1
00:00:00,209 --> 00:00:05,209
(brassy reverberating tone)
(air swooshing)

2
00:00:07,380 --> 00:00:08,213
- [Narrator] Every year,

3
00:00:08,213 --> 00:00:10,440
TxDOT assesses the pavement condition

4
00:00:10,440 --> 00:00:14,190
of nearly 200,000 lane miles of roadway.

5
00:00:14,190 --> 00:00:16,710
Pavement condition data has traditionally

6
00:00:16,710 --> 00:00:19,145
been collected through visual surveys.

7
00:00:19,145 --> 00:00:24,145
But in 2017, TxDOT began using
automated data collection,

8
00:00:24,330 --> 00:00:27,720
which greatly improved
efficiency and safety.

9
00:00:27,720 --> 00:00:29,970
But how accurate is the data?

10
00:00:29,970 --> 00:00:31,796
The Texas Department of Transportation,

11
00:00:31,796 --> 00:00:34,865
and researchers with
Texas State University

12
00:00:34,865 --> 00:00:38,760
have studied the data collected since 2017

13
00:00:38,760 --> 00:00:40,290
to provide answers.

14
00:00:40,290 --> 00:00:42,540
- For the automated
payment data collection,

15
00:00:42,540 --> 00:00:44,460
the vendor have equipment

16
00:00:44,460 --> 00:00:46,920
with the 3D laser and cameras on it,

17
00:00:46,920 --> 00:00:49,980
so they go collect data at highway speed.

18
00:00:49,980 --> 00:00:53,310
The data quality management approach

19
00:00:53,310 --> 00:00:56,474
that was designed by 20 years ago,

20
00:00:56,474 --> 00:01:00,420
really based on the visual
windshield survey approach.

21
00:01:00,420 --> 00:01:03,750
And now we switch to the
automated payment data collection,

22
00:01:03,750 --> 00:01:05,640
and we are facing different challenges.

23
00:01:05,640 --> 00:01:09,510
- We use our understanding
on the payment condition

24
00:01:09,510 --> 00:01:11,670
based on those data collection,

25
00:01:11,670 --> 00:01:13,710
and also make a decision on the

26
00:01:13,710 --> 00:01:17,040
repair and the maintenance,
and using the budget.

27
00:01:17,040 --> 00:01:20,460
So if you have a problem,

28
00:01:20,460 --> 00:01:23,580
accuracy or precision
on the data collection,

29
00:01:23,580 --> 00:01:26,310
then, you could make the wrong decisions.

30
00:01:26,310 --> 00:01:29,949
- The quality or issues
from the data are different

31
00:01:29,949 --> 00:01:32,190
from the old visual days.

32
00:01:32,190 --> 00:01:35,280
So that's why we need to
have a different approach

33
00:01:35,280 --> 00:01:37,680
to audit the data,

34
00:01:37,680 --> 00:01:39,960
and to be able to identify the sections,

35
00:01:39,960 --> 00:01:42,720
has a potential data quality issues.

36
00:01:42,720 --> 00:01:47,720
- In our research, we
actually look at other states,

37
00:01:47,862 --> 00:01:51,360
we had a survey and also
a literature review.

38
00:01:51,360 --> 00:01:53,700
We reached out to all the 50 states,

39
00:01:53,700 --> 00:01:57,420
and we also developed data
analysis of procedures

40
00:01:57,420 --> 00:02:00,900
for how text data can check the quality,

41
00:02:00,900 --> 00:02:02,460
and find out the problems.

42
00:02:02,460 --> 00:02:05,550
- This research team look at mainly

43
00:02:05,550 --> 00:02:08,610
the data collected in 25 districts,

44
00:02:08,610 --> 00:02:12,300
and then they evaluated the
data in the last four years

45
00:02:12,300 --> 00:02:16,230
to identify the issues in the data,

46
00:02:16,230 --> 00:02:21,030
and then they recommend this
stratified sampling approach,

47
00:02:21,030 --> 00:02:22,950
and develop the indexes.

48
00:02:22,950 --> 00:02:25,770
- We use San Antonio
District as a case study

49
00:02:25,770 --> 00:02:29,310
to implement our research results to see,

50
00:02:29,310 --> 00:02:31,050
you know, if this is working.

51
00:02:31,050 --> 00:02:34,410
And our research results
seems like they're working.

52
00:02:34,410 --> 00:02:37,710
- We are planning to
implement the results,

53
00:02:37,710 --> 00:02:40,890
and then we are gonna need
to do a little bit fine tune

54
00:02:40,890 --> 00:02:43,721
in term of how to
incorporate into the system.

55
00:02:43,721 --> 00:02:47,610
So there's a little bit
more work for us to do,

56
00:02:47,610 --> 00:02:51,150
but it's a great research,
approach and the findings.

57
00:02:51,150 --> 00:02:53,220
- How to use the money more effective

58
00:02:53,220 --> 00:02:56,040
and more efficient is our concern.

59
00:02:56,040 --> 00:02:58,650
And also the general public's concern.

60
00:02:58,650 --> 00:03:01,890
And if we can use our data,

61
00:03:01,890 --> 00:03:05,550
I mean research results to
improve the data quality,

62
00:03:05,550 --> 00:03:10,550
and help TxDOT engineers
to make informed decisions,

63
00:03:11,220 --> 00:03:13,170
so that way we can have a hope

64
00:03:13,170 --> 00:03:16,440
to use the money more effectively,

65
00:03:16,440 --> 00:03:18,630
and that also can contribute

66
00:03:18,630 --> 00:03:20,790
to a better condition of the roads.

67
00:03:20,790 --> 00:03:22,050
- [Narrator] For more information,

68
00:03:22,050 --> 00:03:25,080
and to find the publications
for this project,

69
00:03:25,080 --> 00:03:27,720
please visit the TxDOT research library

70
00:03:27,720 --> 00:03:29,463
at the link shown below.