WEBVTT

1
00:00:00.400 --> 00:00:02.950
A survey can have perfectly formed columns

2
00:00:02.960 --> 00:00:05.910
and still describe the wrong road. Engineering

3
00:00:06.160 --> 00:00:08.470
alignment asks whether each record belongs to

4
00:00:08.480 --> 00:00:12.230
the project network, classification and lane references.

5
00:00:12.480 --> 00:00:15.270
Start with a demonstration road, DEMO zero

6
00:00:15.280 --> 00:00:18.310
one. The engine resolves its road identifier

7
00:00:18.440 --> 00:00:22.070
against the reference index. Normalisation handles case

8
00:00:22.080 --> 00:00:25.510
and whitespace; ambiguous candidates are not silently

9
00:00:25.520 --> 00:00:29.190
selected. Next, compare the submitted kilometre range

10
00:00:29.200 --> 00:00:31.430
with the road extent. A segment from

11
00:00:31.440 --> 00:00:33.270
nine point eight to ten point two

12
00:00:33.440 --> 00:00:35.910
cannot fit entirely within a network ending

13
00:00:35.920 --> 00:00:38.710
at ten kilometres. The lane register adds

14
00:00:38.840 --> 00:00:41.830
another relationship: does the submitted lane exist

15
00:00:41.840 --> 00:00:44.630
for that road and location? A valid-looking

16
00:00:44.680 --> 00:00:47.190
lane code alone cannot answer that question.

17
00:00:47.680 --> 00:00:50.920
Reference strength matters. A network reference and

18
00:00:50.920 --> 00:00:53.190
an inferred road extent do not carry

19
00:00:53.320 --> 00:00:56.950
identical evidence. Missing references leave affected checks

20
00:00:57.040 --> 00:01:00.150
not checked, rather than proving alignment. Read

21
00:01:00.160 --> 00:01:02.870
the finding, source reference and proposed action

22
00:01:02.880 --> 00:01:06.150
together. Correct the source or reference deliberately,

23
00:01:06.160 --> 00:01:08.150
then run the check again. Keep the

24
00:01:08.280 --> 00:01:10.630
version used for that review traceable.