 15 23 19 42 CNT 18 9920 04 03 08 00 30 005 1 18 2 6 3 7 4 22 24 29 53 CNT 18 9920 04 03 08 00 35 005 1 23 2 6 3 9 4 27 29 36 65 CNT 18 9920 04 03 08 00 40 005 1 18 2 4 3 8 4 22 22 30 52 CNT 18 9920 04 03 08 00 45 005 1 13 2 4 3 12 4 28 17 40 57 CNT 18 9920 04 03 08 00 50 005 1 8 2 4 3 14 4 29 12 43 55 CNT 18 9920 04 03 08 00 55 005 1 16 2 4 3 8 4 22 20 30 50 CNT 18 9920 04 03 08 01 00 005 1 12 2 3 3 4 4 24 15 28 43 CNT 18 9920 04 03 08 01 05 005 1 19 2 3 3 8 4 19 22 27 49 CNT 18 9920 04 03 08 01 10 005 1 6 2 2 3 8 4 25 8 33 41

PAGE 80

71 Tag County Site Lane Year Month Day Hour Min Int CL01 CL02 CL03 CL04 CL05 CL06 CL07 CL08 CL09 CL10 CL11 CL12 CL13 CL14 CL15 Total Vol. Buses 1 Trucks 1 HV 1 %HV 1 Total % HV 1 CLS 18 9920 1 04 03 08 00 05 005 0 4 5 0 0 0 0 0 5 0 1 0 0 0 0 15 0 6 6 0.4 CLS 18 9920 2 04 03 08 00 05 005 0 6 2 0 0 0 0 0 0 0 0 0 0 0 0 8 0 0 0 0 0.26 CLS 18 9920 3 04 03 08 00 05 005 0 6 2 0 0 0 0 0 0 0 0 0 0 0 1 9 0 1 1 0.11 CLS 18 9920 4 04 03 08 00 05 005 0 9 1 0 0 0 0 0 13 0 0 0 0 0 2 25 0 15 15 0.6 0.47 CLS 18 9920 1 04 03 08 00 10 005 0 9 5 1 0 0 0 0 2 0 0 0 0 0 1 18 1 3 4 0.22 CLS 18 9920 2 04 03 08 00 10 005 0 9 1 0 0 0 0 0 2 0 0 0 0 0 0 12 0 2 2 0.17 0.2 CLS 18 9920 3 04 03 08 00 10 005 0 11 0 0 0 0 0 0 3 1 0 0 0 0 0 15 0 4 4 0.27 CLS 18 9920 4 04 03 08 00 10 005 0 5 4 0 0 0 0 1 10 0 1 0 0 0 1 22 0 13 13 0.59 0.46 CLS 18 9920 1 04 03 08 00 15 005 0 8 3 0 1 1 0 0 3 0 0 0 0 0 0 16 0 3 3 0.19 CLS 18 9920 2 04 03 08 00 15 005 0 7 1 0 0 0 0 0 0 0 0 0 0 0 0 8 0 0 0 0 0.13 CLS 18 9920 3 04 03 08 00 15 005 0 13 0 0 0 0 0 0 1 0 0 0 0 0 0 14 0 1 1 0.07 CLS 18 9920 4 04 03 08 00 15 005 0 11 7 0 0 0 0 0 7 0 1 0 0 0 2 28 0 10 10 0.36 0.26 CLS 18 9920 1 04 03 08 00 20 005 0 9 3 0 1 1 0 0 6 0 0 0 0 0 0 20 0 6 6 0.3 CLS 18 9920 2 04 03 08 00 20 005 0 6 0 0 0 0 0 0 1 0 1 0 0 0 0 8 0 2 2 0.25 0.29 CLS 18 9920 3 04 03 08 00 20 005 0 8 3 0 1 0 0 0 5 0 0 0 0 0 0 17 0 5 5 0.29 CLS 18 9920 4 04 03 08 00 20 005 0 7 3 0 1 0 0 1 16 0 0 0 0 0 0 28 0 17 17 0.61 0.49 CLS 18 9920 1 04 03 08 00 25 005 0 8 0 0 1 0 0 0 6 0 0 0 0 0 1 16 0 7 7 0.44 CLS 18 9920 2 04 03 08 00 25 005 0 5 2 0 0 0 0 0 0 0 0 0 0 0 0 7 0 0 0 0 0.30 CLS 18 9920 3 04 03 08 00 25 005 0 4 0 0 0 0 0 0 0 0 0 0 0 0 0 4 0 0 0 0 CLS 18 9920 4 04 03 08 00 25 005 0 5 2 0 0 0 0 1 7 0 0 0 0 0 0 15 0 8 8 0.53 0.42 CLS 18 9920 1 04 03 08 00 30 005 0 9 2 0 0 1 0 0 5 0 0 0 0 0 1 18 0 6 6 0.33 CLS 18 9920 2 04 03 08 00 30 005 0 5 1 0 0 0 0 0 0 0 0 0 0 0 0 6 0 0 0 0 0.25 CLS 18 9920 3 04 03 08 00 30 005 0 2 4 0 1 0 0 0 0 0 0 0 0 0 0 7 0 0 0 0 CLS 18 9920 4 04 03 08 00 30 005 0 9 2 0 1 0 0 0 10 0 0 0 0 0 0 22 0 10 10 0.45 0.34 CLS 18 9920 1 04 03 08 00 35 005 0 12 1 0 1 0 0 0 9 0 0 0 0 0 0 23 0 9 9 0.39 CLS 18 9920 2 04 03 08 00 35 005 0 6 0 0 0 0 0 0 0 0 0 0 0 0 0 6 0 0 0 0 0.31 CLS 18 9920 3 04 03 08 00 35 005 0 5 3 0 0 0 0 0 1 0 0 0 0 0 0 9 0 1 1 0.11 CLS 18 9920 4 04 03 08 00 35 005 0 9 2 0 1 0 0 1 13 0 0 0 0 0 1 27 0 15 15 0.56 0.44 1 These categories were calculated from the given loop detector data and added to the class data spreadsheets.

PAGE 81

72

PAGE 82

BIOGRAPHICAL SKETCH David S. Kirschner is a 23-year old graduate student at the University of Florida. He is studying towards his Master of Engineering degree, specializing in transportation engineering. He received a Bachelor of Science in Civil Engineering degree from the University of Florida in December of 2004. 73


Permanent Link: http://ufdc.ufl.edu/UFE0010297/00001

Material Information

Title: Development of a Rural Freeway Level of Service Model Based upon Traveler Perception
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0010297:00001

Permanent Link: http://ufdc.ufl.edu/UFE0010297/00001

Material Information

Title: Development of a Rural Freeway Level of Service Model Based upon Traveler Perception
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0010297:00001


This item has the following downloads:


Full Text












DEVELOPMENT OF A RURAL FREEWAY LEVEL OF SERVICE MODEL BASED
UPON TRAVELER PERCEPTION















By

DAVID S. KIRSCHNER


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF ENGINEERING

UNIVERSITY OF FLORIDA


2005
































Copyright 2005

By

David S. Kirschner














ACKNOWLEDGEMENTS

I would like to thank my committee chair, Dr. Scott Washburn, and my committee

members Dr. Lily Elefteriadou and Mr. Bill Sampson.















TABLE OF CONTENTS



A C K N O W L E D G E M E N T S ............................................................................................... iii

LIST OF TABLES .............. ............................................... ........ vi

LIST OF FIGURES .................................................... .............. ......... vii

A B S T R A C T .......................................... .................................................v iii

CHAPTER

1 INTRODUCTION ...... ............................... ........................... 1

B background ............................................................... ... ........ ............. 1
P problem Statem ent ................................................................... .... ...... ........ .. 2
R research O objective and T asks ...................................................................................... 4
Chapter O rganization......................... .... .. ................ 4

2 L ITER A TU R E R E V IEW .................................... ..................................................... 6

H CM Freew ay LO S M methodology .................................... ............................ ....... 6
Studies Investigating Traveler Perception of LOS ..................................................... 8

3 RESEARCH APPROACH ....... ................................................ .............. 16

A alternative Survey M ethods .......................................................... .............. 16
Video D ata Collection ......... .............................. .. ...... ........... 19
Survey Sessions ........................................ 30

4 ANALYSIS AND RESULTS...................................................... ......................... 35

A n aly sis M eth o d ................................................................... ................................ 3 5
Statistical A nalysis............................................ 39

5 CONCLUSIONS AND RECOMMENDATIONS ............................................... 45

Data Collection and Video Clip Creation............................................. ........... 45
Statistical A n aly sis................. .................................. ................. 4 6


iv









Study Limitations and Recommendations for Further Research............................ 47

REFERENCES .... ............... .... .............. .......... .. 49

APPENDIX

A LOCATIONS OF DATA COLLECTION SITES................................................ 51

B V ID E O CLIP SCR EEN SH O T S......................................................... .................... 55

C RURAL FREEWAY TRIP QUALITY SURVEY FORM.............................. .... 63

D SAM PLE LOOP DETECTOR DATA ........................................................................ 68

BIOGRAPHICAL SKETCH ................. ........... ................. 73














LIST OF TABLES



1. H C M L evel of Service Thresholds ....................................................... .................... 8

2. Data Collection Sites and Traffic Data ............................................. 21

3. Data Collection Times, Locations, and Directions ............................... .... ............... 24

4. Traffic D ata for 13 V ideo Clips ......... ................. ......... .................. .............. 29

5. C lip Sites, D ates, and T im es ........... .................................................... .............. 30

6. Dates and Locations of Survey Sessions ........................................................ 33

7. D ensity M odel E stim ation R esults...................................................... .... .. .............. 40

8. Comparison of Estimated and HCM LOS Thresholds ............. .............. 41

9. Traffic Characteristics Model Estimation Results.................................... ............... 42

10. Level of Service M odel Estim ation Results.......................................... .... .. .............. 43

11. Realism of Video Survey Responses .................................................................... 46














LIST OF FIGURES

1. Camera Setup-Front View, Side View, Speedometer................................................. 23

2. In-Vehicle Equipment Setup.......................................................... .............. 24

3. Setup of a Survey Session ............................................................................ .............. 26

4 Sam ple V ideo Screenshot .................................................................................... 26

5. Illustration of an Ordered Probability Model.............................................................. 37

6. Illustration of an Ordered Probability Model with an Increase in fP............................. 38















Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Engineering

DEVELOPMENT OF A RURAL FREEWAY LEVEL OF SERVICE MODEL BASED
UPON TRAVELER PERCEPTION

By

David S. Kirschner

May 2005

Chair: Scott Washburn
Major Department: Civil and Coastal Engineering

The concept of Level of Service (LOS) is meant to reflect the trip quality a traveler will

experience on a roadway or other transportation facility. Despite this, there have been

relatively few studies that have tried to measure the association of prescribed level of service

assessment methods with traveler perceptions. The objective of this study is to provide

insight into how road users perceive their trip quality on rural freeways, and to examine how

the existing service measure (density) relates to these travelers' perceived trip quality.

