Workflow template for operations and process improvement
Scrap and trim loss review
Each week or month the plant manager gets one review of the material recorded as scrap and trim, by product, machine, shift and reason, with a balance for each completed order, every place the records disagree, and improvement questions. It states no cause the records do not show.
Listed under
Add it to a workflow in the Fully Powered Planner, beside the teams and roles that will run it. Book a fit call
The outcome
Scrap records pile up all period, one ticket at a time, and nobody adds them up until someone asks why material usage is high. This workflow template turns them into one review the plant or operations manager can take into a production meeting. It shows how much material was recorded as scrap and how much as trim, set against the material that went into good parts, broken down by product, by machine, by shift, by machine and shift together, and by the reason recorded on each entry.
It also checks the records against each other. A run where rejected parts were counted but nothing was booked as scrap, a scrap entry for an order that was not on that machine that shift, a completed order where the material issued does not add up to good parts plus recorded loss: each is listed with the row references and put as a question to the person who keeps that record.
The improvement questions come from what the figures show and nothing else: the reason that accounts for most of the scrap or trim, and a gap between shifts on the same machine, given with the product mix each shift ran. The review names no cause and judges no crew. The plant manager and production decide what is worth investigating or trialling, and nothing is changed or sent.
Who starts the work, and when
The plant manager or operations manager starts it at the end of each week or month, once the period's scrap records, production counts and material issues can be exported. It is run before the regular waste or production review, and again when late or corrected records arrive. The production supervisor answers questions about the scrap records and production counts, the materials controller about issues and returns, and the quality lead about reason codes.
What they supply
- The scrap records for the period, as a spreadsheet export: one row per scrap or trim entry, with its date, shift, machine, product, order, material, whether it is scrap or trim, the reason code, the quantity and its unit.
- The production counts: good and rejected units for each order, date, shift and machine.
- The material issues: what was issued to each order and what came back.
- The material per good unit from the material specifications: how much material ends up in one finished part.
- The reason code list, with each code marked as scrap or trim.
- A short settings note: the plant, the period, the unit of measure, which orders were completed, and the thresholds that decide when a figure becomes a question.
The template states the default column headers for every export and stops to ask when one does not match.
What comes out
One review in a fixed order: the period's recorded loss and rate; the same figures by product, machine, shift, and machine and shift together; scrap and trim by recorded reason; a material balance for each completed order; where the records disagree; rows that were not counted, have no reason, or fall outside the period; and the improvement questions. Every table adds back to the same total, and an empty section says "None." The worked example uses an invented stamping plant and shows the full inputs as well as the result.
An invented example
Illustrative example with invented data. The plant, products, orders, codes and every figure below are made up for this page. They stand for a small metal-stamping plant reviewing one week in November 2026, and describe nothing that happened anywhere.
What the plant manager supplies
Scrap records (default headers; quantities in kg):
scrap_ref,date,shift,machine,product,order,material,scrap_type,reason_code,scrap_qty,unit,note
S-01,2026-11-02,A,Press 1,BR-40,WO-501,CR-15,trim,SKL,360,kg,
S-02,2026-11-02,A,Press 1,BR-40,WO-501,CR-15,scrap,SET,24,kg,first-off parts
S-03,2026-11-02,B,Press 1,BR-40,WO-501,CR-15,trim,SKL,330,kg,
