WEBVTT

1
00:00:00.400 --> 00:00:02.790
An alignment report can look quiet because

2
00:00:02.800 --> 00:00:05.910
the necessary reference is missing. Start with

3
00:00:05.920 --> 00:00:08.390
what was actually available, not just the

4
00:00:08.400 --> 00:00:11.910
number of findings. For our demonstration project,

5
00:00:11.920 --> 00:00:14.310
imagine a network register and survey file

6
00:00:14.320 --> 00:00:16.710
are present, but lane references and road

7
00:00:16.720 --> 00:00:19.670
geometry are absent. Some relationships can be

8
00:00:19.800 --> 00:00:24.310
examined; others cannot. Keep aligned, warnings, errors

9
00:00:24.320 --> 00:00:27.830
and not checked distinct. A missing prerequisite

10
00:00:27.880 --> 00:00:30.150
must not turn into a successful engineering

11
00:00:30.320 --> 00:00:32.390
result, and a partial run must not

12
00:00:32.400 --> 00:00:35.720
imply complete coverage. Read the reason attached

13
00:00:35.720 --> 00:00:38.470
to the check family. Lane comparison needs

14
00:00:38.480 --> 00:00:41.670
an applicable lane reference; spatial offset needs

15
00:00:41.760 --> 00:00:44.550
usable geometry for the relevant road. A

16
00:00:44.560 --> 00:00:46.950
file being uploaded does not prove either

17
00:00:46.960 --> 00:00:50.710
dependency is usable. The selected schema, available

18
00:00:50.800 --> 00:00:54.390
fields, reference versions and implemented checks define

19
00:00:54.400 --> 00:00:57.750
the evidence scope. Certification records a review

20
00:00:57.920 --> 00:01:00.470
decision and file fingerprint; it does not

21
00:01:00.480 --> 00:01:04.470
create missing engineering evidence. When references change,

22
00:01:04.480 --> 00:01:06.870
run the relevant analysis again and review

23
00:01:06.880 --> 00:01:09.560
the new results. Keep the original scope

24
00:01:09.560 --> 00:01:12.150
and the later result traceable. That is

25
00:01:12.160 --> 00:01:14.390
how a partial check becomes an informed

26
00:01:14.560 --> 00:01:15.190
review.