Study participants were shown a series of video clips of rural freeway travel from a

driver's perspective, then filled out survey forms indicating their opinion of the trip quality

provided by the conditions in the video clip, and ranked these video clips on a scale from

'Excellent' to 'Very Poor'. In addition, the survey participants were asked to give

background information about themselves and their driving habits in case these factors also

turned out to be significant in influencing perceived trip quality. These video clips were

matched with inductance loop detector data that were collected simultaneously at the data









collection sites, in order to see how well the existing service measure (density) corresponded

to the participants' rankings.

The data from the surveys were analyzed using an ordered probability model to

determine which factors influenced the participants' decisions and how. Three models were

created. The first model used only density as a predictive factor. The second took into

account only roadway and traffic characteristics, and the third examined all the significant

factors that could be gathered from the survey. The 'density only' model showed that density

is indeed a strong indicator of travelers' perceptions of trip quality. A set of LOS thresholds

was also calculated using the survey participants' responses. While the survey thresholds and

the HCM thresholds had similar values for facility failure, the intermediate thresholds

estimated from the survey participants' responses were noticeably lower than the HCM

thresholds. This suggests that travelers' tolerance of congestion is lower on rural freeways

than the HCM indicates. The other models showed the significance of other factors in the

perception of trip quality in addition to density, such as socio-economic information and

personal driving habits.

This study provided some preliminary insight into travelers' perception of trip

quality, but further study is needed. It is suggested that more research be conducted regarding

the effects of different factors on the perception of trip quality, such as a more diverse

population sampling. Eventually, the results from this type of video-based study should also

be compared to results obtained from a comparable in-field driving experiment. This study

indicates the need for a further exploration into the differences between urban and rural

freeways, and possibly a different set of thresholds for rural freeways.















CHAPTER 1
INTRODUCTION

Background

Transportation engineers are responsible for targeting roadway infrastructure

improvements where they will have the most beneficial effect. Since the capital available

for these improvements is limited, engineers must carefully select the projects they

choose to fund so that investments will have the best cost-benefit ratio. In a large part

these decisions are guided by the procedures and methodologies found in the Highway

Capacity Manual (HCM) [1]. The HCM is considered to be the definitive reference guide

for traffic operations and analysis in the United States. The procedures in the HCM are

used to estimate the operational performance of a variety of transportation facilities (e.g.,

signalized intersections, two-lane highways) and the corresponding level of service

(LOS). The assignment of a LOS is based on designated performance measures and

corresponding threshold values for individual facilities. The HCM is published by the

Transportation Research Board (TRB) and its development and maintenance is the

responsibility of the Highway Capacity and Quality of Service (HCQS) committee of the

TRB. The current edition of the HCM was published in 2000.

The concept of LOS is a foundation of the HCM. The LOS of a facility is used in

the HCM as a qualitative indicator of the operating conditions being experienced by

travelers of that facility, under specific roadway, traffic, and control conditions. The

HCM describes LOS as "A qualitative measure describing operational conditions within









a traffic stream, based on service measures such as speed and travel time, freedom to

maneuver, traffic interruptions, comfort, and convenience." LOS is divided into six

categories, A through F in the 2000 HCM. LOS A indicates excellent service and LOS F

indicates extremely poor service. An analysis yielding LOS A would indicate that the

facility is performing extremely well, with low volumes and little congestion. If an

analysis shows a facility to be performing at LOS C, it is in the middle range of

congestion. If a facility is at LOS E, it is still permitting traffic flow but is experiencing

significant delays with conditions approaching capacity. At LOS F, a facility is

experiencing oversaturated conditions and the demand has exceeded the capacity of the

facility.


Problem Statement

The performance measures that are used to calculate LOS for a facility are

referred to as service measures. The currently designated service measures) for each

facility is (are) based on the collective experience and judgment of the members of the

HCQS committee. The same is true with the selection of the threshold values for the

various LOS designations. There is currently no quantitative procedure to define which

values are used as LOS thresholds. The LOS determination process, therefore, is based on

the perspective of transportation professionals. The selection of service measures by the

HCQS committee is, however, guided by two principles: 1) the service measure for each

facility should represent speed and travel time, freedom to maneuver, traffic

interruptions, and comfort and convenience in a manner most appropriate to

characterizing quality of service for the particular facility being analyzed, and 2) the

service measure chosen for a facility should be sensitive to traffic flow such that the









service measure accurately describes the degree of congestion experienced by users of the

facility [2].

The 1985 HCM described LOS as "A qualitative measure that characterizes

operational conditions within a traffic stream and their perception by motorists and

passengers. The descriptions of individual levels of service characterize these conditions

in terms of factors such as speed and travel time, freedom to maneuver, traffic

interruptions, and comfort and convenience" [3]. This statement indicates that the

selection of performance measures and thresholds for the determination of level of

service should be consistent with how operating conditions are perceived by the traveling

public. Until recently, road users' perceptions of quality of service were rarely compared

to the LOS assigned to a facility by the HCM, despite the above definition emphasizing

the importance of reflecting road users' perceived quality of service.

There have been suggestions from within the HCQS committee that a new

approach needs to be explored when selecting a service measure for a facility. Instead of

the measure and corresponding thresholds that transportation professionals (the HCQS

committee) believe represent the quality of service as perceived by travelers, the public's

opinion should be taken into account so as to determine what measure or measures they

associate with quality of service on a transportation facility. Under the current

methodology, the HCQS committee believed that the service measures were highly

correlated with public perception, but this was not known for sure [4]. Since billions of

dollars of transportation investment decisions are made every year based upon the

outcome of HCM level of service analyses, it is essential that the transportation









engineers' assessments of the impact of these investments be consistent with traveler

perception of the investment impacts.


Research Objective and Tasks

The objective of this study was to develop a model for assessing the LOS of a

roadway facility that takes into account the road user's perceived quality of service.

Specifically, this study was focused on rural freeways.

The following tasks were carried out in supporting the above research objective.

Determine appropriate rural freeway sites to perform field data collection

Collect video of roadway and traffic conditions from these sites

Collect traffic data from count stations at these sites

Produce video clips to be shown to survey participants

Develop a survey instrument

Recruit survey participants

Conduct survey sessions

Perform an analysis of survey responses

Develop a level of service model


Chapter Organization

Chapter 2 contains an overview of the current HCM freeway analysis

methodology as well as an overview of relevant literature. Chapter 3 describes the

research approach for this study, including the field data collection, survey instrument

development, survey response data collection, and the statistical analysis method used to

analyze the data. Chapter 4 contains the analysis results. Chapter 5 contains the






5


conclusions and recommendations. Additionally, several appendices with supporting data

and information are included.














CHAPTER 2
LITERATURE REVIEW

The Highway Capacity Manual [1] states that the level of service of a roadway

section should accurately reflect the perceptions of travelers, yet the current methodology

does not directly take these perceptions into account. There have been some recent

studies performed seeking travelers' opinions about what factors and qualities are

important to them in assessing the quality of their trip. A literature review was conducted

to identify these studies and note their findings with regard to the traveler's perception of

LOS.


HCM Freeway LOS Methodology

A freeway is a section of divided roadway with controlled access and two or more

lanes in one direction. Within this definition there are significant differences between

urban and rural freeways. Rural freeways have greater distances between interchanges

than urban freeways, higher speed limits than urban freeways, and a higher percentage of

social and recreational trips than urban freeways. Urban freeways have a higher

percentage of work and shopping trips than rural freeways. Despite these differences,

urban and rural freeways both use density as their service measure with the same

thresholds for LOS.

Traveler expectations and perceptions of quality of service are different for rural

and urban freeways. While urban freeways experience the full range of LOS conditions

from A to F, rural freeways rarely drop below LOS C. Rural freeway travelers have come









to expect these higher levels of service, therefore while urban freeway travelers are

concerned with their overall travel time and the reliability of this travel time, rural

freeway drivers take travel time for granted. Urban freeway drivers expect their ability to

change lanes to be restricted, while a restricted ability to change lanes negatively impacts

a rural freeway user's perceived quality of service [5].

The original HCM had a basic three-point scale to define level of capacity. In

1963 the Level of Service concept was introduced and replaced the previous scale. In

1965 the six-point LOS scale (from A to F) was introduced. In 1985 this six-point scale

was redefined to use traffic density (vehicles per unit length of roadway) as the service

measure for defining LOS on freeway sections. This is the method that is still used today.

Although the concept of LOS is meant to reflect the operational conditions as perceived

by motorists, no freeway LOS methodology in the history of the HCM has been based on

driver perception studies. Therefore, there can be no way to make sure that the LOS

thresholds freeways (as well as any other type of facility) accurately reflect users'

perception of the quality of service they receive.

Under the existing LOS methodology, rural and urban freeways have the same

service measure density, as well as the same thresholds for each rank on the LOS scale.

These thresholds for all freeway sections are shown in the table below.

Do these thresholds accurately reflect the quality of service perceived by travelers

on all freeways, urban and rural? In particular, the studies by Hostovsky [5] and

Washburn [4] indicate that rural freeway travelers may judge the quality of their trip

based on different qualities and criteria. A potential outcome of this study is a set of LOS

thresholds unique to rural freeways. This idea of differing service measures for different










categories of a specific facility type is not a new one. Currently there are two service

Table 1. HCM Level of Service Thresholds

Level of Service Density (pc/mi/ln)
A 0-11
B 11-18
C 18-26
D 26-35
E 35-45
F >45



measures used for assessing the LOS of a two-lane highway. These classes share a

common service measure, but the thresholds are different (one of the classes also uses an

additional service measure). In addition, the HCM procedure for analyzing arterial streets

uses the same service measure for all arterial streets but includes four sets of thresholds

for four different classifications of arterials [1].


Studies Investigating Traveler Perception of LOS


A study by Pecheux et al. [6] noted that the Transportation Research Board's

Committee on Highway Capacity and Quality of Service recognized a need to improve

the HCM methodology of assessing LOS. Specifically, concerns were raised that the LOS

of a roadway section did not correspond to road users' perceptions. The authors felt that

for LOS to accurately reflect travelers' perception of quality of service they would first

have to find out what performance measures were significant to travelers.

The study method involved test participants driving along a pre-selected 40-

minute route, encompassing mostly arterial streets, accompanied by an interviewer and a

traffic engineer. The participant would discuss what factors they personally found









important to the quality of their trip. Participants identified over 40 factors that were

important. These included such factors as intersection efficiency-if the intersection was

being utilized by opposing traffic while travelers were waiting, and the aesthetic qualities

of the intersection. Both of these topics are not covered by the HCM. The study

concluded that more research was needed to focus on traveler perception.

