Every plant has a repair bench, and most know roughly how busy it is. Far fewer can say which defects arrive there, how long each repair takes, which serial numbers come back a second time or what all of that costs. Repair stations are often the last place on the line still running on paper tags and memory. Good repair station data collection turns that bench into a source of causes and costs instead of a quiet place where problems disappear. See a repair log built by voice in a short repair station demo.
Log Every Repair by Serial, by Voice, Before It Leaves the Bench
iFactory captures the defect and the repair action by voice, blocks release until the unit passes re-inspection and rolls repair labour into cost of poor quality. Technicians keep their hands on the work.
Two Words That Mean Different Things to an Auditor
ISO 9000:2015 draws a clear line. Rework (3.12.8) is action on a nonconforming product to make it conform to the requirements. Repair (3.12.9) is action to make it acceptable for the intended use, and a note says a successful repair does not necessarily make the product conform. The difference decides who must approve the work and what the record must hold. Check how your own codes map with our quality support team.
- The unit ends up meeting the original drawing and specification
- Example: a misplaced resistor replaced with the correct part
- Example: a weld spatter spot ground and blended within tolerance
- Confirmed against the same method the line uses
- The unit works, but no longer matches the original specification
- Example: a jumper wire added where a board trace was damaged
- Example: an oversize fastener fitted after a stripped thread
- Often needs a concession from the customer
One Line per Visit, Kept Against the Serial Number
A repair log is only useful if each entry belongs to one unit. Logging by serial shows which units came back, what was done each time and whether the re-check passed. Logging by shift total hides all of that. The entries below are illustrative. See a serial history pulled up in a live log view.
Count the Laps, Not Just the Repairs
Armand Feigenbaum described a "hidden plant" inside every factory: the capacity spent fixing work that was not right the first time. Rework loops are where it hides. A unit that fails, is repaired, fails again and is repaired again counts as one good unit at the end, but it took three trips. Find the loops in your own data with our analytics team.
Why a 99% Line Can Build Only 87% Right the First Time
First pass yield (FPY) is the share of units that pass a step the first time, with no repair. Rolled throughput yield (RTY) multiplies the FPY of every step to show the chance a unit gets through the whole line untouched. Final yield after repair can look excellent while RTY shows how busy the repair bench really is. The numbers below are illustrative. Plot your own steps in a yield walkthrough.
Ask the Repair Bench What It Saw This Week
Every repair is logged against a serial, a defect and a technician, so supervisors can ask plain questions and get answers drawn from real records rather than tally sheets.
No Repaired Unit Moves On Without a Fresh Check
A repair changes the unit, so the original pass result no longer applies. The unit should be checked again by the same method that caught the defect, and any step the repair could have disturbed. A system hold makes that rule hard to skip. Set up hold rules with our support team.
What IATF 16949 Asks for Reworked and Repaired Product
For automotive suppliers, IATF 16949 sets out controls in clause 8.7.1.4 for reworked product and 8.7.1.5 for repaired product. The two are close, with stricter customer approval for repair. The matrix summarises the main points; read the standard and your customer-specific requirements for the exact wording. Map them to your repair stations in a compliance session.
Say the Repair, Keep Your Hands on the Unit
A technician holding a soldering iron or a torque wrench should not have to stop and type. With voice entry, they say what they found and what they did, and the system reads it back as structured fields before saving. Hear it on a noisy bench with our voice specialists.
Roll Repair Labour Into Cost of Poor Quality
Repair minutes are an internal failure cost, part of cost of poor quality (COPQ), yet they often sit inside direct labour where nobody sees them. Once each repair carries its own minutes, the cost adds up by itself. Every input below is assumed. Run the sum with your own figures in a COPQ review.
Parts used in repair, scrapped units and extra test time come on top. Add them from your own records, not estimates.
Release Check for Every Repaired Unit
Use this list at the bench before any repaired unit returns to flow. Tick each line as it is done. Ask our support team for a print-ready version.
Three Benches, Three Things the Data Showed
These illustrations show the kind of pattern repair data brings out. They are not case studies. Talk through your own benches in a short working session.
Six Measures Worth a Weekly Look
These come straight from a well-kept repair log. Read them by product and by station, not as one plant number. Set targets for them with our quality team.
Repair Station Module and AI Server in One Delivery
iFactory delivers the Repair Station Module as a turnkey package. A pre-configured NVIDIA AI server ships racked and ready, with voice models and links to your test stations set up before it arrives. You rack it, plug in power and Ethernet, and the AI is live. Get it scoped in a turnkey planning call.
- Pre-configured NVIDIA AI server, racked and ready
- Voice entry for defect and repair action at each bench
- Repair logs kept by serial number
- Release holds until re-inspection passes
- Links to AOI, ICT and functional test results
- Repair labour rolled into COPQ reports
- 24×7 remote monitoring of models and hardware
Server installed, test stations and serial labels linked, defect and action codes agreed.
Voice entry and holds run on one repair bench, with technicians shaping the phrases.
Remaining benches added, COPQ reports started, weekly repair review in place.
Repair Data Built at the Bench, Not After the Shift
iFactory is built for technicians with busy hands and quality teams who need clean records. Our support team can map it to your current repair flow.
What Plant Managers Ask About Rework Tracking
Turn the Repair Bench Into a Source of Causes
Log every repair by serial and by voice, hold every repaired unit until it passes a fresh check and see repair labour as a cost. Delivered turnkey and live in 6 to 12 weeks.