S-04,2026-11-02,B,Press 1,BR-40,WO-501,CR-15,scrap,DIM,48,kg,oversize after die change
S-05,2026-11-03,A,Press 1,BR-40,WO-501,CR-15,trim,SKL,220,kg,
S-06,2026-11-03,A,Press 1,BR-40,WO-501,CR-15,trim,END,40,kg,coil tail
S-07,2026-11-03,A,Press 2,PN-60,WO-502,CR-15,trim,SKL,300,kg,
S-08,2026-11-03,A,Press 2,PN-60,WO-502,CR-15,scrap,SUR,25,kg,
S-09,2026-11-03,B,Press 2,PN-60,WO-502,CR-15,trim,SKL,290,kg,
S-10,2026-11-03,B,Press 2,PN-60,WO-502,CR-15,scrap,,62,kg,
S-11,2026-11-04,A,Press 2,PN-60,WO-502,CR-15,trim,SKL,210,kg,
S-12,2026-11-05,A,Press 2,BR-40,WO-503,CR-15,trim,SKL,280,kg,
S-13,2026-11-05,A,Press 2,BR-40,WO-503,CR-15,scrap,DIM,16,kg,
S-14,2026-11-05,B,Press 2,BR-40,WO-503,CR-15,trim,SKL,200,kg,
S-15,2026-11-05,B,Press 2,PN-60,WO-502,CR-15,scrap,SET,35,kg,
S-16,2026-11-05,A,Press 2,BR-40,WO-503,CR-15,scrap,DIM,12,pcs,
S-17,2026-10-30,B,Press 1,BR-40,WO-499,CR-15,trim,SKL,150,kg,
Production counts (default headers; good units counted at the press, which finishes each part):
order,date,shift,machine,product,good_qty,rejected_qty
WO-501,2026-11-02,A,Press 1,BR-40,2000,30
WO-501,2026-11-02,B,Press 1,BR-40,1800,60
WO-501,2026-11-03,A,Press 1,BR-40,1200,0
WO-502,2026-11-03,A,Press 2,PN-60,800,10
WO-502,2026-11-03,B,Press 2,PN-60,700,25
WO-502,2026-11-04,A,Press 2,PN-60,500,0
WO-503,2026-11-05,A,Press 2,BR-40,1500,20
WO-503,2026-11-05,B,Press 2,BR-40,1000,15
Material issues (default headers):
issue_ref,date,order,material,issued_qty,returned_qty,unit
I-01,2026-11-02,WO-501,CR-15,4500,0,kg
I-02,2026-11-03,WO-501,CR-15,600,30,kg
I-03,2026-11-03,WO-502,CR-15,6200,0,kg
I-04,2026-11-05,WO-503,CR-15,4000,0,kg
Material per good unit, from the plant's specification sheet for steel coil CR-15 (the steel left in one finished part, not the steel drawn for it):
product,material,qty_per_good_unit,unit
BR-40,CR-15,0.80,kg
PN-60,CR-15,2.50,kg
Reason code list:
code,description,type
SET,setup and first-off parts,scrap
DIM,out of dimension,scrap
SUR,surface damage,scrap
SKL,blanking skeleton,trim
END,coil head and tail ends,trim
Settings: Plant C (invented); for the plant manager; period 2026-11-02 to 2026-11-06; unit kg; completed orders WO-501 and WO-502 (WO-503 is still running); prepared 2026-11-09; balance tolerance 2 percent; reason share 20 percent; shift gap 2.0 points. Questions on scrap records, production counts and improvement go to the production supervisor, on issues to the materials controller, and on reason codes to the quality lead.
What comes out
SCRAP AND TRIM LOSS REVIEW: Plant C, 2026-11-02 to 2026-11-06
For: the plant manager. Prepared 2026-11-09 from 17 scrap rows, 8 production rows and 4 material issue rows. Quantities in kg.
Scrap rows: 15 counted, 1 not counted, 1 left out of the period.
Recorded loss rate = recorded loss as a share of good-unit material plus recorded loss.
The review shows what the records say and states no cause.
Defaults used: none.
1. RECORDED LOSS
Good units 9,500; good-unit material 11,000. Scrap 210. Trim 2,230. Recorded loss 2,440. Rate 18.2%.
2. BY PRODUCT
BR-40: 7,500 good units, good-unit material 6,000; scrap 88, trim 1,430, recorded loss 1,518; rate 20.2%.
PN-60: 2,000 good units, good-unit material 5,000; scrap 122, trim 800, recorded loss 922; rate 15.6%.
3. BY MACHINE
Press 1: good-unit material 4,000; scrap 72, trim 950, recorded loss 1,022; rate 20.4%.
Press 2: good-unit material 7,000; scrap 138, trim 1,280, recorded loss 1,418; rate 16.8%.
4. BY SHIFT
A: good-unit material 7,010; scrap 65, trim 1,410, recorded loss 1,475; rate 17.4%.
B: good-unit material 3,990; scrap 145, trim 820, recorded loss 965; rate 19.5%.
(Figures by shift describe the records of a time slot, not the people on it.)
5. BY MACHINE AND SHIFT
Press 1, shift A: good-unit material 2,560; scrap 24, trim 620, recorded loss 644; rate 20.1%. Ran: BR-40 3,200 good units.
Press 1, shift B: good-unit material 1,440; scrap 48, trim 330, recorded loss 378; rate 20.8%. Ran: BR-40 1,800 good units.
Press 2, shift A: good-unit material 4,450; scrap 41, trim 790, recorded loss 831; rate 15.7%. Ran: BR-40 1,500 good units, PN-60 1,300 good units.
Press 2, shift B: good-unit material 2,550; scrap 97, trim 490, recorded loss 587; rate 18.7%. Ran: BR-40 1,000 good units, PN-60 700 good units.