A study by Hostovsky et al. [5] used focus group participants to identify factors

important to trip quality on rural freeways and then compared those findings with those

from a focus group study using regular urban commuters and commercial truck drivers.

The participants in the rural freeway focus group identified three factors that were most

important to trip quality-low density, regular (predictable) travel time, and maintaining

a steady travel speed. Other topics discussed were the safety issues inherent to the

isolated locations of rural freeways, aesthetics, speed differential between cars, the

presence of heavy vehicles, and the need for better traveler information.

When compared to the results of a focus group study involving urban commuters,

it was found that urban commuters placed high importance on the overall speed of their

trip, where rural freeway travelers felt that the ability to choose their speed was a

positive. This reflects the fact that urban drivers rarely have the opportunity to choose

their speed in the traffic stream, so a faster speed is usually preferable over a slower one.

Urban commuters were also not as concerned with the ability to change lanes and move

about the facility at will. Most of the urban drivers were happy if they could stay in one

lane and maintain a desired speed for their trip. The rural drivers were pleased if the

density of the freeway section was low enough to allow movement between lanes and

passing at will. This study was significant due to the fact that it recognized the









differences in how travelers rate their trip quality on an urban versus a rural freeway. The

HCM uses the same methodology for freeways in both types of areas.

A study by Nakamura et al. [7] evaluated traffic flow conditions along an

expressway in Japan from a driver's viewpoint. The study intended to quantitatively

analyze the relationship between traffic flow conditions, drivers' perceptions, and

drivers' behaviors. The field data portion of this study was intended to collect data on

drivers' behavior and perception under various flow conditions. Drivers had a video

camera mounted in their own vehicle and were asked to drive a section along an

expressway. After each trip the subject was asked to complete a survey about the traffic

flow conditions. Twenty-two subject vehicles were used and 105 surveys were collected.

The behavioral data collected was number of lane changes, travel time by lane, and

percent time spent following.

This study found that the most important factor influencing drivers' satisfaction

with their trip was the traffic flow rate. Other factors affecting trip quality were found to

be number of lane changes, and the percent time spent following. Additionally, choosing

the LOS based on the driver's level of satisfaction was attempted and then compared to

the conventional LOS methodology. The results of this comparison suggested that the

traffic conditions on Japanese expressways are not satisfying drivers. The realistic

meaning of this result was that if facilities were designed to the driver's satisfaction level

rather than the conventional LOS it would require an enormous investment.

Several studies have been identified using road-user surveys and video selections

to evaluate LOS methodology. The first study, by Sutaria and Haynes [8], used a road

user survey to evaluate the LOS methodology for signalized intersections. Over 300









drivers were shown video clips taken both from a driver's perspective and from an

overhead camera at an intersection. The film segments were specifically chosen to

represent a specific LOS and were intended to be shown to drivers for one or two signal

cycles. The final compilation shown to drivers included both types of view and the clips

were put in a random order.

Their road user survey consisted of two parts-a group attitude survey and a film

survey. The group attitude survey used a questionnaire, to be answered before the film

portion of the survey. The questionnaire included demographic information such as

gender, age, and education, as well as questions about the participants' driving

experience and the type of roadways the participants usually drove on. They were then

asked to give the relative importance of factors including delay, number of stops, traffic

congestion, heavy vehicle density, and ability to change lanes as these factors applied to

the quality of service at an intersection. After the initial questionnaire the participants

were shown the video clips, consisting of a driver's view of a vehicle approaching,

waiting, and passing through an intersection. After each of these clips the participants

rated the quality of service they felt the intersection provided. At the end of the video

portion the participants were again asked to rate the factors important to quality of service

at an intersection to see if their initial opinion had changed. In all, 310 drivers

participated in this survey.

The results from the survey showed delay to be the most important factor both

before and after the film portion of the survey. This study provided the first results that

took into account the perceptions of travelers and changed the performance measure used

by the HCM to evaluate LOS in the 1985 edition. This study also recommended further









similar studies, and for further studies to simultaneously collect video and traffic flow

data to allow for accurate measurements of what is depicted on the films.

A study conducted by Pecheux et al. [9] addressed the issue of developing a study

method to assess the perceived LOS at signalized intersections. The first objective of all

the study methods was to determine how well the current LOS methodology reflects the

opinion of road users. The second objective was to determine the factors affecting users'

perceptions at signalized intersections.

The participants in this study represented a wide range of ages, education levels,

and incomes. The participants were first shown a series of approaches to signalized

intersections from a driver's perspective. After being shown a sequence of these clips, the

participants were asked to fill out a survey including their attitudes about certain driving

situations as well as their socio-economic information. After filling out these surveys the

participants were asked to discuss the factors that influenced their perception of quality of

service as a group. The study results showed that on average, the participants' delay

estimates were fairly accurate, however individual delay perceptions varied significantly.

Fifteen factors were identified that contributed significantly to quality of service. Finally,

the study found that participants tended to perceive service quality on three or four

distinct levels as opposed to the six HCM levels of service.

Another study using video clips and road user surveys was performed by

Choocharukul et al. [10] with the intention of evaluating the current HCM methodology

of assessing LOS. This study intended to provide a multivariate statistical analysis of the

factors that were important to road users' perception of trip quality and to compare those

factors to the current performance measures for LOS.









The data for this study were collected at various urban freeways. Cameras

mounted on overpasses were used and were focused on sections that included inductance

loop detectors. The cameras were focused so that only one direction of travel could be

observed. The data from the loop detectors were collected and synchronized with the time

of the video clips so the researchers would know the actual traffic flow conditions during

the time the video clips were recorded. Two sets of video clips were chosen, each

containing twelve clips. Two video clips were used for each HCM LOS designation, A-

F, with one clip on the high end of an LOS designation and the other clip at the low end.

These designations were determined by the loop detector information.

There were two groups of survey participants in this survey, one consisting of

students, transportation professionals, and environmental management professionals, and

the other consisting of commercial truck drivers and clerical and support staff. The

participants were provided with written descriptions of the six HCM LOS designations

(directly from the HCM). They were then asked to view the twelve video clips and rank

each of them with the LOS they thought was appropriate for the conditions. The

participants were also surveyed for demographic information such as age and education

levels, as well as information about their driving habits.

This study used an ordered probit statistical model to assess how users perceive

the LOS of the roadway sections. The results of the survey and analysis revealed that

perceived levels of service do not closely follow the HCM. Almost all the participants in

this study had a lower tolerance for LOS A than the HCM, with the average cut-off for

LOS A among the study participants shown to be 7 passenger cars per mile per lane

(pc/mi/ln) as opposed to the HCM cutoff of 11 pc/mi/ln. The HCM threshold for LOS F









also does not correspond with the findings of this study, with the participants selecting an

average of 82 pc/mi/ln as the upper bound of LOS F as opposed to the HCM LOS F of 45

pc/mi/ln. The study also found that factors other than density are likely to influence road

users' perception of quality of service. The results from both groups indicated that a

freeway with 4 lanes (instead of 6 or 8), an increase in traffic density, and an increase in

the standard deviation of vehicle speeds all contributed strongly to a worse perception of

LOS. It should be noted that the use of an overhead view of traffic could likely affect

survey participants' perceptions in a different way than that of an-vehicle view of traffic

and roadway conditions.


Background Study

A study was performed at the University of Florida by Washburn et al. [4] with

the objective of discovering what factors are important to drivers when evaluating the

quality of their trip on a rural freeway. Several methods were considered for this study

(focus groups, video/simulation viewing and review, interviews, etc.) with the final

choice being an in-field survey-based approach. Two hundred and thirty-three travelers

were surveyed at rest stops and service plazas along rural freeways in Florida. These

locations were chosen due to their access to travelers in the process of a rural freeway

trip. It was believed that this in-field survey approach would provide more reliable data,

than mail-back surveys for example, as the drivers' experiences would still be fresh in

their minds when filling out the surveys.

Drivers were asked to rank the factors that contributed to the quality of their trip

on a scale from 1 to 7. The most important factor, ranked in the top three 64.3% of the

time, was the ability to consistently maintain the desired travel speed. The factor with the









next highest ranking was the ability to change lanes freely and pass other vehicles. This

was ranked in the top three 33.3% of the time. The third most important factor was the

ability to maintain a speed no less than the posted speed limit. This factor was ranked in

the top three 33.0% of the time. This preliminary study showed that though density is

important to rural freeway travelers, it is not the most important factor in determining trip

quality. It also showed that drivers consider many other factors when determining trip

quality.


Conclusions

The studies detailed in this chapter have shown that, while some research has

been done on travelers' perception of quality of service, there is a need for more study.

The current HCM methodologies for evaluating LOS may be insufficient for determining

the perceived quality of service from the traveler's point of view. From the studies

summarized in this chapter, we can see that it is possible to understand and approximate a

traveler's perception of quality of service using the factors that are found to be important

to them. This type of research may ultimately assist decision makers when planning for

new roadways and roadway improvements.















CHAPTER 3
RESEARCH APPROACH

This chapter will describe the methods used in collecting the sample data for this

study as well as the methods used to refine the data for use in public surveys .Detailed

within the chapter are the choice of a survey method, the selection of data collection sites,

the creation of the survey form, and the process for conducting a road user survey.


Alternative Survey Methods

Common methods of data collection include the following: focus groups, field

surveys, in-vehicle surveys driven by a researcher, in-vehicle surveys driven by the

research participant, driving simulators, and video surveys.

Focus Groups This consists of recruiting test participants in order to arrange a

roundtable-type discussion about rural freeway travel. Participants would discuss

their rural freeway trip experiences and relate which aspects of rural freeway

travel are most important to them when evaluating the quality of their trip. The

advantage of a focus group is the relative ease of the survey, there is no video data

collection, field work, or liability on the part of the researchers. The disadvantage

is that participants may influence each other's responses and one particularly

vocal participant could swing the rest of the group towards his or her opinion.

Another disadvantage is the lack of a control element for the researchers there is

no one experience on which the participants are basing their opinions, so the

researchers can not look at the data or video record to interpret the responses.

16









Additionally, the potential lack of quantitative feedback upon which to build an

analytical model limits researchers in their ability to predict the responses of other

travelers faced with similar roadway and traffic conditions.

* Field Surveys Researchers distribute survey forms at locations frequented by

rural freeway travelers, such as rest stops or service plazas. The participants give

their opinions on rural freeway travel in a survey form, rating which factors are

most important when they judge their trip quality. One advantage to this method

is that participants surveyed have recently driven on a rural freeway and have this

experience fresh in their mind. Another advantage is that it is relatively easy to

recruit participants for this sort of survey; there is always a ready supply of people

in this type of location. The disadvantages are similar to the focus group.

