WEBVTT

1
00:00:00.400 --> 00:00:04.230
Two combined-instrument rows describe different kilometre ranges,

2
00:00:04.280 --> 00:00:07.030
yet repeat the same measurement pattern. That

3
00:00:07.040 --> 00:00:09.270
deserves more attention than checking whether each

4
00:00:09.280 --> 00:00:11.830
cell contains a number. A C I

5
00:00:11.840 --> 00:00:14.870
D record links road, lane, range and

6
00:00:14.960 --> 00:00:17.590
test date to roughness, rutting and texture

7
00:00:17.680 --> 00:00:20.630
measurements. Those measurements must stay attached to

8
00:00:20.640 --> 00:00:24.150
their original location and survey context. One

9
00:00:24.160 --> 00:00:27.350
implemented check finds the same road, lane,

10
00:00:27.440 --> 00:00:30.630
date and coordinate reused across different ranges.

11
00:00:31.120 --> 00:00:33.910
Compare the source rows: a copied coordinate

12
00:00:33.960 --> 00:00:36.550
can hide behind otherwise valid field types.

13
00:00:37.120 --> 00:00:40.310
Another compares the instrument-value vector across different

14
00:00:40.320 --> 00:00:42.550
ranges in the same road, lane and

15
00:00:42.560 --> 00:00:45.590
date group. At least three recognised measurement

16
00:00:45.600 --> 00:00:47.950
columns must be present for this check.

17
00:00:48.240 --> 00:00:51.030
Every included measurement must be populated and

18
00:00:51.040 --> 00:00:54.710
non-zero. The rule deliberately avoids treating blank

19
00:00:54.720 --> 00:00:57.990
or zero-filled, partially measured vectors as copied

20
00:00:58.000 --> 00:01:01.990
combined-instrument readings. Equal roughness alone is not

21
00:01:02.000 --> 00:01:04.790
enough. An identical vector is a review

22
00:01:04.800 --> 00:01:08.360
signal, not proof of fabricated data. Inspect

23
00:01:08.360 --> 00:01:11.160
the original export and survey records, correct

24
00:01:11.160 --> 00:01:14.870
the actual cause, then rerun. Preserve unmeasured

25
00:01:14.880 --> 00:01:16.310
values honestly.