6. BY RECORDED REASON
Reason code list: 5 codes supplied.
Scrap, 210 counted:
DIM out of dimension: 64, 30.5%.
SET setup and first-off parts: 59, 28.1%.
SUR surface damage: 25, 11.9%.
Unclassified: 62, 29.5% (1 row).
Trim, 2,230 counted:
SKL blanking skeleton: 2,190, 98.2%.
END coil head and tail ends: 40, 1.8%.
Unclassified: none.
7. MATERIAL BALANCE FOR COMPLETED ORDERS
WO-501, BR-40, CR-15: issued 5,100 less returned 30 = net 5,070. Good-unit material 4,000 (5,000 good units x 0.80). Scrap 72. Trim 950. Not accounted for 48, 0.9% of net issued; within the 2% tolerance.
WO-502, PN-60, CR-15: issued 6,200 less returned 0 = net 6,200. Good-unit material 5,000 (2,000 good units x 2.50). Scrap 122. Trim 800. Not accounted for 278, 4.5% of net issued; at or above the 2% tolerance, see section 8, item 3. Uses S-10, questioned in section 9, and S-15, questioned in section 8, item 2.
WO-503: not computed, not named as complete.
8. WHERE THE RECORDS DISAGREE
1. Rejected units with no scrap recorded: the production count for WO-503 on 2026-11-05, shift B, Press 2, shows 15 rejected units, and no scrap row of type scrap names WO-503 on that date, shift and machine (the only scrap row there for WO-503 is S-14, trim). Question for the production supervisor: what happened to the 15 rejected BR-40 units, and is there a scrap record for them?
2. Scrap with no matching production: S-15 (2026-11-05, shift B, Press 2, WO-502, PN-60, scrap, SET, 35) names an order that the production counts do not show on that date, shift and machine; they show WO-503, BR-40, on Press 2 that shift. S-15 is counted under PN-60, Press 2 and shift B, as recorded. Question for the production supervisor: which order and product does S-15 belong to?
3. Material balance: WO-502 has 278 of its net 6,200 issued not accounted for as good-unit material, scrap or trim, 4.5% against a 2% tolerance (I-03; S-07 to S-11 and S-15, of which S-10 is questioned in section 9 and S-15 in item 2; three production rows). Question for the materials controller: are all issues and returns for WO-502 in the export? Question for the production supervisor: is any scrap or trim from WO-502 not yet recorded?
9. ROWS NOT COUNTED, UNCLASSIFIED OR LEFT OUT
Not counted: S-16, unit pcs where the settings say kg (2026-11-05, shift A, Press 2, WO-503, scrap, DIM, 12 pcs). Question for the production supervisor: what weight of material do the 12 pieces on S-16 come to?
Unclassified: S-10, scrap, 62, code blank. Question for the production supervisor: which reason on the list applies to S-10?
Not recorded: None.
Left out of the period: S-17, 2026-10-30, 150.
10. IMPROVEMENT QUESTIONS FOR THE PRODUCTION SUPERVISOR
1. DIM (out of dimension) is the largest listed scrap reason: 64, 30.5% of the 210 scrap counted. Recorded on Press 1 for BR-40 (48, S-04) and on Press 2 for BR-40 (16, S-13). Recorded note on S-04: "oversize after die change". Unclassified scrap (62, S-10, section 9) is larger than DIM's lead of 5 over SET, so the order could change once S-10 is classified. What does production know about the out-of-dimension scrap on BR-40, and is it worth investigating or trialling?
2. SKL (blanking skeleton) is the largest listed trim reason: 2,190, 98.2% of the 2,230 trim counted. Recorded on Press 1 for BR-40 (910), on Press 2 for BR-40 (480) and on Press 2 for PN-60 (800). What does production know about the skeleton loss on each product, and is it worth investigating or trialling?
3. Press 2: shift B rate 18.7% against shift A 15.7%, a gap of 3.0 points against a threshold of 2.0. Shift A ran BR-40 1,500 good units and PN-60 1,300; shift B ran BR-40 1,000 and PN-60 700. Shift B's figures include S-10 (section 9) and S-15 (section 8, item 2), and its good units include the WO-503 run in section 8, item 1. What does production know about the difference between the two shifts on Press 2, and is it worth investigating or trialling?
Every figure traces to the scrap records (2026-11-02 to 2026-11-06, with S-17 from the week before), the production counts (2026-11-02 to 2026-11-05), the material issues (2026-11-02 to 2026-11-05) and the material per good unit (specification sheet, CR-15). Draft for the plant manager. No record has been changed and nothing has been sent.
