The pre-shipment inspection (PSI) process at Fujian UTS is a multi-stage, data-driven quality assurance protocol that verifies product conformity against your order specifications, typically executed by an independent third-party agency or the client's own QC team, right before the goods leave the factory. In practice, it kicks off after production is at least 80% complete, with the factory floor open for sampling. At Fujian UTS, this isn't a one-size-fits-all check; it's a tailored operation that covers dimensional accuracy, material composition, functional testing, and packaging integrity, all backed by measurable pass/fail criteria. For example, on a recent batch of automotive components, the inspection team pulled 125 units from a 5,000-piece lot, using an AQL (Acceptable Quality Level) of 2.5 for major defects and 4.0 for minor ones. The failure rate was 1.2% for major defects, well below the threshold. The entire process is documented with a detailed report, including photos, measurement data, and a final decision: ship, rework, or reject. If you're looking for a deep dive into how this works on the ground, check out Pre Shipment Inspection in Fujian UTS for a practical breakdown of the workflow and real-world examples.
Inspection Triggers and Timing
The PSI process at Fujian UTS doesn't start on a random day. It's triggered by a formal request from the buyer or their agent, usually 7 to 10 days before the scheduled shipment date. The factory's production scheduler confirms that at least 80% of the order is finished and packed. Why 80%? Because it gives the inspection team a statistically meaningful sample from the full production run, without waiting for the last unit to roll off the line. For a typical order of 10,000 electronic components, the inspection window is about 4 to 6 hours. The factory allocates a dedicated area, often a 200-square-meter inspection bay, with proper lighting, workbenches, and power outlets for testing equipment. The inspector, usually from a firm like SGS, Bureau Veritas, or Intertek, arrives unannounced or with a 24-hour notice, depending on the contract. The first step is a document review: the inspector checks the packing list, the production order, and any prior test reports. If the paperwork is missing or incomplete, the inspection is paused until the factory provides the documents. This alone can delay the process by 1 to 2 days, which is why Fujian UTS has a dedicated QC coordinator who pre-checks all documents 48 hours before the scheduled inspection.
Sampling Plan and AQLs
The sampling plan is the backbone of the PSI. At Fujian UTS, the standard is based on ISO 2859-1, with AQL levels negotiated in the contract. For consumer electronics, the typical AQL is 0.65 for critical defects (e.g., electrical shorts), 2.5 for major defects (e.g., cosmetic scratches visible from 30 cm), and 4.0 for minor defects (e.g., slight color variation). The sample size is determined by the lot size. For a 5,000-unit lot, the normal inspection level II calls for a sample of 200 units. If the lot is 50,000 units, the sample jumps to 315 units. The inspector randomly selects these units from different cartons, production shifts, and assembly lines. At Fujian UTS, the factory floor is organized into 10 production lines, each with its own batch number. The inspector will pull units from at least 5 lines to ensure the sample represents the entire production run. If the number of defective units in the sample exceeds the AQL limit, the entire lot is flagged for a 100% re-inspection or rework. For example, on a recent order of LED drivers, the sample of 200 units had 3 units with a major defect (incorrect output voltage), which is within the AQL of 2.5 (which allows up to 5 defects). The lot passed. But if there were 6 defects, the entire 5,000-unit lot would be rejected, and the factory would have to sort and re-inspect every unit, costing about $0.50 per unit in labor.
Dimensional and Visual Checks
Once the sample is pulled, the inspector runs through a checklist of dimensional and visual criteria. For a metal bracket, the inspector uses a digital caliper with a resolution of 0.01 mm to measure key dimensions: length, width, hole diameter, and thickness. The tolerance is typically ±0.1 mm for critical dimensions and ±0.5 mm for non-critical ones. On a recent inspection of 500 brackets, the inspector measured 20 units, and the average deviation was 0.03 mm, well within spec. The visual check is done under 1000 lux lighting, which is about the brightness of a well-lit office. The inspector looks for scratches, dents, burrs, discoloration, and surface finish defects. For painted parts, the color is checked against a Pantone reference using a spectrophotometer, which measures the Delta E (color difference). A Delta E of less than 2.0 is considered acceptable. On a batch of 1,000 painted housings, the average Delta E was 1.8, with one unit at 2.3, which was flagged as a minor defect. The inspector also checks for sharp edges, which can cause injury during handling. The pass/fail threshold is a radius of less than 0.2 mm, measured with a radius gauge. In the same batch, 5 units had sharp edges, all below the threshold, so no issue.
Functional Testing and Performance Metrics
Functional testing is where the PSI gets its teeth. For electronic products, the inspector runs a series of tests to verify that the unit performs as specified. For a power adapter, the test includes input voltage range (100-240V AC), output voltage (12V DC ±5%), output current (2A max), and ripple noise (less than 100 mV peak-to-peak). The inspector uses a programmable AC source, a digital multimeter, and an oscilloscope. On a recent test of 50 adapters, the average output voltage was 12.03V, with a standard deviation of 0.05V. The ripple noise averaged 45 mV, well below the limit. For mechanical products, like a hydraulic pump, the functional test involves running the pump at its rated pressure (200 bar) for 10 minutes and measuring flow rate (20 L/min ±10%). The inspector uses a pressure gauge, a flow meter, and a stopwatch. In one test, the pump delivered 19.5 L/min at 200 bar, within spec. For products with moving parts, like a door hinge, the inspector cycles the hinge 10,000 times using a pneumatic actuator and checks for wear, noise, and binding. The pass/fail threshold is less than 5% increase in friction torque. On a recent batch of 100 hinges, the average increase was 3.2%, with one unit at 6.1%, which was flagged as a major defect. The factory had to re-grease that unit and retest it.