* In-Vehicle Surveys (driven by research personnel) Participants are recruited and

driven along a section of rural freeway, then surveyed about their perception of

the trip quality. Advantages to this method include all participants would have

the same experience to draw upon for their responses, and there would be no need

to attempt to simulate the driving experience as participants would be

experiencing the conditions firsthand. The disadvantages to this method include

the liability to the researchers should the vehicle be involved in an accident, and

the time and effort involved in conducting a survey of this manner. The

controllability and repeatability of the conditions are also disadvantages because it

is not possible to ensure the same conditions will be experienced by multiple

survey participants.









* In-Vehicle Surveys (driven by research participants) Participants are recruited to

drive along a section of rural freeway and provide the researchers with feedback

on their trip once they return. Once again this method is advantageous in that it

would provide participants with a firsthand look at the conditions involved. The

disadvantages to this are similar to the previous method in that there is significant

liability attached to a method like this, and this method would be even more time-

consuming than the previous one. This method also suffers from the same lack of

control and repeatability as any in-car survey.

* Driving Simulator Participants are put behind the wheel of a real vehicle, but

the driving environment is simulated with the use of computer animation and

video display monitors. They would then participate in the virtual driving of a

rural freeway segment. This would give participants a closer likeness of actual

freeway travel without the liability of having them drive a real section themselves.

Disadvantages include cost (simulator time is expensive) and the well-

documented motion sickness problem for participants (which increases

recruitment time and costs).

* Video Surveys This method involves participants viewing pre-recorded video

scenes from actual rural freeway sites. The clips could be from one of two

perspectives:

o Overhead View A camera placed over the test section of rural freeway

records the traffic flow for survey participants to review at a later time.

While this method does not give a simulation of actually driving the









freeway section, it does give the participant a broader overview of the

traffic stream.

o Driver's Perspective A vehicle is equipped with a video camera to

record the rural freeway trip from the driver's perspective. This method

would better simulate actual rural freeway travel than an overhead view.



After considering all advantages and disadvantages, the method chosen was a

video survey from the driver's perspective. This method would allow larger groups of

people to be surveyed simultaneously while giving a reasonably accurate depiction of

rural freeway travel This method allows for complete control and repeatability of the

conditions experienced by the participants, as well as eliminating the liability issues

inherent in an in-vehicle survey.


Video Data Collection

The data collection method was developed after selecting the form the survey

would take. It included five specific tasks Site Selection, Equipment and Setup, Video

Data Collection, Video Clip Creation, and Loop Data Collection.


Site Selection

The sites at which the video clips were captured were all within Florida. Reasons

for this include the proximity to the University of Florida and the access to the Florida

Department of Transportation's (FDOT) network of more than 7,500 traffic monitoring

stations. The FDOT maintains a network of inductance loop detectors (ILD) along

Florida's Intrastate Highway System. There are two types of ILD stations permanent









and portable. The permanent stations were used in this study since they are continuously

recording data 24 hours a day, 365 days a year. The portable stations require a data

recorder to be hooked up at the location of the ILD station in a roadside cabinet. The

permanent stations are telemetered such that data can be downloaded and archived on a

daily basis. The data archived from the permanent stations are compiled every year by

the FDOT and published on the Florida Traffic Information (FTI) CD. Also included on

this CD for each state highway are the number of lanes in each direction, the ILD station

type (permanent or portable), a description of the site, the Average Annual Daily Traffic

(AADT), the percentage of trucks on the highway, and the peak hour in each direction,

Multiple sites were examined around the state. There were several factors leading

to the final site selections. South Florida was excluded due to the limited number of rural

freeway segments and the long driving times required to get there and back. Sites west of

Tallahassee were also not considered due to driving distance. These conditions hinged

upon the availability of suitable sites in north-central Florida. The final sites selected

represented a mix of four-lane and six-lane freeways, level and rolling terrain, truck

percentages, and a wide range of volume. This information was obtained from the Florida

Traffic Information CD published by the FDOT [11]. All sites selected were permanent

count stations instead of portable stations. Additional data were used from a similar study

conducted months before at the University of Florida. A list of the data collection sites

and their associated traffic data follows in Table 2. Maps of the locations of the data

collection sites can be found in Appendix A.















Table 2. Data Collection Sites and Traffic Data


Volume Two-way K D Truck
Site Site Type Description Direction Direction AADT Factor Factor %
1 2
189920 Telemetered SR-93/I-75, 3.5 mi south of Turnpike, Sumter Co. 20472 N 21250 S 41722 10.1 59.84 21.66
360317 Telemetered 1-75, 0.35 mi north of Williams Rd overpass, Marion Co. 37630 N 37844 S 75474 11.14 55.41 21.76
140190 Telemetered SR-93/I-75, 0.6 mi. south of SR 54, Pasco County 37443 N 37203 S 74646 8.76 53.67 11.71
730292 Telemetered SR-9/I-95, 1.4 mi south of Palm Coast Pkwy, Flagler Co. 29276 N 29980 S 59256 9.91 54.92 17.82
269904 Telemetered SR-93/I-75, 3 mi. north of Marion Co. line, Alachua Co. 31304 N 31023 S 62327 11.96 55.97 19.09
970428 Telemetered SR-91/F1. Turnpike, 797 ft. south of CR561, Lake Co. 17655 N 18088 S 35743 11.09 55.42 12.34










Equipment and Setup

The objective of video data collection was to depict travel along a section of rural

freeway from a driver's perspective. In order to give the survey participants a more

complete representation of the conditions on the freeway section, it was decided that two

more video images would be captured the vehicle's speedometer and the driver side

rear-view mirror. This brought the total number of cameras needed to three. The images

would be combined during the creation of the video clips.

The vehicle used for the video data collection was a minivan. As mentioned

above, three cameras were placed in the vehicle in order to capture different aspects of

the rural freeway trip. The first camera captured the view through the front windshield,

including a view of the interior rear-view mirror. It was placed on a mount secured to the

right side of the driver's seat. The second camera captured a view of the speedometer. It

was mounted on a suction mount affixed to the steering column. It was found during a

preliminary test that the instrument cluster needed to be shaded to reduce glare, so the

image would not appear washed-out. The final camera captured a view from the vehicle's

driver-side rear-view mirror. This camera was mounted on a pole secured between the

driver's seat and the door. Figure 1 shows these cameras as they were mounted in the

vehicle. The video images were captured by three portable VCRs placed inside the

vehicle. A microphone was also connected to one of the VCRs allowing the researcher to

announce when they crossed a detector loop and any other potentially important

information. This would allow the researcher to match the captured video clip to the loop

data collected in a later step. All these devices were powered by three 12-volt deep-cycle

batteries. A schematic of the equipment and connections is found in Figure 2.






























Figure 1. Camera Setup-Front View, Side View, Speedometer


Video Data Collection

The method used to capture the video clips remained constant throughout the

collection of data. The researcher would activate the three VCRs and start recording.

Then the researcher merged onto the freeway. The cameras captured conditions between

the exit ramps that came before and after the ILD station. The researcher would speak

into the microphone when the detector loop was crossed, giving the exact time and site

number so the clip could be matched with the relevant loop data. Up to four runs were

made at each location giving a number of clips to choose from when creating the clips for

the survey. The data collection for this project was performed during November of 2003

and March 6-9, 2004. A summary of each video data collection session is shown in

Table 3.











Camaer 2 Camera 3 Mirophone


Monitor


Figure 2. In-Vehicle Equipment Setup


Table 3. Data Collection Times, Locations, and Directions


Date
11/4/2003
11/5/2003
11/21/2003
11/21/2003
11/21/2003
11/21/2003
3/7/2004
3/7/2004
3/8/2004
3/8/2004
3/8/2004
3/8/2004
3/8/2004


Site
730292
269904
140190
140190
140190
140190
730292
730292
189920
360317
360317
360317
970428


Freeway
1-95
1-75
1-75
1-75
1-75
1-75
1-95
1-95
1-75
1-75
1-75
1-75
Turnike


Direction
NB
SB
NB
SB
SB
SB
SB
NB
SB
SB
NB
SB
SB


After reviewing the video data gathered on the first day of data collection, it was


deemed unusable. The mounting for the camera allowed too much vibration in the picture


and the video would not work for a public survey. Although the second round of video


Time
12:55
11:01
6:48
7:04
7:14
2:31
2:35
2:47
12:36
11:50
12:04
2:07
1:49


Camwa 1









data collection was scheduled for March 6-8, the runs made on the first day needed to be

redone due to problems with the camera placement, necessitating a fourth day of data

collection on March 9, 2004.


Video Clip Creation

The survey participants were shown a single video display that contained the

video scenes of the front windshield and interior rear-view mirror, the driver's side rear-

view mirror, and the speedometer. The display used was a video projector and a wall-

mounted screen, located between 5 and 20 feet away from the participants depending on

the specific survey location. The setup of one of the survey sessions is depicted below in

Figure 3. The majority of the screen was taken up by the view through the front

windshield. Since the front windshield view captured a portion of the dashboard as well

as the view from the front of the vehicle, the other two images could be overlaid on this

area. A screenshot from one of the video clips used in the survey is shown in Figure 4.

Screenshots from all 13 clips can be found in Appendix B.

The clips were assembled using a video-editing program (Adobe Premiere) [12].

They first had to be captured from the VHS tapes using an ADS digital encoder [13].

After they were stored on the computer hard drive they were combined using Adobe

Premiere into clips from 1.5 to 2.5 minutes in length. The length of an individual clip was

chosen based on events in the video that the researchers wanted to include or exclude, as

well as with a survey participant's attention span in mind. The clips shown to viewers

were chosen based on conditions they represented that were unique or different from

other clips. This selection process is explained further in the section entitled "Video Clip

Selection".























Figure 3. Setup of a Survey Session


Figure 4. Sample Video Screenshot


Inductance Loop Detector Data Collection

It was desired to calculate the LOS of these sites according the HCM

methodology in order to assess how strongly correlated it was with the responses









provided by the survey participants. In order to determine the HCM LOS of the rural

freeway segments, data were collected from the inductance loop detector (ILD) stations

at each test site. The data collected came in three files-speed, volume, and vehicle

classification. FDOT personnel programmed the detectors at the sites selected for the

study to record data in five-minute intervals (the hardware minimum interval) rather than

the usual one-hour interval. This shorter interval allowed for traffic data that more

accurately reflected the conditions depicted in the video clips. It should be noted that

even with a five-minute data collection interval the conditions shown in the video clips

could potentially vary from the average provided by the ILD data. These ILD data were

used to categorize the collected video data and provided a starting point for selecting a

range of conditions to be represented in the survey.

The ILD data were provided in the form seen in Appendix D. When available,

there were three data files for each site speed, count, and class. In the speed file, counts

are provided for each speed range. The midpoints of the speed ranges are shown at the

top of the table. In the class file, descriptions such as "CL01" are given to the columns.

These refer to the specific class of vehicle counted in that group and are explained by the

figure provided. From the data provided it was possible to calculate descriptive statistics

for the traffic flow at each site, such as the percentage of heavy vehicles in the traffic

stream, the total 5-minute volume, the average speed, and the density.


