Case Studies
Six examples of problems diagnosed correctly and solved practically. Names and specific locations are not disclosed — but the underlying situations are real.
Blade wear & maintenance
Blade Sharpening Jig Case Study
Lead: Tom BrennanWhile I was working on a contract in a textile company, I was asked to investigate a bottleneck on a cutting line just over 12 months old. The cutting blade kept jamming, which caused the motor shaft to shear into pieces. A new motor was installed, then the bearing on the cutting shaft failed. Each time a new blade was installed, the problem repeated. The line was a significant bottleneck.
Root cause
The blade was struggling to cut the material, causing excessive stress on the motor and shaft.
Solution
I designed a jig for the grinder that allowed me to change the pitch of the blade and sharpen it. I used the old sheared shaft that had been replaced and modified it to fit on the grinder — turning scrap into a precision tool. I then iterated through two more versions of the jig, improving it each time (three versions total).
Results
Overall downtime reduced by 50% initially, improved to 80–85%. Cutting-related downtime reduced by approximately 90% (blades still blunt naturally but can now be sharpened rather than replaced). The line ran noticeably quieter — operator remarked on this. Less stress on motor and shaft. Old blades previously discarded could now be salvaged and reused — ongoing cost saving.
Process & quality
Production Recovery
Lead: Priya KondalChallenge
A critical quilting machine had remained out of operation for an extended period following multiple unsuccessful repair attempts. The machine was central to production scheduling, and the situation had become increasingly costly.
Investigation
Each repair attempt had addressed a different component — belts, motors, sensors — based on what seemed like plausible causes at the time. None of the attempts had been informed by systematic observation of what was actually happening when the machine ran.
Root Cause
Timing had not been correctly synchronised following a belt replacement. The machine was running, but out of phase — producing output that appeared broken to the operators but had no obvious single-component failure.
Solution
The timing synchronisation was corrected. The machine was re-commissioned and returned to full operation.
Outcome
Production successfully restored. The repair cost a fraction of the accumulated attempts.
Lesson
Repeated repair attempts without root cause identification can mask the actual problem. Each failed attempt adds cost and can introduce new variables that complicate diagnosis.
Reliability Improvement
Lead: ReThink InternalChallenge
A cross-lapping machine was experiencing repeated shaft wear. Each occurrence required maintenance intervention, consumed spare parts, and interrupted production. The pattern had continued long enough to be considered normal by the team.
Investigation
Previous maintenance had focused on replacing worn components — shafts, bearings, seals. Each replacement addressed the consequence, not the cause. The pattern suggested a systemic design issue rather than a parts quality or maintenance scheduling issue.
Root Cause
The bushing arrangement was unsuitable for the load and speed conditions. The material and geometry of the bushing created friction and wear that no component replacement could permanently resolve — because the underlying design flaw remained.
Solution
The support was redesigned with a more appropriate bearing solution matched to the actual load and speed conditions. The new arrangement eliminated the conditions causing wear.
Outcome
Reliability significantly improved. Repeated maintenance visits for this machine ceased.
Lesson
Recurring maintenance problems are rarely about parts — they are usually about design. Replacing components that are wearing out without addressing why they are wearing out simply generates more wear, more cost, and more downtime.
Alignment / dimensional
How We Solved a Recurring Webbing Misalignment on a Cross-Lapper
Lead: Avery WhitlockA nonwoven producer had been running the same cross-lapper for years with no issues. The webs had begun drifting — half a degree at first, then a full degree, then enough to misalign the lap every shift. Operators reset the alignment at the start of each run, and the fault returned within hours. The drift was small, repeatable, and quietly costing production time.
The usual suspects all looked plausible on inspection: web-stock tension, edge alignment at the feed rollers, operator loading at the input. None of them correlated with the drift when the machine was watched over several shifts.
I was asked to find the real cause and stop the fault from coming back.
The Investigation
I observed a complete run from raw web through the lap and onto the downstream feed, returning for two more shifts so the cycle could be watched under normal operating conditions. I captured the drift on video at the lap edge across each shift. Web-stock tension at the creel, edge alignment of the feed rollers, and operator loading were each checked and ruled out — none of them correlated with the drift when the machine was watched over time.
