What factory evaluation documents on site: part 2

Key takeawaysFloor records matter more than office certificates because they are generated continuously under production pressure and are difficult to stage convincingly. · Sampling evidence must show the quantity checked, the criteria applied, and the verdict reached, following the accept-or-reject logic described in acceptance sampling references. ^[1] · Live demonstrations of equipment and tests add weight that paper alone cannot, for the same show-it-working reason that underpins factory acceptance testing. ^[2] · Calibration and maintenance logs decide whether every other measurement can be trusted, so expired instruments must leave acceptance duty at once. · Cross-checking records against each other, rather than reading each in isolation, is how assessors detect the gaps worth pursuing.

The first part of this series covered the documents factories prepare before the visit: licenses, manuals, certificates, and organization charts. Those establish identity and intent, and most suppliers have them ready. This second part moves onto the floor, where the evaluation examines the records production actually generates.

Floor records differ from office certificates in one decisive way: they are produced continuously, by many hands, under time pressure. That makes them hard to fabricate convincingly and rich in information for an assessor who knows how to sample. Batch travelers, in-process measurements, sampling decisions, and calibration logs together show whether the documented system runs when nobody is watching.

The sections below explain which on-site records matter, how acceptance sampling evidence should read, what demonstrations add, and how assessors cross-check one record against another. Buyers can use the same map to brief factory contacts on what will be examined.

From certificates to working records

Office documents answer whether the factory is legitimate and organized. Working records answer whether it is controlled, and the two questions diverge more often than buyers expect. A supplier can hold a valid business license, a printed quality manual, and framed certificates while running production on verbal instructions and unrecorded adjustments. The evaluation therefore treats the meeting-room folder as background and the floor records as the main text.

The transition usually happens within the first hour. After introductions, the assessor asks to trace a shipped order or to see the records of the line currently running. How quickly complete records appear, and whether they agree with each other, sets the tone for the rest of the day. Suppliers assessed by UTS Inspection are told this sequence in advance so the right records are retrievable without drama.

Records the floor should be able to produce

How sampling evidence should read

Every acceptance decision on the floor follows the same underlying logic: a defined quantity is judged from examination of a part of it, against stated criteria, producing a verdict that determines what happens next. References on acceptance sampling describe this statistical reasoning and its limits, including how sample size and acceptance numbers relate to the risk accepted. ^[1] An assessor reading sampling records looks for exactly these elements: which lot, how many checked, against what criteria, with what result, and what happened to rejected units.

The most common weakness is verdict-only recording, where the log says pass or fail without quantities or criteria. Such entries cannot be reconstructed, compared, or trended, and they leave the assessor unable to judge whether inspection is strict, lenient, or decorative. The fix is a simple form discipline: every sampling entry names the lot, the sample size, the defect found with its classification, and the disposition. Factories that adopt this discipline often discover their own inspection inconsistencies before any visitor does.

What demonstrations add to paper

Paper records describe control; demonstrations display it. Watching an operator set up a machine, run the normal checks, and record the results tests instruments, methods, training, and record discipline in a single sequence. This show-it-working principle is the same one behind formal acceptance testing of equipment before handover, as described in references on factory acceptance testing. ^[2] The evaluation borrows the logic: claims about capability are confirmed by watching the capability exercised.

Two conditions keep demonstrations honest. First, they must use routine settings, materials, and instruments rather than specially prepared ones, so the live run matches filed records from ordinary days. Second, the assessor should compare the demonstrated result with the written record on the spot, checking that readings, verdicts, and signatures follow the documented method. A smooth demonstration backed by matching records is among the strongest evidence a visit can produce.

Cross-checking records against each other

1. Match settings to instructionsPick the active line's current batch and ask for its traveler, then verify that the settings recorded match the posted work instruction.
2. Tie batches to incoming lotsTake one incoming lot number from the traveler and request its receiving record, checking criteria, measurements, and verdict.
3. Verify instruments behind readingsCompare the gauge serial used for a recorded measurement against the calibration log to confirm current status.
4. Follow deviations to responsesFor any deviation noted in the records, ask for the response entry and confirm it names who decided what.
5. Close the loop at packingTake one packed carton number and confirm its contents trace to the batches claimed on the packing record.

Frequently asked questions