Video Clip Selection

There were thirteen video clips chosen for the final survey. The final number of

clips chosen was a result of five pilot test sessions, striking a balance between coverage

of alternatives and attention/focus span of participants. These preliminary tests had






28


shown that many participants lost interest after two minutes and had already started

writing their opinions down. The final video clips were chosen to represent a variety of

conditions in categories including lane configuration, traffic density, terrain, truck

percentage, the presence of a median or guardrail, and shoulder configuration. The

relevant data for each video clip included in the final survey is included below in Table 4

and Table 5.


















Table 4. Traffic Data for 13 Video Clips


ILD LD ILD
Clip Clip Truck Inside Middle Outside Inner Middle Outer Terran
# Road Dir Lanes Length Volume' Density Truck Speed Speed Speed Avg Lane Lane Lane 5min Terrain Speed
# Length %0 Speed Speed Speed Av Lane Lane Lane Volume
% Speed Volume
1 1-75 S 2 2:10 low none 8.00 0.13 77.1 72.2 74.30 42 57 99 flat 75
2 1-75 S 3 1:52 med-high high 63 75 66 204 flat 70-75
3 1-75 N 2 2:00 med-high med 13.79 0.20 76.4 69.4 72.20 67 99 166 flat 60-70
4 1-95 N 2 1:35 very high 26.45 56.5 55.4 56.00 102 145 247 flat 40-55
5 1-75 S 3 1:40 low-med med 6.30 0.17 77.6 74.0 66.4 72.30 26 51 37 114 rolling 70-75
6 1-95 S 2 1:59 med 10.21 76.9 74.7 75.80 63 66 129 flat 70
7 1-75 S 2 2:00 med-high high 26.31 0.34 71.6 65.5 68.90 167 135 302 flat 67-72
8 1-75 N 3 2:01 med-high low 61 98 80 239 flat 67-72
9 1-75 S 2 2:00 high high 26.11 0.32 71.3 66.1 69.20 179 122 301 flat 55-65
10 1-95 N 2 1:43 med 10.86 78.4 69.4 72.90 80 52 132 flat 75
11 1-75 S 3 1:26 med med 48 82 48 178 flat 70-75
12 1-75 S 2 1:27 med-high med 17.04 0.15 73.6 68.2 71.10 110 92 202 flat 60-65


13 Turnpike S 2 2:03 med high


rolling 75-80


1 These levels (low, med, high) indicate subjective judgments that were used to choose between clips.










Table 5. Clip Sites, Dates, and Times


Clip # Clip Site Time Date Closest City
1 189920 run 1 189920 12:36 3/8/2004 Wildwood
2 360317 run 1 360317 11:50 3/8/2004 Ocala
3 Tampa 0648 140190 6:48 11/21/2003 Tampa
4 730292 run 4 730292 14:47 3/7/2004 Daytona Beach
5 Micanopy 1101 269904 11:01 11/5/2003 Micanopy
6 730292 run 3 730292 14:35 3/7/2004 Daytona Beach
7 Tampa 0714 140190 7:14 11/21/2003 Tampa
8 360317 run 2 360317 12:04 3/8/2004 Ocala
9 Tampa 0704 140190 7:04 11/21/2003 Tampa
10 Daytona 1255 730292 12:55 11/4/2003 Daytona Beach
11 360317 run 3 360317 12:07 3/8/2004 Ocala
12 Tampa 1431 140190 14:31 11/21/2003 Tampa
13 970428 run 1 970428 13:49 3/8/2004 Winter Garden



Survey Sessions


Development of Survey Form and Participant Instructions


The survey form for this study had to serve two purposes-record the

participants' opinions about the rural freeway video clips and their reasons for these

opinions, and record characteristics about the participants that might influence their

ratings. Thus the form is divided into two sections.

The first section of the survey form is for personal information about the traveler

taking the survey. Examples of this information include education level, income, and

number of years possessing a driver's license. This section also records information about

the participant's rural freeway travel habits. It asks for information such as the amount of

rural freeway trips taken per month and the average length of the participant's rural

freeway trips. Finally it asks for some driving habits, such as any changes in the









participant's driving style when driving alone versus with a passenger. It also asks the

participant to rate their usual driving style, from Conservative to Aggressive.

The second section of the survey is for recording the participant's opinions and

rankings of the video clips. It is divided into two sections for each of the thirteen clips.

The first section asks the participant to rank the quality of the trip depicted in the video

clip on a scale from 'Very Poor' to 'Excellent' with 6 total ranking levels. A total of six

ranking levels was chosen so that there would be general correspondence with the six

levels of the HCM (A-F). Participants were asked to use the word ranking rather than a

numerical ranking (e.g., 1-6) to minimize the possibility that those familiar with the HCM

might try to equate the numerical rankings with the HCM LOS rankings. The second

section asks the participant to record why they ranked the video clip as they did, listing

all factors that significantly contributed to their ranking. The participants were to then

number these according to their relative significance to each other.

Finally the form includes questions about the survey itself. These include the

participant's opinion on the video clips as a representation of rural freeway travel and if

the participant would have changed their rankings based on the purpose of the trip (e.g.,

business, recreational, or social).

A one page written survey instruction sheet was developed because there was a

significant amount of information that needed to be communicated to the participants in

order for them to complete the survey form in a manner which would be useful as study

data. The participants could refer back to it if there were any questions about the survey

process. The instructions given to each survey participant are provided in Appendix C.









Conducting the Survey Sessions

Survey participants were recruited from various sources. They include the

following:

Undergraduate students in the University of Florida civil engineering program,

recruited from the introductory transportation engineering course,

Graduate students in the University of Florida civil engineering program,

recruited from the transportation degree program,

Employees of the University of Florida Technology Transfer Center,

Employees of the Florida Department of Transportation, and

Alachua county residents (Random participants recruited for a fee by the Florida

Survey Research Center)

The undergraduate students were recruited from the Principles of Highway

Engineering and Traffic Analysis course during the Fall 2004 semester. The graduate

students were those enrolled in a transportation engineering degree program during the

Fall 2004 semester. The University of Florida Technology Transfer Center is an

organization that provides training and technical assistance to Florida's transportation and

public works professionals. Their survey session was conducted at their off-site

headquarters in Gainesville, FL, with participants ranging from high-school educated

support staff to professionals with graduate degrees. The FDOT survey session was

conducted at the central office in Tallahassee, FL. This session also included participants

of varying backgrounds and demographics. The public sample was comprised of Alachua

county residents, recruited by the University of Florida Survey Research Center. The

survey center was instructed to recruit individuals with varying socio-demographic










characteristics and also make sure that the participants had experience driving on rural

freeways. Additionally, they did not recruit college students as there was already a

sufficient number in this group.

In total there were 126 surveys filled out for this study. The locations, dates, and groups

of participants taking the survey during each session are given in Table 6.


Table 6. Dates and Locations of Survey Sessions


Survey # of
Session Date City Location Participants Surveys
1 8/4/04 Gainesville UF Technology Transfer Center T2 employees 16
2 11/16/04 Tallahassee Florida DOT Central Office DOT employees 11
3 12/2/04 Gainesville University of Florida undergraduate students 14
4 12/2/04 Gainesville University of Florida undergraduate students 9
5 12/4/04 Gainesville UF Hilton Conference Center public' 13
6 12/4/04 Gainesville UF Hilton Conference Center public' 15
7 12/4/04 Gainesville UF Hilton Conference Center public' 11
8 12/9/04 Gainesville University of Florida undergraduate students 20
9 1/22/05 Gainesville University of Florida public' 9
10 1/27/05 Gainesville University of Florida graduate students 8
Total Number of Surveys 126
Participants were recruited through the University of Florida Survey Research Center




Because of the video format of the survey, multiple surveys could be filled out at

a time, the main limitations being the ability of the participants to comfortably view the

video screen and the length of time for which the participants could be expected to focus

on this task. The screen was placed as close as possible to eye level so participants

looking at the screen saw it as they would a car's windshield. Before viewing the clips

the participants were given the instruction sheet and time to read it. These written

instructions were also verbally reviewed by the session moderator, as well as some

supplemental information. The participants were also told that they could ask









interpretation questions in-between the viewing of the video clips. It was decided to

create two example clips, each 20 seconds long, to show the upper and lower ends of the

range of possible traffic flows. The first was a nearly empty four-lane freeway and the

second was stop-and-go traffic along a four-lane freeway. The participants were then

shown each of the 13 video clips and instructed to watch each clip entirely before writing

their responses. Since it was not intended for the order of the clips to have any effect on

the participants' rankings, the order was shifted for each survey session. After each clip

was finished, the participants were given time to record their rankings.















CHAPTER 4
ANALYSIS AND RESULTS

This chapter contains information about the methodology used to analyze the

survey data, as well as the results of these analyses.


Analysis Method

To determine how or if the participants' responses correspond to the six LOS

rankings, a statistical analysis was needed to predict the probability of selecting discrete

rankings (1-6 as included in the survey). While one of several multinomial discrete-

choice modeling methods would suffice to predict a discrete outcome, most do not take

into account the ordered nature of the responses in this survey (1 is better than 2, which is

better than 3, etc.). Using a standard multinomial discrete model, such as a multinomial

logit model, would still yield consistent parameter estimates, but with a loss of efficiency

[14]. In order to account for the discrete and ordered responses in this survey, an ordered

probability model was chosen as the statistical analysis approach.

An ordered probability model is derived by defining an unobserved variable, z,

that is the basis for modeling the ordinal ranking of data (in this case the six clip

rankings) [15]. This variable is specified as a linear function for each observation n such

that



Zn = Xn + En (1)









where X, is a vector of variables determining the discrete ordering for observation n, f is

a vector of estimable parameters, and En is a random disturbance. In this analysis, y is

defined as each participant's evaluation of each of the 13 video clips. Since there are 126

participants and 13 clips, there are a total of 1638 observations. Using this equation, the

observed clip ranking, yn for each observation is written as



Yn = 1 ifzn < u

Yn = 2 if/Ul z, < aU2

yn = 3 ifU2< z, < U3 (2)

Yn = 4 if U3< Zn < p4

Yn = 5 ifi4< Zn <_U/

Yn = 6 ifz, > us



where the p values are the thresholds that define Yn. The u values are estimated jointly

with the model parameters (P). The estimation problem then becomes one of determining

the probability that a participant will select a particular ranking for each clip. In using the

ordered probit model, it is assumed that the error term, en, is normally distributed with a

mean of 0 and a variance of 1. The resulting ordered probit model has the following

probabilities corresponding to each clip ranking:



P(yn = 1) = 0(-fX,)

P(yn = 2) = 1(u, PX,) 0(-fX,,)

P(~n = 3) = 0(U2 PX,) O(j1 fX,) (3)









PO2n

P(y'

P(y'


1(U flX.) 1(U2- flX.)