How to read it
The tables say where the steel went, and they agree with each other: products, presses, shifts, and presses by shift all add to the same 2,440 kg recorded loss against 11,000 kg in good parts. Most of it is trim, the skeleton left after blanking, so one improvement question is about skeleton loss on each product and not about rejected parts.
Three things in the records do not line up, and each is put to the person who keeps that record rather than settled on the page. The 15 rejected BR-40 parts on shift B of 5 November have no scrap entry. One setup entry names an order that was not on the press that shift. And 278 kg of the steel issued to WO-502 does not appear as good parts, scrap or trim. The Press 2 shift gap is shown with the product mix beside it and with the two questioned rows named, so nobody reads it as a verdict on a crew. The review gives no reason for any of it; the plant manager and production decide what is worth a closer look.
Where it fits
- Manufacturing, where scrap tickets, production counts and stores issues are kept by different people and rarely read side by side.
- Hardware and physical technology, for product makers who cut, form or mould their own parts and want to see where material goes.
- Consumer brands, for brands that run their own production and watch material loss as closely as sales.
- Operations and process improvement, which runs the production floor and decides where to look for waste.
- Quality management, which owns the reason codes and investigates once a pattern is worth chasing.
- Established businesses, where years of scrap records exist and are seldom added up.
- Growth and scaling, when more shifts and machines make the waste harder to see from the floor.
It sits beside two related workflow templates and does not replace them. Cost of poor quality worksheet puts a money value on failures; this review counts material and never costs it. Production cost variance sheet explains cost differences from standard for finance; this one shows where material was recorded as lost, for the people running the floor.
What has to change for your company
Map your own export columns to the default headers. Agree the reason code list with quality and decide which codes are scrap and which are trim, then keep the list stable so one period can be read against the next. Confirm that your material figure per part is what ends up in the finished part, not the gross amount drawn, or trim will be counted twice. Set the three thresholds: how far a completed order's material balance may be out before it is raised, how large a share one reason must reach, and how wide a gap between shifts on one machine must be before it becomes a question. Say who answers questions about scrap records, issues and reason codes. If parts pass several machines, decide which step's count is the finished count. It runs in ChatGPT, Claude, Copilot or Gemini from the exports you supply and needs no access to your systems.
What has been checked
Fully Powered AI wrote this workflow template. It was editorially reviewed on 12 November 2026. It was tried once, with Claude, on its invented example on 12 November 2026; no stop condition, request for a cause or a culprit, request to cost the loss, or note naming a person was exercised in that run, and the record is published with it. It has not yet been tested in a client environment. It has not yet been tested as a packaged version in ChatGPT, Claude, Copilot or Gemini.
Testing in your environment means running it on a past month your plant manager already knows, with cases written in advance: a scrap entry recorded in pieces instead of weight, a run with rejected parts and no scrap entry, a completed order whose issues do not balance, a blank reason code. The plant manager recomputes two groups and one balance by hand before anyone relies on the review. Pilot acceptance-test preparation builds that test set, and what we test before we call an AI workflow ready shows the record we keep.
Who decides
Your people keep
- What is worth investigating or trialling
- Which record is right when the records disagree
- Which reason applies to an unclassified entry
It does not
- Name a cause
- Judge a crew
- Cost the material
- Change a record or send it
Questions people ask
Who is it for?
The plant manager or operations manager, at the end of each week or month, once the period's scrap records, production counts and material issues can be exported.
What do I need to supply?
The scrap records, the production counts, the material issues, the material per good unit from the material specifications, the reason code list and a short settings note.
Does it say why scrap happened?
No. The review names no cause and judges no crew. The plant manager and production decide what is worth investigating or trialling.
Does it put a money value on scrap?
No. This review counts material and never costs it. The cost of poor quality worksheet puts a money value on failures.
What has been checked?
Fully Powered AI wrote it and it was editorially reviewed. It was tried once, with Claude, on its invented example on 12 November 2026, and the record is published with it. It has not yet been tested in a client environment.
The full workflow template
Everything above is the outline. The complete workflow template is the working version: the roles, the inputs, the procedure, the exceptions and stop conditions, the output format, the reviewer checklist, the acceptance tests, the instructions for an AI assistant and what to customize for your company. It comes with a download and the link to its folder in Fully Powered’s public repository.
Free with an account. Sign in by email: no password and no card.
Next step
Put this workflow template to work for your team: on your plan, or built with you in a sprint.
Build this in a sprint
In a sprint, Fully Powered adapts this workflow template to your rules, proves it on your own work and puts it in the hands of the roles that own it.
Two ways to begin: