Manufacturing Loss Investigation
- Context
- Beverage manufacturing — PET preform production (organisation anonymised)
- Status
- Delivered
- Tools
- Excel dashboard, Production data analysis, Fishbone (Ishikawa) diagrams, Pareto analysis
Executive summary
An Industrial Engineering investigation into PET preform material losses in a manufacturing operation. Using production data and root-cause analysis, I quantified approximately US$10,000 in identified material losses and linked them to their causes, and delivered the findings as an Excel dashboard report that gave the operation a prioritised basis for corrective action.
~US$10,000
Material losses identified
PET preform material losses quantified during the investigation
How to read this result. This figure is the value of losses identified. It is not recovered savings.
Business context
PET preforms are the moulded tubes later blown into bottles. Material that is scrapped, rejected or otherwise lost during production is a direct cost, and in high-volume manufacturing small loss rates accumulate quickly.
Challenge
The operation knew material was being lost but did not have a clear account of how much, where in the process, or why. Without that, corrective effort could not be prioritised.
My contribution
- Gathered and organised production data relevant to material usage and losses.
- Quantified the losses and expressed them in monetary terms.
- Applied root-cause analysis to link losses to underlying causes.
- Built the Excel dashboard report that presented the losses, their causes and the recommendations.
- Prepared recommendations for addressing the main causes.
Investigation and methodology
The investigation followed a standard Industrial Engineering sequence: define the loss, measure it, then analyse causes. Losses were categorised so the largest contributors could be separated from minor ones (Pareto analysis), and cause-and-effect diagrams were used to trace each major category back through people, methods, machines and materials.
Solution design and implementation
The findings were delivered as an Excel dashboard report: an evidence-based account of material losses, a cause structure, and recommendations ranked by their share of the identified loss.
Results and business value
The investigation identified approximately US$10,000 in PET preform material losses. This is the value of losses identified, not money recovered. Its business value is in making a hidden cost visible and giving management a prioritised starting point for corrective action.
Technical details
- Quantification: material loss volumes converted to monetary value to make categories comparable.
- Prioritisation: Pareto ordering of loss categories.
- Cause analysis: Ishikawa (fishbone) structure for the major categories.
Lessons, constraints and improvement opportunities
- Putting a currency value on loss changes the conversation; it moves the issue from operations to management attention.
- Identified loss is not saved cost — savings depend on implementation and should be measured separately.
- Ongoing loss tracking per shift and per cause would show whether corrective actions hold.
- Root-cause analysis
- Industrial Engineering
- Loss quantification
- Operational analysis
Interactive demonstration
Explore a synthetic version
Built independently with generated data and illustrative business rules, to show the analytical approach without exposing any client system or record.
Illustrative demonstration using synthetic data. Not an original client system. Company, store and people names are fictional (Delta Polymers).
PET preform material-loss investigation
Weekly material losses across three fictional production lines, with cause categories, the relationship to downtime, and a root-cause structure that follows your filters.
- Material lost
- 33,243 kg
- Identified loss value (synthetic)
- US$48.2k
- Identified, not recovered
- Loss rate
- 0.84%
- of material processed
- Downtime
- 360 h
- 1,278.6 kg lost per week on average
Show data table
| Week | Loss (kg) |
|---|---|
| Week 1 | 1,565 |
| Week 2 | 1,135 |
| Week 3 | 1,560 |
| Week 4 | 1,392 |
| Week 5 | 1,228 |
| Week 6 | 1,097 |
| Week 7 | 1,363 |
| Week 8 | 1,150 |
| Week 9 | 1,333 |
| Week 10 | 1,136 |
| Week 11 | 1,432 |
| Week 12 | 1,198 |
| Week 13 | 1,356 |
| Week 14 | 1,206 |
| Week 15 | 1,080 |
| Week 16 | 1,400 |
| Week 17 | 1,001 |
| Week 18 | 1,179 |
| Week 19 | 1,408 |
| Week 20 | 1,290 |
| Week 21 | 1,275 |
| Week 22 | 1,339 |
| Week 23 | 1,186 |
| Week 24 | 1,282 |
| Week 25 | 1,297 |
| Week 26 | 1,354 |
Show data table
| Cause | Loss (kg) | Cumulative share |
|---|---|---|
| Start-up scrap | 12,170 | 37% |
| Colour changeover purge | 9,556 | 65% |
| Short shots and voids | 6,439 | 85% |
| Contamination | 3,085 | 94% |
| Handling damage | 1,993 | 100% |
Show data table
| Line-week | Downtime (min) | Loss (kg) |
|---|---|---|
| Line 1, wk 1 | 357 | 513 |
| Line 2, wk 1 | 347 | 494 |
| Line 3, wk 1 | 353 | 558 |
| Line 1, wk 2 | 282 | 426 |
| Line 2, wk 2 | 252 | 341 |
| Line 3, wk 2 | 161 | 368 |
| Line 1, wk 3 | 352 | 520 |
| Line 2, wk 3 | 279 | 426 |
| Line 3, wk 3 | 455 | 614 |
| Line 1, wk 4 | 380 | 488 |
| Line 2, wk 4 | 264 | 395 |
| Line 3, wk 4 | 289 | 510 |
| Line 1, wk 5 | 330 | 454 |
| Line 2, wk 5 | 281 | 416 |
| Line 3, wk 5 | 186 | 358 |
| Line 1, wk 6 | 190 | 327 |
| Line 2, wk 6 | 154 | 268 |
| Line 3, wk 6 | 356 | 501 |
| Line 1, wk 7 | 221 | 307 |
| Line 2, wk 7 | 240 | 338 |
| Line 3, wk 7 | 431 | 718 |
| Line 1, wk 8 | 240 | 370 |
| Line 2, wk 8 | 256 | 339 |
| Line 3, wk 8 | 233 | 441 |
| Line 1, wk 9 | 260 | 387 |
| Line 2, wk 9 | 287 | 421 |
| Line 3, wk 9 | 331 | 524 |
| Line 1, wk 10 | 234 | 384 |
| Line 2, wk 10 | 185 | 295 |
| Line 3, wk 10 | 267 | 457 |
| Line 1, wk 11 | 303 | 424 |
| Line 2, wk 11 | 286 | 463 |
| Line 3, wk 11 | 389 | 545 |
| Line 1, wk 12 | 118 | 295 |
| Line 2, wk 12 | 265 | 433 |
| Line 3, wk 12 | 320 | 470 |
| Line 1, wk 13 | 251 | 376 |
| Line 2, wk 13 | 262 | 369 |
| Line 3, wk 13 | 451 | 612 |
| Line 1, wk 14 | 264 | 442 |
| Line 2, wk 14 | 233 | 398 |
| Line 3, wk 14 | 258 | 366 |
| Line 1, wk 15 | 180 | 324 |
| Line 2, wk 15 | 238 | 279 |
| Line 3, wk 15 | 325 | 477 |
| Line 1, wk 16 | 276 | 419 |
| Line 2, wk 16 | 251 | 374 |
| Line 3, wk 16 | 370 | 607 |
| Line 1, wk 17 | 89 | 178 |
| Line 2, wk 17 | 237 | 364 |
| Line 3, wk 17 | 351 | 458 |
| Line 1, wk 18 | 213 | 344 |
| Line 2, wk 18 | 202 | 333 |
| Line 3, wk 18 | 304 | 503 |
| Line 1, wk 19 | 268 | 387 |
| Line 2, wk 19 | 378 | 516 |
| Line 3, wk 19 | 314 | 505 |
| Line 1, wk 20 | 117 | 238 |
| Line 2, wk 20 | 215 | 356 |
| Line 3, wk 20 | 489 | 696 |
| Line 1, wk 21 | 183 | 250 |
| Line 2, wk 21 | 380 | 524 |
| Line 3, wk 21 | 344 | 501 |
| Line 1, wk 22 | 210 | 346 |
| Line 2, wk 22 | 278 | 451 |
| Line 3, wk 22 | 352 | 542 |
| Line 1, wk 23 | 101 | 241 |
| Line 2, wk 23 | 300 | 472 |
| Line 3, wk 23 | 280 | 474 |
| Line 1, wk 24 | 220 | 341 |
| Line 2, wk 24 | 342 | 472 |
| Line 3, wk 24 | 291 | 469 |
| Line 1, wk 25 | 106 | 272 |
| Line 2, wk 25 | 347 | 427 |
| Line 3, wk 25 | 390 | 598 |
| Line 1, wk 26 | 109 | 281 |
| Line 2, wk 26 | 334 | 513 |
| Line 3, wk 26 | 349 | 561 |
Recommendations, ranked by share of loss
- 1Start-up scrap (37% of loss). Standardise start-up settings per mould and record first-good-shot time each shift.
- 2Colour changeover purge (29% of loss). Sequence production to reduce colour changes; measure purge volume per changeover.
- 3Short shots and voids (19% of loss). Review injection pressure and drying temperature checks; add a drying log.
- 4Contamination (9% of loss). Seal resin handling between silo and hopper; inspect regrind before reuse.
- 5Handling damage (6% of loss). Revise carton packing and train operators on handling at the take-off station.