1(4 -- '-(U2 -PXn)

\-^4~-PXn)


It can be shown that threshold yi can be set equal to 0 without loss of generality [15]. In

the above equations, 0(.) represents the cumulative normal distribution:


1 ~1,
(u) = \ e 2 dw
v^-00


This model can be estimated using maximum likelihood procedures.

The thresholds u1 and u1 define the upper and lower thresholds for outcome i.

This is illustrated in Figure 5.


y -4


y=2


y= I


y 5


y=6


Figure 5. Illustration of an Ordered Probability Model


A positive increase in the f term implies that an increase in x will increase the probability

that the highest category response will be returned (in this case, y = 6). An increase in the


-f fly it) -/ fly n.,. P -PC









p term also implies that the probability of returning the lowest response (y

decreased. This is illustrated in Figure 6.


1) is


y=4


y=2


y=l


y=5


y=6


Figure 6. Illustration of an Ordered Probability Model with an Increase in f

A unique issue was present in this data set that complicated the analysis

procedure. Each of the 126 participants viewed 13 clips and thus generated 13

observations. The issue is that there are unobserved characteristics that are unique to each

participant that will be reflected in all 13 of their rankings. If this is not accounted for in

the model, the model will be estimated as though each of the 1638 observations came

from a unique participant. This approach would result in lower standard errors in the

model's estimated parameters, leading to inflated t-statistics and exaggerated degrees of

significance.

The solution to this problem is found in a standard random effects approach. The

first equation is rewritten as


- fly iA -tix -A 4-











zic = PXic + E,c + P, (5)



where i denotes each participant (i = 1,...,126), the c denotes each video clip (c =

1,...,13), p, is the individual random effect term and all other terms are as previously

defined. The random effect term p, is assumed to be normally distributed with mean 0

and variance oa. When this random effects model is estimated, an estimate of o is also

calculated, the significance of which determines the significance of the random effects

model relative to the standard ordered probit model [16].


Statistical Analysis

The results of the surveys were put into spreadsheet form, with unique cases for

each clip viewing. Each participant's rankings were kept together within the spreadsheet

for analysis purposes. The data were analyzed using LIMDEP [17] with a random effects

approach as detailed in the previous section.

The first analysis was performed to explore how the quality of service perceptions

of the participants in this survey correlated with the HCM LOS thresholds. The density

for each of the video clips was calculated from the loop detector data (and the video data,

in cases where the loop detector data was incomplete). A statistical analysis was

performed using density as the only independent variable to find out where the thresholds

of the survey participants fell relative to the six clip rankings. The results are given below

in Table 7.

The very high level of significance indicated by the t-statistic (coefficient divided

by standard error) calculated for density in the above model offers some evidence that










this performance measure correlates well with perceived LOS. The reference t-statistic

for these analyses is 1.282, representing a 90% confidence level in a one-tailed t-test. The

positive coefficient calculated for density indicates that, as density increases, the

likelihood of a traveler perceiving a worse LOS increases. The random effects term, a, is

also highly significant, meaning that the choice of a random effects model for this data

set was correct. Had this term not been significant, a normal ordered probability model

would have been sufficient.

One test for the goodness-of-fit of a model is calculating that model's p2 value.

The p2 value of a model is between 0 and 1. A p2 value of 1.0 indicates a perfect model

fit. The p2 value of a model is calculated as follows:

p2 LL()- K(6)
LL(0)

where K represents the number of variables in the model, LL(fl) represents the log

likelihood at convergence, and LL(O) represents the initial log likelihood [15].

Table 7. Density Model Estimation Results

Standard
Variable Coefficient Error t-statistic
Constant -0.138 0.076 -1.82

Traffic ( I i ,. i ....
Density (pc/mi/ln) 0.096 0.003 34.37

Threshold Values
/l 0.918 0.038 23.89
P2 1.922 0.048 39.92
/3 2.863 0.053 53.88
/P4 4.112 0.066 62.47

Standard Deviation of Random Effects
a 0.455 0.050 9.12

Initial Log Likelihood -2710.16
Log Likelihood at Convergence -2314.60
p2 0.15









Using the participants' responses it was possible to calculate a set of thresholds

for the participants' assigned LOS rankings. Using the calculated values in Table 7, the

threshold values can be calculated as (Uk o)/fll. In this equation, k designates the five

threshold values, uj = 0, and the other threshold values are given in Table 7. A

comparison between the calculated threshold values from this survey and the HCM LOS

thresholds is given in Table 8.


Table 8. Comparison of Estimated and HCM LOS Thresholds

Estimated Thresholds HCM thresholds
LOS
(pc/mi/ln) (pc/mi/ln)
A 0-2 0-11
B >2-11 >11-18
C >11-21 >18-26
D >21-31 >26-35
E >31-44 >35-45
F >44 >45


These thresholds are generally lower than the HCM thresholds for corresponding

rankings, indicating the participants in this survey had a lower tolerance for high-density

traffic conditions than could be inferred from the HCM LOS thresholds.

The second analysis that was performed was intended to take into account all the

traffic and roadway characteristics influencing the participants' perception of trip quality.

The results of this table are given below in Table 9.

The traffic characteristics examined produced effects according to expectations.

The calculated difference in speed between the inner lane and the outer lane was in the

model as "speed differential". As this value increased, participants were more likely to

assign a worse LOS to a given set of conditions. A higher average speed resulted in a

more favorable LOS ranking. Motorists in this survey found three lanes in one direction










to be a preferred configuration over two lanes and were more likely to assign a favorable

LOS ranking to those roadways with three lanes in one direction.


Table 9. Traffic Characteristics Model Estimation Results


Standard t-
Variable Coefficient Error statistic

Constant 6.296 0.597 10.55

Traffic ( Ii. i ....
Speed Differential (mi/h) 0.163 0.027 6.08
Average Speed (mi/h) -0.096 0.009 -10.97
3 Lanes (1 Yes, 0 No) -1.848 0.210 -8.82
Truck % 0.005 0.004 1.04
Density (pc/mi/ln) 0.061 0.006 10.59

Threshold Values
p/ 0.949 0.064 14.88
P2 2.192 0.077 28.48
P3 3.258 0.092 35.60
/P4 4.630 0.106 43.80

Standard Deviation of Random Effects
o 0.522 0.060 8.76

Initial Log Likelihood -2710.16
Log Likelihood at Convergence -1472.53
p2 0.45


An increase in the truck percentage resulted in a higher possibility of a worse LOS

ranking. While the t-statistic for the truck percentage was below 1.282, it was decided to

leave this variable in the model because it was felt that this was a very important variable

from a policy standpoint. As expected, the participants preferred not to have a high

percentage of trucks in the traffic stream. Finally, density was very significant in this

model as it was in the first. A higher density led to an increased possibility of a worse

LOS ranking. The random-effects term was again significant in this analysis, justifying

the use of a random-effects model.











The third analysis that was performed was aimed at discovering which factors are

important to travelers when judging their trip quality. This model was estimated

including demographic data as well as roadway and traffic flow characteristics. The

values given in Table 10 should be interpreted such that a positive parameter estimate

means that an increase in that variable will lead to a better perceived quality of service,

and a negative parameter estimate means that an increase in that variable will lead to a

worse perceived quality of service.


Table 10. Level of Service Model Estimation Results


Standard t-


Variable
Constant

Demographic and Background Information
Age > 35 (1 Yes, 0 No)
Income (thousands of $)
Average Number of Rural Freeway Trips per Month
Average One-Way Trip Distance > 100 miles? (1 Yes, 0 -
No)
Less Aggressive Driver with Passengers? (1 Yes, 0 No)

Traffic ( i ... .i,..i .
Speed Differential (mi/h)
Average Speed (mi/h)
3 Lanes (1 -Yes, 0 -No)
Truck %
Density (pc/mi/ln)

Threshold Values


Standard Deviation of Random Effects


Initial Log Likelihood
Log Likelihood at Convergence
p2
P


Coefficient Error statistic


6.156 0.622


-0.358
-0.003
0.025

0.395
0.267


0.162
-0.095
-1.836
0.005
0.062


0.939
2.181
3.247
4.613


0.435


0.121
0.002
0.017

0.127
0.186


0.028
0.009
0.217
0.005
0.006


0.065
0.078
0.093
0.107


0.059


9.90


-2.96
-1.89
1.49

3.11
1.43


5.85
-10.58
-8.47
1.03
10.56


14.56
27.93
34.90
43.21


7.42

-2710.16
-1447.34
0.46









In Table 10, a positive coefficient value indicates that as the variable increases,

there is an increased likelihood of a worse perception of LOS. Likewise, a negative

coefficient value indicates that as the variable increases, there is an increased likelihood

of a better perception of LOS.

The results indicate that, while density is important to travelers, it is not the only

factor influencing perceived quality of service. The survey results showed significant

effects of demographic and background information on drivers' LOS rankings. Table 10

indicates that participants with over 35 are more likely to assign a given set of conditions

a better LOS, as are those with higher incomes.

Travelers who drive on rural freeways more frequently are more likely to perceive a

worse LOS, as are those whose average rural freeway trip is over 100 miles in one-way

length. Those participants who indicated that they tend to drive less aggressively with

passengers in the car as opposed to driving alone were more likely to assign a worse LOS

to a given set of conditions. A possible explanation is that these drivers are more

aggressive than the average motorist. Participants were asked if they considered

themselves to be an aggressive driver, and the results of that model did not display

significance. Perhaps motorists were more reluctant to admit they drive aggressively, but

this tendency manifests itself in their responses to this question.

The results estimated using the traffic and roadway characteristic variables

showed similar significance and magnitude to the model estimated only using these

variables. The random effects term was once again significant.















CHAPTER 5
CONCLUSIONS AND RECOMMENDATIONS

Since 1963, the Level of Service concept has been integral to the Highway

Capacity Manual methodology for assessing the performance of transportation facilities.

There is, however, still relatively little known about how the HCM methodologies for

assigning LOS correspond to road users' perceptions of their quality of service. The

purpose of this study was to investigate what factors influenced road users' perceptions

of quality of service, and how that perception compares to HCM calculated LOS.


Data Collection and Video Clip Creation

The data collection process used for this study proved successful in gathering the

necessary video data. After deciding on the best camera positions and mounting

techniques, all cameras recorded clear, steady views of their intended targets. The

equipment in the vehicle performed exactly as intended, capturing the necessary

information while keeping all three VCR timers consistent so the video data could be

synchronized at a later time.

The sites chosen generally provided a good variety of traffic conditions, but some

clips from a pilot study were also used to provide additional roadway and traffic

conditions that were not captured in the data collection effort for this project. These clips

were re-edited using the same process as the clips filmed for this study so there would be

consistency in the screen views.









The loop detector data did not work out as well for some of the sites as was

initially hoped. Due to malfunctioning detectors or construction at the selected sites,

some of the desired data were not available.

