Overview
The daily read: what is bleeding, what is verified, what to say first.
Built Wednesday, Jun 24
$661,941

This uncoded loss is $661,941, and at 26.70% of total reject dollars it's more than a quarter of everything you're losing. It's spread across 273 events, so it's not one bad ticket, it's a habit at the reject station.

Share of all reject dollars26.7%
273events logged as "No Code Selected"
$8.61Mannualized, if the habit holds
First move

Enforce defect coding at the reject station: no reject gets logged without a code, so next month this money shows up as a named, fixable line instead of a black hole.

Reject cost · 4 weeks
$2.48M
scrap $1.90M · rework $581K
Verified queue
4 of 8 proposed
3 consolidated · 1 held for review
Biggest defect family
Polishing - Weld Seam
$115K coded loss this window
Trap avoided
KW 4.5 TRACER TIP,PAINT
#2 by dollars, healing -68%, not chased

Today's verified opportunities

tap one to open the full case
Python computes every number→ AI maker explains→ Independent AI checker verifies→ Deterministic cross-check AI never does the math. Every claim cites its numbers.
1

7X60,MIT,ST(OD) CHR POL

Material - Bad Pin Stamp
$101,940

Bad pin stamp on 7X60 is spiking, and it's mostly true scrap, so the money is gone.

First move

This looks upstream: pull incoming inspection and the supplier scorecard for the part before you touch the line.

Why it matters

Running $101,940 recent-weighted with the trend climbing +73%, the steepest on the board. The defect is Material - Bad Pin Stamp and it's mostly true scrap, at $2,464 every event.

Why this confidence

Four weeks of data, a hard +73% climb, and a mostly-true-scrap signal, no spike noise, so the data earns High.

Cited numbers

$101,940 recent-weightedtrend +73%$2,464/eventMaterial - Bad Pin Stampmostly true scrap4 weeks present

Independent checker

Steepest climb on the board (+73.5%) across 4 weeks with no spike, mostly true scrap at $2,464/event, and no known-work match on Material - Bad Pin Stamp.

Real vs noise

pass: 4 weeks present, is_spike False, trend +73.5%, $2,464/event

Known work

clear: no initiative or tracker part matches Material - Bad Pin Stamp

Over-claim audit

clean on every figure ($101,940, +73%, $2,464, mostly true scrap, 4 weeks); minor wording only: headline says 'spiking' though is_spike is False, and the body correctly calls it a 4-week climb with no spike noise

2

TUNGSTEN SPROCKET DEBUR

Plating - Rectifier Failure
$78,076

Rectifier failure on the sprocket line is climbing +67%, a plating problem, not a polish one.

First move

Pull True Cam tank parameters for the line and check rack contact and agitation on the top part numbers.

Why it matters

This part is running $78,076 recent-weighted with the trend climbing +67%. The dominant defect is Plating - Rectifier Failure, costing $2,408 every event.

Why this confidence

Four weeks of data with a strong +67% climb and no spike, a stable, steep trend the data can stand behind.

Cited numbers

$78,076 recent-weightedtrend +67%$2,408/eventPlating - Rectifier Failure4 weeks present

Independent checker

+67.2% across 4 weeks with no spike on Plating - Rectifier Failure, a defect family with no open, held, or dismissed initiative in the tracker.

Real vs noise

pass: 4 weeks present, is_spike False, trend +67.2%, $2,408/event

Known work

clear: no plating initiative exists in known_projects

Over-claim audit

numbers all match ($78,076, +67%, $2,408, 4 weeks); minor: 'the sprocket line' is inferred from the part name since the data carries no line field, and the first step is the engine's canned plating step, not a maker invention

3

4" PIPE-CRV STACK 439SST NiCR

Polishing - Tool Marks
$71,885

Tool marks on the 4" pipe-crv stack are climbing +49%.

First move

Run a cycle-time and fixture check on the polishing cell and confirm operator ramp on that line.

Why it matters

Running $71,885 recent-weighted with the trend climbing +49%. The dominant defect here is Polishing - Tool Marks, and it's costing $2,575 every event, the highest per-event hit on the board.

Why this confidence

Four weeks of data and a +49% climb, no spike, but a mixed scrap/rework picture keeps it Medium.

Cited numbers

$71,885 recent-weightedtrend +49%$2,575/eventPolishing - Tool Marks4 weeks present

Independent checker

+49.0% across 4 weeks with no spike, and Polishing - Tool Marks is a different defect family from PRJ-114's weld-seam scope, so this is novel work.

Real vs noise

pass: 4 weeks present, is_spike False, trend +49.0%, $2,575/event

Known work

clear: PRJ-114 covers Polishing - Weld Seam only, and no initiative exists for Polishing - Tool Marks

Over-claim audit

clean: $71,885, +49%, $2,575/event match, and the 'highest per-event hit on the board' claim checks out against all eight engine cost_per_event values

4

SCREW,FLANGE,HEX HEAD

Polishing - Out of Round
$89,964

Out-of-round on the flange screw is flat but expensive, mostly true scrap.