The Root Cause
The main drive pulley had drifted approximately 1.5 mm off-true after a bearing change six months earlier. Static inspection did not show it — the pulley looked aligned, the bolts were torque-checked, and the belt sat within tension specification. Under load, with the bearing warm from a full run, the misalignment became visible and large enough to push the web half a degree for every metre of travel through the lap. The bearing swap itself had not been wrong, but the re-convergence around it left a small residual geometry on the machine that nobody had re-checked once the new bearing had settled.
The Solution
The pillow-block was re-shimmed and the drive pulley re-aligned to less than 0.3 mm of total runout, measured hot after a full warm-up run. A short visual-check cadence was added at the start of every shift for the first month; once the new alignment held consistently, the cadence was relaxed to a weekly check folded into the routine maintenance walk. No replacement parts were required.
The Result
The webbing misalignment fault stopped appearing. The lap returned to consistent alignment across shifts, the day-start reset disappeared from the routine, and production time lost to drift recovered immediately. The investigation took less than a day on the floor and the engineering change took an afternoon. The same class of fault has not re-occurred in the 14 months since.
Material handling
Solving a Hidden Labelling Problem Through Observation
Lead: Tom BrennanA bedding manufacturer was experiencing customer returns because some finished quilts were leaving the factory without their product labels.
The labels appeared to be stitched onto the quilts correctly, but somewhere during production they were disappearing. The problem was not being identified before the products were packed and shipped, resulting in customer returns.
I was asked to investigate the labelling machine because it was believed the fault was occurring there.
The Investigation
Rather than focusing only on the labelling machine, I observed the complete production process from the point where the label was sewn onto the quilt through to the end of the production line. During observation, I confirmed that the labelling machine was working correctly and that every label was being stitched onto the product. By following the product through the entire process, I discovered the real cause. As the quilt transferred onto the conveyor, a toothed drive belt was running approximately 3 mm beyond the edge of its pulley. The belt was catching the sewn-in label and tearing it from the quilt before the product reached the end of the production line. The labelling machine was not the problem.
The Root Cause
The root cause was a conveyor guide arrangement that allowed the drive belt to contact the label during product transfer. The symptom appeared at the labelling machine, but the actual cause was further down the production line.
The Solution
I designed and implemented a simple modification to the conveyor guide, preventing the drive belt from contacting the labels during transfer. The modification required only a minor engineering change and was completed with minimal disruption to production.
The Result
Following the modification, the labels remained securely attached throughout the production process. Finished quilts left the production line correctly labelled, eliminating the issue that had been causing customer returns and restoring confidence in the quality of the finished product.
Lessons Learned
The first suspected cause is not always the real cause. By observing the complete manufacturing process instead of focusing on a single machine, it was possible to identify the true root cause and solve the problem with a simple engineering modification. This project reinforced one of the core principles of ReThink Manufacturing: Observe the whole process before changing any part of it.
Quality Improvement
Lead: Tom BrennanChallenge
Recurring product label tears were causing customer returns. The labels were tearing at the point of application, which was attributed at various times to label stock quality, applicator calibration, and operator handling.
Investigation
None of the previously considered causes had been verified through observation of the actual process. When the label path was observed in operation, the tearing was occurring consistently at the same point — suggesting a mechanical cause rather than a materials or handling cause.
Root Cause
A conveyor belt was making contact with the label at the point of application, causing the tear. The contact was subtle but consistent, and only observable when watching the process run.
Solution
A simple engineering modification eliminated the contact between the belt and the label path. No change to label stock, applicator settings, or operator handling was required.
Outcome
Customer returns eliminated. The modification required minimal materials and labour.
Lesson
Quality problems with external symptoms often have simple, observable internal causes. The noise around a problem — multiple theories, attempted fixes, blame spread across people and materials — can obscure a mechanical root cause that is visible when the process is simply watched.
What the work feels like on the floor
Stay in the loop
Reading through these examples but not ready to start an enquiry yet? Leave your email and we'll send you future case work and notes from the production floor.