The final form of the video clips and the presentation to survey participants

worked very well, exactly as intended. The last question on the survey form (as seen in

Appendix A) asked participants to rate how well the video clips simulated the driving

experience for the conditions depicted on the screen. The majority of participants found

the survey to be a "very good" representation of the actual driving experience, with 95%

of the participants rating the survey as a "good" or better representation of the actual

driving experience. The responses to this question are tabulated in Table 11. As shown in

this table, the average response from participants was approximately a 2 out of 6,

corresponding to "very good".


Table 11. Realism of Video Survey Responses


Ranking Excellent Very Good Good Fair Poor Very Poor
1 2 3 4 5 6
Frequency 21 64 36 5 1 0
Percent of Total 1
17 50 28 4 1 0
Responses (%)
Average Rank 2.2


Statistical Analysis


The analysis process chosen for this survey was an ordered probability model,

specifically the ordered probit model. The structure of the standard ordered probit model

formulation does not account for each participant providing 13 responses, so a random-

effects formulation was used. This modeling choice was justified, with the standard

deviation of random effects showing significance in all statistical analyses.









The first model developed was one incorporating only density as an independent

variable. This produced results that were as expected, that density is very significant to

travelers when they are judging the quality of service provided by a rural freeway.

A complimentary outcome of this analysis was that density thresholds for each

LOS were estimated according to the survey participants' responses. For LOS A-E, the

survey participants showed a lower tolerance for high-density traffic conditions, hence

their estimated thresholds were lower. The HCM thresholds and the estimated thresholds

showed similar values for LOS F.

The second model was estimated to include the influence of other roadway and

traffic characteristics. The results of this model showed that while density is significant to

user perception of LOS, there are other significant factors influencing this perception,

such as average speed of the traffic stream and the speed differential between lanes.

The final model included all factors from the survey that were found to be

significant, including demographic factors as well as roadway and traffic characteristics.

The results of this model indicated that the background and characteristics of the

individual road user can influence their perception of LOS. While this result was

expected, it is still significant due to the implications for a potential future modification

to the HCM LOS methodology.


Study Limitations and Recommendations for Further Research

Since the scope of this study was limited to North Central Florida, additional

testing with participants from a variety of other geographic regions would be needed to

adopt any findings on a national level. An expanded sample, both geographically and in

roadway conditions, would provide much more comprehensive coverage of the roadway









and traffic condition combinations. The video survey format has inherent limitations as

well. In a future study, it would be desirable to allow road users to drive in a traffic

stream with known characteristics (density, truck percentage, etc.), then express their

opinion regarding the LOS of the roadway section. This was not considered for this study

due to cost and liability. The results of this survey could be compared to the results of the

video survey to assess the accuracy of the video survey. If the video survey is shown to

be an accurate method of simulating traffic conditions, it can be used in future studies and

will be more effective than in-field surveys. Finally, although participants were told to

imagine the conditions in the video scenes as if they were occurring throughout the

duration of a trip, it is not known whether actually experiencing these conditions for an

equivalent time to an entire trip would change the outcome.

It is hoped that the findings of this study will lead to further developments in this

area. The study does show that density is significant in determining a road user's

perception of trip quality. It is also known that there are significant factors influencing

LOS other than density and these should be explored more completely. Ultimately, a

better understanding of travelers' perceptions of quality of service will lead to a better use

of the available resources to improve the roadway network where it is really needed, and

to more accurate planning and accommodating for future demands.















REFERENCES


1. Transportation Research Board (2000). Highway Capacity Manual. TRB,
National Research Council. Washington, D.C

2. Harwood, D., Flannery, A., McLeod, D., Vandehey, M. (July 2001). The Case for
Retaining the Level of Service Concept in the Highway Capacity Manual.
Presented at the 2001 Transportation Research Board Committee A3A10 -
Highway Capacity and Quality of Service Midyear Meeting, Truckee, California.

3. Transportation Research Board (1985). Special Report 209: Highway Capacity
Manual. TRB, National Research Council. Washington, D.C

4. Washburn, S., Ramlackhan, K., McLeod, D. (2004). Quality of Service
Perceptions by Rural Freeway Travelers: Exploratory Analysis. Transportation
Research Record: Journal of the Transportation Research Board, No. 1883.
Washington, D.C., pp. 132-139.

5. Hostovsky, C., Wakefield, S, Hall, F. (2004). Freeway users' Perception of
Quality of Service: A Comparison of Three Groups. In Transportation Research
Record: Journal of the Transportation Research Board, No. 1883. TRB, National
Research Council. Washington, D.C., pp. 150-157.

6. Pecheux, K., Flannery, A., Wochinger, K., Rephlo, J., Lappin, J. (2004).
Automobile Drivers' Perceptions of Service Quality on Urban Streets.
Transportation Research Record: Journal of the Transportation Research Board,
No. 1883. TRB, National Research Council. Washington D.C. pp. 167-175.

7. Nakamura, H., Suzuki, K., Ryu, S. (2000). Analysis of the Interrelationship
Among Traffic Flow Conditions, Driving Behavior, and Degree of Driver's
Satisfaction on Rural Motorways. Transportation Research Circular E-C018:
Proceedings of the Fourth International Symposium on Highway Capacity.
National Research Council. Washington, D.C., pp. 42-52


8. Sutaria, T.C., and Haynes, J.J. (1977). Level of Service at Signalized
Intersections. Transportation Research Record: Journal of the Transportation
Research Board, No. 644. TRB, National Research Council. Washington, D.C.,
pp. 107-113.









9. Pecheux, K., Pietrucha, M., Jovanis, P. (2000). User Perception of Level of
Service at Signalized Intersections: Methodological Issues. Transportation
Research Circular E-C018: Proceedings of the Fourth International Symposium
on Highway Capacity, National Research Council. Washington, D.C., pp. 322-
335.

10. Choocharukul, K., Sinha, K., Mannering, F. (2004). User Perceptions and
Engineering Definitions of Highway Level of Service: an Exploratory Statistical
Comparison. Transportation Research Part A, 38. pp. 677-689.

11. Florida Traffic Information 2003. (2003). Florida Department of Transportation,
Tallahassee, FL, CD-ROM.

12. Users Guide for Adobe Premiere Pro Software. (n.d.). Last Accessed November
17, 2003, from http://www.adobe.com/products/premiere

13. Users Guide for ADS Pyro A/V Link. Last Accessed March 15, 2005, from
http://www.adstech.com/products/API-555/intro/api555_intro.asp?pid=API-555

14. Amemiya, T. (1985). Advanced Econometrics. Harvard University Press.
Cambridge, MA.

15 Washington, S., Karlaftis, M., Mannering, F., 2003. Statistical and Econometric
Methods for Transportation Data Analysis. Chapman & Hall/CRC. Boca Raton,
FL.

16 Greene, W., 2003. Econometric Analysis. Prentice Hall. Upper Saddle River, NJ.

17. Users Guide for LIMDEP 8.0. 2004. http://www.limdep.com Econometric
Software, Inc. Last Accessed March 23, 2005.





















APPENDIX A
LOCATIONS OF DATA COLLECTION SITES







LEWISR

9189920

189920
-I I ',-- \






52








47 ,iop
L 7. .
-* PU- N :wl


M ant* .


6.
--- I_ I -
I %,- ,.- _Q


Jl -iK A,+.
"'


I, *.^A,
S73029








20
AGLER -- 1 ':' I




,jqi.
-, ...\, ..



200
K .






53








-i' / A/0190
,581








*--, HI BOROUGH /























415A
7 30317
328- -






54









/ ^26




234








,269904
'3 46 3 46. .3
S241 .. ,


d? Li





















APPENDIX B
VIDEO CLIP SCREENSHOTS









56





















. .. .-






57




58






I
mI






59






60








61
















































. .. .








62




































II





















APPENDIX C
RURAL FREEWAY TRIP QUALITY SURVEY FORM




















UNIVERSITY OF
FFLORIDA RC
Transportation Research Center

Rural Freeway Trip Quality Survey

In the exercise you are about to participate in, you will be watching a series of 13 short video
segments of various roadway and traffic conditions on rural frec'. a' s A rural freeway is a
freeway that travels through relatively unpopulated areas. Rural freeways are typically used for
longer trips, such as city-to-city trips. All freeway segments (whether in urban, rural, or other
types of areas) are characterized by opposing directions of traffic being separated by either a
physical barrier or open space. All freeways are also characterized by limited access, that is,
entry to and exit from a freeway can only be made at interchanges (on- and off-ramps). For rural
freeways, interchanges are spaced much further apart than along freeways in urban areas.

Each of the video clips is approximately 1.5 to 2 minutes in length. Each clip is intended to give
you a "snapshot" of the typical conditions experienced over the course of an extended trip on a
rural freeway. When watching each video clip, please imagine and/or keep the following points
in mind:
The conditions viewed on the video clip for about 2 minutes are intended to be
representative of what you would experience for a much longer trip (30 minutes or more).
Imagine how you would personally drive, or try to drive, in the given conditions. You
are not limited to the driving behavior of the vehicle from which the video is being.
viewed. The intent of the video vehicle is to provide you with a reasonable
representation of the typical conditions being experienced by ALL motorists on that
section of rural freeway. Therefore, your survey responses should not be specific to how
the video vehicle was being driven. If you feel like you would, and could, drive
differently under the given conditions, then base your survey responses on that. It is
important that your survey responses reflect how the given conditions affect your
perception of trip quality based upon your own desired driving behavior.

After watching each video clip, we ask that you do the following on the survey form:
Rank (from Very Poor to Excellent) the travel conditions
In the space provided, briefly list the reasons/factors for why you ranked the conditions in
that video clip as you did. Please be as specific as possible-for example, you might say
'opportunities to pass other vehicles in order to maintain my desired speed were limited',
as opposed to 'speed was too low'.

The video clips are intended to be weather neutral-that is, in developing the video clips it was
not our intent to have weather be a significant factor in your trip quality perceptions. Although
the lighting conditions may vary somewhat, please do not factor in the environmental conditions
unless you feel very strongly about a certain condition.

If you recognize the freeway section, disregard previous knowledge and experience and base
your ranking strictly upon the conditions observed in the video clip.


Thank you for ;, our ciotperaliori and participation.





















