What Can an IPI Catch Before Mass Production?

Our Initial Production Inspection (IPI) can catch wrong parts, material mismatches, dimensional errors, poor assembly, workmanship defects, failed operating checks and incorrect labels or packaging before most of an order is made. We compare available early-production goods with your approved sample and specifications.

At UTS, IPI normally takes place at 5%–10% production completion. Production has already started and may already be running at scale; IPI does not necessarily precede mass production. If you need a sample checked before the order enters production, our sample evaluation service covers that earlier stage.

Problems IPI Can Catch Early

These checks depend on the goods, documents and test setup available. An unavailable variant or packaging item is not automatically a defect or delay; we compare its availability with the agreed schedule.

How Our Team Checks Early Production

In our on-site inspection work, we check against the purchase order, specifications, approved sample, drawings and packing files. We measure agreed points and run the agreed operating checks where the required setup is available. For sampled inspections, we use the buyer-approved inspection level and sampling plan.

Our reports distinguish available, finished and packed quantities, identify the sampled SKUs and record measurements and test results. We state which checks could not be completed. Before inspection, resolve any conflict between the approved sample and written specifications, and agree which requirements govern acceptance.

Wrong Parts and Material Mismatches

We compare components with the approved parts list and accessories with the SKU they accompany. For textiles, an agreed fabric-weight check in g/m² can identify a mismatch that appearance alone may not reveal.

Visual checks cannot establish chemical composition or restricted-substance content. Those points need laboratory evidence. If a report names a different material or component, we review its coverage and the records linking the tested sample to the goods; a different name alone does not prove the report is inapplicable.

Dimensional and Fit Errors

We measure agreed points against the current drawing and check part fit. This can reveal misplaced holes or incorrect sizes, but a measurement alone does not establish whether tooling, assembly or another factor caused the deviation.

In a hypothetical check, the drawing specifies hole spacing of 100.0 ± 0.5 mm. An assumed reading of 101.2 mm is 0.7 mm above the 100.5 mm upper limit.

This is an assumed tolerance, not an industry standard. NIST notes that a measurement result is complete only when accompanied by a quantitative statement of uncertainty. [1] For an actual check, we need suitable equipment, an agreed method and an acceptance rule that accounts for measurement uncertainty.

Workmanship Faults and Failed Functions

We check stitching, joints and finishes against the approved sample and defect criteria. Agreed operating checks can reveal a sticking zipper, a lid that will not seat or an unresponsive control.

A brief function check does not establish long-term durability. Safety, strength and endurance tests need defined methods, acceptance limits, equipment and durations. If the required setup is unavailable during IPI, the result remains unverified and suitable testing must be arranged.

Incorrect Labels and Unreadable Codes

IPI can reveal a superseded label revision, the wrong model identifier, incorrect instructions or packaging that does not fit. We compare actual production items with the current approved files for each checked variant. Reviewing artwork alone cannot confirm printed-code readability or packaging fit.

For barcode or QR-code checks, 100% of the checked codes must scan successfully on the specified compatible device under agreed conditions. The returned data must match the approved SKU, identifier or destination. An unreadable code or incorrect data requires correction and rechecking.

We record the checked quantity, variants, device and conditions. This requirement applies to the codes inspected. It does not establish that every code in the order was scanned or will work on every future device.

What Early Findings Cannot Confirm

Where acceptance sampling is used, its main purpose is to decide whether a lot is acceptable, not to estimate its quality, as NIST explains. [2] Our early findings do not establish the condition of unmade units or unchecked SKUs. IPI does not replace required laboratory testing or any planned final random inspection of the completed order.

Correcting Problems Before More Units Are Made

For a hypothetical 10,000-unit order, 5%–10% completion means 500–1,000 finished units and 9,000–9,500 unfinished units. These are production quantities, not inspection sample sizes. Unfinished units may already contain affected components, so the supplier should identify both finished goods and work in progress that need correction.

We recommend recording both the number of checked units showing each defect and the total checked for that SKU. Match the follow-up check to the specific correction:

• Old drawing: Confirm the approved revision and reassess the affected measurements against it.
• Replacement component: Check its identity, fit and affected function on newly assembled units.
• Failed safety test: Verify the correction using the applicable test. An appearance check cannot resolve a safety-test failure.

Before affected production continues, ask the supplier to document the correction and how existing goods will be handled. Verify the change on newly produced units, not only on a repaired sample.