A factory audit finds the real problems by following one part from the wafer to the box, because the failures appear at the hand-offs between the stages. The wafer that looked clean, the assembly that looked correct, and the test that looked complete can all be right on their own and still deliver a bad part if the hand-off breaks the chain. This article builds the follow-the-batch audit, defines the scope across wafer, assembly, test, and documentation, lists the five red flags to note on any line visit, checks the failure-analysis capability that most audits never see, and closes with a supplier scorecard template.
Following One Batch Through the Factory
The audit’s framing is a single batch followed through the factory, and the framing is the method: track one lot from its wafer through the assembly, the test, and the packing, and audit each hand-off. The problems are at the hand-offs.
The batch-tracking audit starts by choosing a real lot, not a demo. The auditor follows that lot through the fabrication stages, checking at each hand-off that the lot’s identity, the process records, and the quality data travel with the parts. A batch that changes identity, loses its records, or skips a test step at a hand-off is the batch that hides the problem. The audit article documents the manufacturer audit, and the incoming inspection article covers the lot-identity discipline; this article adds the physical-visit checklist.
The reason the hand-offs are where the problems appear is that each stage’s team assumes the previous stage was right. The wafer team records its results, the assembly team assumes them, the test team assumes the assembly — and a single broken assumption at any hand-off propagates to the finished part. The audit that follows the batch catches the assumption that broke.
The Audit Scope: Wafer, Assembly, Test, and Docs
The audit scope is the four domains that define a diode manufacturer, and the scope is what keeps the visit structured. Each domain has its evidence and its red flags.
The wafer domain covers the fabrication: the wafer source, the process records, the electrical test of the wafer, and the traceability of the wafers into the lots. The assembly domain covers the packaging: the die attach, the wire bond, the molding, and the environmental control of the assembly line. The test domain covers the finished-part testing: the test program, the test conditions, the sample rates, and the test equipment’s calibration. The documentation domain covers the whole record: the datasheet version, the qualification reports, the lot traceability, and the compliance declarations. The audit article and the AEC-Q101 verification article document the evidence each domain should carry, and the datasheet comparison article covers the test-condition discipline.
The scope map is the audit’s checklist: the auditor walks the four domains, collects the evidence in each, and the scorecard at the end grades the supplier across the four. A supplier that is strong in the wafer domain and weak in the test domain is a supplier whose parts may not be tested as claimed.
The audit scope is best shown as the scorecard skeleton, with the domains and the evidence each visit grades:
| Domain | Evidence collected | Red flags to note | Weight |
|---|---|---|---|
| Wafer | Wafer source, process records, wafer test | No wafer traceability | High |
| Assembly | Die attach, bond, molding, environmental control | Environmental lapse | High |
| Test | Test program, conditions, sample rates, calibration | Test gap or skip | High |
| Documentation | Datasheet version, qualification, traceability, compliance | Documentation vacuum | Medium |
| Failure analysis | FA tools, returned-part workflow, response time | No FA capability | Medium |
The table is the scorecard skeleton: each domain names the evidence, the red flag, and the weight, and the filled version at the end of the visit becomes the supplier verdict. The audit article and the AEC-Q101 verification article provide the evidence standards each row needs.
Five Red Flags to Note in Any Line Visit
The line visit is where the real operation shows, and five red flags separate a controlled factory from a spot operation. The flags are observable on a single visit.
The first is the missing or inconsistent lot identity: parts on the line with no lot code, or codes that do not match the records, mean the traceability chain is broken. The second is the test gap: a test station that is skipped, under-sampled, or running out-of-calibration equipment, which means the parts are not verified as claimed. The third is the environmental lapse: a bonding or assembly area with uncontrolled humidity or temperature, which is a yield and reliability risk. The fourth is the documentation vacuum: no process records, no test logs, no qualification reports at the line, which means the quality system exists only on paper or not at all. The fifth is the knowledge gap: operators who cannot explain what they are doing or what the test limits mean. The audit article and the counterfeit verification article document the red-flag evidence, and this article names the five to note in person.
The five red flags are the quick audit: any one of them on a line visit is a finding, and two or more are a reason to reconsider the supplier before the next purchase. The flags are observable because they are operational, not promotional.
Failure-Analysis Capability: The Capability You Rarely See
The failure-analysis capability is the capability that most audits never check, and it is the one that predicts how the supplier responds when a part fails in the field. A manufacturer’s ability to analyze a returned part — to find the root cause and correct the process — is the difference between a quality system and a paper one.
The FA capability has three checks. First, the equipment: does the supplier have the failure-analysis tools — the decapsulation, the microscopy, the curve-tracing — to actually find a failure mode? Second, the process: is there a defined workflow for a returned part, from receiving it, to analyzing it, to issuing a report? Third, the responsiveness: can the supplier analyze a returned part in a useful timeframe, or does the field failure sit unanswered? The rectifier failure modes guide and the incoming inspection article document the failure evidence a supplier must be able to read; the FA check verifies the supplier can read it.
The FA check is worth the audit time because it is the supplier’s feedback loop. A manufacturer that can analyze failures corrects its process and improves; one that cannot is destined to repeat the same failure lot after lot. The capability is the audit’s deepest question.
The five red flags also have a timing dimension that a single visit cannot fully show but a trend can: the same line visited twice, a year apart, tells the auditor whether the flags are improving or decaying. A supplier whose lot identity was inconsistent on the first visit and clean on the second has acted on the finding; one whose test gap persists across visits has not. The audit is most powerful when the scorecard is kept year to year, because the trend is what separates the supplier that fixes problems from the one that repeats them. The audit article and the sourcing strategy article both treat the repeatability of the quality system as the long-term signal, and this repeat-visit logic is the reason the scorecard is kept as a file, not as a memory.
The same timing logic applies to the failure-analysis capability. The FA check is not a one-time capability question; it is a workflow that must operate every time a field failure arrives. A supplier that can analyze failures in the audit week but has no defined workflow for the next returned part has demonstrated equipment, not a system. The audit should confirm the workflow exists as a documented, staffed process – the receiving form, the analysis path, the report template – because the system, not the equipment, is what the next field failure will exercise. The rectifier failure modes guide documents the failure types this workflow must be able to read.
Turning Findings into a Supplier Scorecard
The audit closes with the scorecard, and the scorecard is what turns the visit’s findings into a usable verdict. The four domains and the FA capability are scored, weighted, and summarized into a supplier decision.
The scorecard grades each domain — wafer, assembly, test, documentation, and failure analysis — on the evidence collected, with the red flags subtracted and the FA capability weighted for its long-term value. The result is a numeric profile that places the supplier on a spectrum from approved to rejected, with the borderline suppliers flagged for a follow-up visit. The audit article and the sourcing strategy article provide the scoring criteria the scorecard uses, and the automotive-grade framework raises the bar for the safety-rated programs.
The scorecard is the reusable deliverable: the same form, filled after every visit, gives the buyer a trend across time and across suppliers. A supplier whose score improves across visits is closing its gaps; one whose score decays is failing, and the trend is the decision.
The factory audit is the follow-the-batch visit that finds the real problems. Track one lot from wafer to box and audit each hand-off; scope the four domains of wafer, assembly, test, and documentation; note the five red flags on the line; check the failure-analysis capability that most audits miss; and score the findings into a supplier scorecard. The audit turns the visit from a tour into a verdict, and the verdict keeps the next purchase honest. The general rectifier category supplies the parts the audit verifies, and the audit article is the companion reference.