L UNIVERSITY OF ,
FLORIDA
-- Transportation Research Center

About Yourself

Gender: 0 Male a Female

Age: 0 16 to 25 years o 26 to 45 years D 46 to 65 years E Over 65 years

Marital Status: D Single L Married [ Separated/Divorced E Widowed

Highest level of education:
j Some or no high school 0 High school diploma or equivalent
P Technical college degree (A.A.) D College degree a Post-graduate degree

Approximate annual household income:
Dl No income u Under $25,000 0 $25,000- 49,999 0 $50,000 -74,999
j $75,000 99,999 o $100,000 149,999 O $150,000 or more

Number of years possessing a driver's license:

About Your Rural Freeway Driving

Typical number of rural freeway round trips made during a month?
O 1 to2 o 3 to4 E 5 to6 E 7 to 8 9 to 10 0 11 to 12 Over 12

Typical percentage of these trips made as a driver __ as a passenger__ (should sum to 100)

Typical one-way length of trip made on a rural freeway (in miles)?
C ess than16 miles n 16to30 0 31 to 45 E 46 to 60 E 61 to 75 o 76 to 100
0 101 to 125 [ 126 to 150 D 151 to 175 0 176to200 E Over200

Vehicle type most often used for rural freeway trips:
Sedan L Sports car E Pickup truck E SUV 0 Minivan
Full-size van L RV/Motorhome 0 Motorcycle D Other

Typical number of passengers in vehicle for rural freeway trips?
O 0-Driver only 0 1 E 2 o 3 o 4 or more

Typical driving style on rural freeways (on a scale from 1-5, with 1 being 'Very Conservative' and
5 being 'Very Aggressive'):

When driving alone, versus driving with passengers, does your driving style become:
E Less aggressive L Stay the same r More aggressive

























Your Opinions

Rank the overall quality of your trip (Excellent, Very Good, Good, Fair, Poor, Very Poor) for the given
roadway and traffic conditions observed in each video clip. In the space provided, list all the significant
factors/reasons that influenced your ranking of the trip quality for each video clip. After listing the factors,
please number them from most significant to least significant (with 1 being the most significant).

Video Clip Rank Comments







2



3



4



5



6



7



8



9


























10



11



12



13




In general, how would the purpose of your trip (such as business, recreational, social) affect the trip quality
rankings assigned above (e.g., higher, lower, not at all)?








If the conditions in the video clips were encountered in an urban setting, and the trip length was relatively
short, how would this affect the trip quality rankings assigned above (e.g., higher, lower, not at all)?









How would you rate this exercise in terms of its ability to give you a reasonable feel for the traffic and
roadway conditions you would experience if you were actually driving your vehicle along this roadway
under these traffic conditions?


Excellent E Very Good a Good F Fair L Poor 0 Very Poor





















APPENDIX D
SAMPLE LOOP DETECTOR DATA
























Tag County Site Lane Year Month Day Hour Min Int 15 23 28 33


SPD 18 9920 1 04 03

SPD 18 9920 2 04 03

SPD 18 9920 3 04 03

SPD 18 9920 4 04 03

SPD 18 9920 1 04 03

SPD 18 9920 2 04 03

SPD 18 9920 3 04 03

SPD 18 9920 4 04 03

SPD 18 9920 1 04 03

SPD 18 9920 2 04 03

SPD 18 9920 3 04 03

SPD 18 9920 4 04 03

SPD 18 9920 1 04 03

SPD 18 9920 2 04 03

SPD 18 9920 3 04 03

SPD 18 9920 4 04 03

SPD 18 9920 1 04 03

SPD 18 9920 2 04 03

SPD 18 9920 3 04 03

SPD 18 9920 4 04 03

SPD 18 9920 1 04 03

SPD 18 9920 2 04 03

SPD 18 9920 3 04 03

SPD 18 9920 4 04 03

SPD 18 9920 1 04 03

SPD 18 9920 2 04 03

SPD 18 9920 3 04 03

SPD 18 9920 4 04 03

SPD 18 9920 1 04 03

SPD 18 9920 2 04 03

SPD 18 9920 3 04 03

SPD 18 9920 4 04 03


0 0 0 0

0 0 0 0

0 0 0 1

0 0 0 0

0 0 0 0

0 0 0 0

0 0 0 0

0 0 0 0

0 0 0 0

0 0 0 0

0 0 0 0

0 0 0 0

0 0 0 0

0 0 0 0

0 0 0 0

0 0 0 0

0 0 0 0

0 0 0 0

0 0 0 0

0 0 0 0

0 0 0 0

0 0 0 0

0 0 0 0

0 0 0 0

0 0 0 0

0 0 0 0

0 0 0 0

0 0 0 0

0 0 0 0

0 0 0 0

0 0 0 0
o o o

o o o

o o o

o o o

o o o

o o o

o o o

o o o

o o o

o o o

o o o

o o o

o o o

o o o

o o o

o o o

o o o

o o o

o o o

o o o

o o o

o o o

o o o

o o o

o o o

o o o

o o o

o o o

o o o


38 43 48 53 58 63 68 73

0 0 0 0 2 1 3 4

0 0 0 0 0 0 0 2

0 0 0 0 0 0 0 2

0 0 0 0 1 2 6 10

0 0 0 0 1 2 3 3

0 0 0 0 0 0 0 3

0 0 0 0 0 0 0 5

0 0 0 1 0 1 5 10

0 0 0 0 0 2 2 7

0 0 0 0 0 0 0 1

0 0 0 0 0 0 1 7

1 0 0 0 2 5 8 7

0 1 0 0 0 5 3 6

0 0 0 0 0 0 4 0

0 0 0 0 0 2 3 3

0 0 0 0 0 4 4 8

0 0 0 0 0 0 5 5

0 0 0 0 0 0 0 2

0 0 0 0 0 0 0 0

0 0 0 0 0 1 7 2

0 0 2 0 0 1 2 9

0 0 0 0 0 0 0 3

0 0 0 0 0 0 1 2

0 0 0 0 0 2 5 8

0 0 0 0 3 3 3 6

0 0 0 0 0 0 0 5

0 0 0 0 0 0 0 3

0 0 0 0 0 1 7 11

0 0 0 0 0 2 6 7

0 0 0 0 0 0 0 1

0 0 0 0 0 0 0 1

0 0 0 0 2 2 3 7


Total Avg. 5 min

78 83 91 Vol. Spdi vol1

3 1 1 15 722

5 1 0 8 774 23

4 1 1 9 73 9

5 1 0 25 71 8 34

7 2 0 18 733

6 1 2 12 793 30

7 2 1 15 77 9

4 1 0 22 71 9 37

3 0 2 16 743

2 2 3 8 835 24

5 1 0 14 75 1

4 1 0 28 685 42

4 0 1 20 702

3 0 1 8 746 28

6 1 2 17 754

12 0 0 28 73 0 45

3 2 1 16 748

3 2 0 7 780 23

2 2 0 4 80 5


0 0

0 0

1 0


4 2 1

3 5 0

1 0 0

2 4 0

8 0 0

2 1 0

2 1 0

4 1 2

6 1 1


19


veh/hr/ln' Density'



138 1 86



204 282



180 238



222 299



144 1 86



252 3 56



168 235



270 3 65



138 1 82



114 155


24 144 201



29 174 232



29 174 241



36 216 291



22 132 1 82



30 180 240


1These categories were calculated from the given loop detector data and added to the speed data spreadsheets.















Total Total Total

Tag County Site Yr. Mo. Day Hour Min Int Lane # Lane # Lane # Lane # NB SB Volume

CNT 18 9920 04 03 08 00 05 005 1 15 2 8 3 9 4 25 23 34 57

CNT 18 9920 04 03 08 00 10 005 1 18 2 12 3 15 4 22 30 37 67

CNT 18 9920 04 03 08 00 15 005 1 16 2 8 3 14 4 28 24 42 66

CNT 18 9920 04 03 08 00 20 005 1 20 2 8 3 17 4 28 28 45 73

CNT 18 9920 04 03 08 00 25 005 1 16 2 7 3 4 4 15 23 19 42

CNT 18 9920 04 03 08 00 30 005 1 18 2 6 3 7 4 22 24 29 53

CNT 18 9920 04 03 08 00 35 005 1 23 2 6 3 9 4 27 29 36 65

CNT 18 9920 04 03 08 00 40 005 1 18 2 4 3 8 4 22 22 30 52

CNT 18 9920 04 03 08 00 45 005 1 13 2 4 3 12 4 28 17 40 57

CNT 18 9920 04 03 08 00 50 005 1 8 2 4 3 14 4 29 12 43 55

CNT 18 9920 04 03 08 00 55 005 1 16 2 4 3 8 4 22 20 30 50

CNT 18 9920 04 03 08 01 00 005 1 12 2 3 3 4 4 24 15 28 43

CNT 18 9920 04 03 08 01 05 005 1 19 2 3 3 8 4 19 22 27 49

CNT 18 9920 04 03 08 01 10 005 1 6 2 2 3 8 4 25 8 33 41



























CL CL CL CL CL CL CL CL CL CL CL CL CL CL CL Total
Tag County Site Lane Year Month Day Hour Min Int
01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 Vol.


18 9920

18 9920

18 9920

18 9920

18 9920

18 9920

18 9920

18 9920

18 9920

18 9920

18 9920

18 9920

18 9920

18 9920

18 9920

18 9920

18 9920

18 9920

18 9920

18 9920

18 9920

18 9920

18 9920

18 9920

18 9920

18 9920

18 9920

18 9920


Trucks
Buses'


%HV Total %
HV1
HV1


1 04

2 04

3 04

4 04

1 04

2 04

3 04

4 04

1 04

2 04

3 04

4 04

1 04

2 04

3 04

4 04

1 04

2 04

3 04

4 04

1 04

2 04

3 04

4 04

1 04

2 04

3 04

4 04


1These categories were calculated from the given loop detector data and added to the class data spreadsheets.







72



CLASSIFICATION SCHEME "F"


DESCRIPTION


NO. OF
AXLES


1 MOTORCYCLES 2
ALL CARS 2
2 CARS W/ 1-AXLE TRLR 3
CARS W/2-AXLE TRLR 4


PICK-UPS &VANS
I & 2 AXLE TRLRS


2,3. & 4


BUSES 2 & 3


2-AXLE, SINGLE UNIT 2


p 3-AXLE, SINGLE UNIT 3


B 4-AXLE, SINGLE UNIT 4


2-AXLE TRACTOR.
S-AXLE TRLR(2S1) 3

2-AXLE TRACTOR. 4
2-AXLE TRLR(2S2)

H 3-AXLE TRACTOR, 4
1 -AXLE TRLR(3S 1)

13-AXLE TRACTOR,
2-AXLE TRLR(3S2) 5


JI 3-AXLE TRUCK.
W/2-AXLE TRLR 5

TRACTOR W/ SINGLE
TRLR 6 & 7


5-AXLE MULTI- 5
TRLR

6-AXLE MULTI-
TRLR 6


7 or more


System Usage Data 1/9/90


CLASS.
GROUP


3

4

5


6


7


9




10



11



12

13


ANY 7 OR MORE AXLE















BIOGRAPHICAL SKETCH

David S. Kirschner is a 23-year old graduate student at the University of Florida.

He is studying towards his Master of Engineering degree, specializing in transportation

engineering. He received a Bachelor of Science in Civil Engineering degree from the

University of Florida in December of 2004.