First move

Run a cycle-time and fixture check on the polishing cell and confirm operator ramp, this one is a standing cost to knock down, not a fire.

Why it matters

This part is running $89,964 recent-weighted, but the trend is essentially flat at +4%, so it's steady dollars rather than a growing problem. The defect is Polishing - Out of Round, it's mostly true scrap, and it costs $2,527 every event.

Why this confidence

Four weeks of data and mostly true scrap, but a flat +4% trend gives the data less to lean on, so Medium.

Cited numbers

$89,964 recent-weightedtrend +4%$2,527/eventPolishing - Out of Roundmostly true scrap4 weeks present

Independent checker

Flat at +3.7% but heavily true scrap (engine chip 'Mostly true scrap', scrap $163,184 of $197,084 raw) at $89,964 recent-weighted, which is the rubric's true-scrap path to confirmation, and the maker honestly framed it as a standing cost, not a fire.

Real vs noise

concern: trend is flat at +3.7%, so it is not worsening; it passes on the heavy-true-scrap basis with 4 weeks present, is_spike False, and $2,527/event

Known work

clear: no initiative for Polishing - Out of Round and the part is not in the tracker

Over-claim audit

clean: maker discloses the flat trend itself ('steady dollars rather than a growing problem'), and $89,964, +4%, $2,527/event all match the engine

What the checker removed today, and why
5" PIPE-CRV STACK 439 SST NiCR

Dominant defect Polishing - Weld Seam is already owned by open initiative PRJ-114, whose scope explicitly rolls up any weld-seam part; attach it there instead of opening a duplicate ticket.

PRJ-114
10G CARBON STEEL 120" X 30"

Second Polishing - Weld Seam part, already owned by open initiative PRJ-114; the maker itself flags 'same cell as rank 1', which is the consolidation signal, not grounds for a new ticket.

PRJ-114
6X108,MIT,ST(OD) CHR POL

Third Polishing - Weld Seam part, already owned by open initiative PRJ-114; attach to the active kaizen rather than opening a duplicate.

PRJ-114
RETAINING RING

Exact part and defect match to dismissed PRJ-097 (tank filtration + fixture change, closed 2026-04-29) with the trend now climbing again (+24.6%), which per the tracker note requires human re-review by process engineering, not an automatic re-open.

re-review

Pareto: where the coded loss lives

No single villain: the top two families carry only 11% and it takes 8 families to reach half the coded dollars. A spread this even points at process discipline, not one bad machine. Red line = cumulative share.

$29K$58K$87K$115KPolishing - Weld Seam: $115,488Weld Seam$115KPolishing - Tool Marks: $89,212Tool Marks$89KPlating - Rack Mark: $87,337Rack Mark$87KPlating - Wrong Process: $78,720Wrong Process$79KPolishing - Out of Round: $78,092Out of Round$78KPlating - Burn: $63,830Burn$64KPlating - Contamination: $60,218Contamination$60KAll other families: $1,244,090All other$1.24Mcumulative 6%6%cumulative 11%11%cumulative 16%cumulative 20%cumulative 25%cumulative 28%cumulative 32%cumulative 100%100%

Week by week: scrap vs rework

Worst week is Jun 8 at $753K. True scrap is 77% of every week's loss.

$376K$753KJun 1 scrap: $519,809Jun 1 rework: $194,219$714KJun 1Jun 8 scrap: $596,697Jun 8 rework: $156,003$753KJun 8Jun 15 scrap: $561,943Jun 15 rework: $169,211$731KJun 15Jun 22 scrap: $219,060Jun 22 rework: $61,986$281KJun 22
True scrap (gone) Rework (recoverable)

The blind spot: coded vs uncoded

$662K has no defect code. Fix the coding discipline and the Pareto gets sharper next month.

26.7%no defect code
Uncoded: $661,941 (26.7%)
Coded: $1,816,987
273 uncoded events in 4 weeks

Top parts by recent pull

The four queue parts pull $407K of the recent window. Green tag = survived the checker.

5" PIPE-CRV STACK 439 SST NiCR $131,314
7X60,MIT,ST(OD) CHR POLin queue $101,940
TUNGSTEN SPROCKET DEBURin queue $78,076
10G CARBON STEEL 120" X 30" $95,390
6X108,MIT,ST(OD) CHR POL $75,621
4" PIPE-CRV STACK 439SST NiCRin queue $71,885
SCREW,FLANGE,HEX HEADin queue $89,964
RETAINING RING $80,280
Rank the board by and you chase what already hurt. Rank it by (recent pull × direction of travel × cost intensity, with a ) and the list reorders itself. Click any row to see exactly why it ranks where it does.
By raw dollars · the gut list
By evidence · what to chase

Hover a row to trace where it moves. The emerald line is TUNGSTEN SPROCKET DEBUR jumping #10 → #3; the rose row is KW 4.5 TRACER TIP,PAINT healing on its own and dropping off the chase list.