WEBVTT 00:00:00.000 --> 00:00:01.000 From network to investigation. Nested synthetic review sets; no predictive classifier. 00:00:01.000 --> 00:00:07.000 MATHEMATICS · define the relationship and the meaning of each symbol before calculating. 00:00:07.000 --> 00:00:14.000 Nj: records retained at stage j. Each conditional denominator is the prior stage. Counts are 100, 20, 16 and 6. All review sets are nested. 00:00:14.000 --> 00:00:23.000 INPUT · records at each stage · read the axis units and the numerical values. 00:00:23.000 --> 00:00:25.000 Nested synthetic review sets; no predictive classifier. 00:00:25.000 --> 00:00:30.000 Screened: 100 of 100 scheduled records (100.0%). Grey records remain visible. 00:00:30.000 --> 00:00:35.000 Flagged: 20 of 100 scheduled records (20.0%). Grey records remain visible. 00:00:35.000 --> 00:00:40.000 Retained after QA: 16 of 100 scheduled records (16.0%). Grey records remain visible. 00:00:40.000 --> 00:00:45.100 Investigated: 6 of 100 scheduled records (6.0%). Grey records remain visible. 00:00:45.100 --> 00:00:54.100 OUTPUT · conditional retention · read the axis units and the numerical values. 00:00:54.100 --> 00:01:01.100 Only 6% of the original set reaches investigation. The conditional 37.5% at the last stage uses a denominator of 16, not 100. 00:01:01.100 --> 00:01:07.100 CALCULATION · substitute the defined inputs; keep units through each intermediate step. 00:01:07.100 --> 00:01:10.100 Flagged fraction: 20 / 100 = 0.20 fraction of screened 00:01:10.100 --> 00:01:13.100 QA retention: 16 / 20 = 0.80 fraction of flagged 00:01:13.100 --> 00:01:16.100 Investigation retention: 6 / 16 = 0.375 fraction of QA set 00:01:16.100 --> 00:01:19.100 Combined fraction: 0.20 × 0.80 × 0.375 = 0.06 fraction of original set 00:01:19.100 --> 00:01:28.500 SENSITIVITY · change one declared input · read the axis units and the numerical values. 00:01:28.500 --> 00:01:31.900 Investigate 4: 4 % of original 100. Four of the sixteen QA sections proceed. 00:01:31.900 --> 00:01:35.300 Investigate 6: 6 % of original 100. Six of the sixteen QA sections proceed. 00:01:35.300 --> 00:01:38.300 Investigate 8: 8 % of original 100. Eight of the sixteen QA sections proceed. 00:01:38.300 --> 00:01:44.300 Check: Verify set membership, not just decreasing totals. Confirm the six investigated identifiers occur in the sixteen-section QA set and reconcile additions, removals and deferred sections between snapshots. 00:01:44.300 --> 00:02:02.300 ONE ROW · The brightest warning on a network dashboard may also be the least corroborated. 00:02:02.300 --> 00:02:12.300 PARAMETER MAP · 12000 evaluated cells. The colour limits stay fixed across every row. 00:02:12.300 --> 00:02:22.633 Each of the 12,000 cells joins four synthetic deliveries on a route-section key. A source is present unless its declared modular missingness rule removes that row. Count the 00:02:22.633 --> 00:02:32.258 present sources and divide by four; the downloadable network exercise below contains the individual deliveries. Use the map to ask where source availability is thin before 00:02:32.258 --> 00:02:40.383 attaching a structural interpretation to a screening flag. Height shows coverage only; the route axis is an identifier, not a geographical coordinate. 00:02:40.383 --> 00:02:52.383 VALUE LANDSCAPE · Height and colour encode Available-source fraction. This is a graph over the two input axes, not automatically a physical pavement shape. 00:02:52.383 --> 00:03:00.800 This computed source fraction is not classifier confidence, probability of failure or a real evidence inventory. Source presence, independence, quality and relevance must be checked separately in a field application. 00:03:00.800 --> 00:03:06.217 Selection rates describe the process, not the probability of failure.