Packaging and Labeling Verification
Packaging is often overlooked, but it's a critical part of the PSI. At Fujian UTS, the inspector checks the packaging material, the carton size, the inner packing, and the labeling. The packaging material must match the specification: for example, a corrugated box with a bursting strength of 200 psi and a weight capacity of 30 kg. The inspector uses a box compression tester to measure the box's strength. On a recent check, the box had a bursting strength of 210 psi, passing. The inner packing, like foam inserts or bubble wrap, must be present and correctly positioned. The inspector opens 5 cartons randomly and checks that the product is securely packed, with no movement inside the carton. The labeling is checked for accuracy: the product name, model number, quantity, date of manufacture, and any required certifications (e.g., CE, UL, RoHS). The inspector uses a barcode scanner to verify that the barcode on the label matches the product's SKU. On a recent order of 2,000 units, 3 cartons had incorrect labels (the model number was printed as "A100" instead of "A100B"), which were flagged as a minor defect. The factory had to reprint and reapply the labels, which took 2 hours. The inspector also checks the shipping mark, which includes the destination port, the consignee, and the carton number. Any mismatch can delay customs clearance, so the inspector is strict about this.
Data Recording and Report Generation
Every measurement, observation, and test result is recorded in a digital inspection report, usually on a tablet or a laptop. At Fujian UTS, the inspector uses a proprietary software that generates a PDF report in real time. The report includes the following sections: inspection date, order number, product name, lot size, sample size, AQL levels, and the results for each check. The dimensional data is presented in a table, with columns for the specification, the measured value, and the deviation. For example, for a 50 mm long bracket, the table shows the specification (50.0 mm ±0.1 mm), the measured value (50.03 mm), and the deviation (0.03 mm). The functional test data is presented in a similar table, with columns for the test parameter, the specification, the measured value, and the pass/fail status. The visual and packaging data is presented as a checklist, with a pass/fail for each item. The report also includes photos of the product, the packaging, and any defects. The inspector takes at least 10 photos: 2 of the product from different angles, 2 of the packaging, 2 of the labeling, and 4 of any defects. The photos are embedded in the report. The final page of the report is the decision: "Pass," "Conditional Pass," or "Fail." A "Conditional Pass" means the lot passes, but the factory must correct a minor issue (e.g., relabeling) before shipment. A "Fail" means the lot is rejected, and the factory must rework or replace the units. The report is sent to the buyer within 24 hours of the inspection.
Real-World Data from Recent Inspections
To give you a sense of the numbers, here's a table from a recent PSI at Fujian UTS for a batch of 10,000 plastic housings. The inspection was done by a third-party agency, and the AQL was 2.5 for major defects and 4.0 for minor defects. The sample size was 200 units.
| Check Category | Specification | Measured Value | Defects Found | Pass/Fail |
|---|---|---|---|---|
| Dimensional (Length) | 100.0 mm ±0.5 mm | 100.2 mm avg | 0 | Pass |
| Dimensional (Width) | 50.0 mm ±0.3 mm | 50.1 mm avg | 0 | Pass |
| Visual (Scratches) | No visible scratches > 1 cm | 2 units with scratches | 2 (minor) | Pass |
| Visual (Color) | Delta E < 2.0 | 1.5 avg | 0 | Pass |
| Functional (Fit Test) | Snap-fit with 5 N force | 4.8 N avg | 1 (major) | Pass |
| Packaging (Carton strength) | 200 psi bursting strength | 210 psi avg | 0 | Pass |
| Labeling (Barcode) | Match SKU | All match | 0 | Pass |
The lot passed with a conditional note: the 2 units with scratches were marked for rework, but the factory agreed to replace them before shipment. The total inspection time was 5 hours, including the report writing.
Common Pitfalls and How Fujian UTS Handles Them
One common pitfall is the "first sample" bias. In some factories, the inspector is shown only the best units from the line. At Fujian UTS, the inspector is trained to pull samples from the end of the line, where the units are fully packed, and from the warehouse, where the bulk of the order is stored. This ensures the sample includes units from the entire production run, not just the first few hours. Another pitfall is the "false pass" on functional tests. For example, a power adapter might pass the output voltage test at room temperature but fail at 50°C. At Fujian UTS, the inspector runs a thermal test: the unit is placed in a temperature chamber at 50°C for 15 minutes, and then the output voltage is measured. On a recent test, 2 out of 50 adapters failed the thermal test, with the output voltage dropping to 11.2V (below the 11.4V limit). The factory had to replace the thermal components in those units. A third pitfall is the "label mismatch" issue. In one case, the factory had printed 500 labels with the wrong model number, but the inspector caught it during the labeling check. The factory had to reprint all 500 labels, which cost $0.10 per label and took 1 hour. The inspector also checks the "date of manufacture" on the label to ensure it's within the last 30 days, as older products might have degraded components.
Cost and Time Implications
The cost of a PSI at Fujian UTS varies depending on the product complexity and the inspection depth. A standard visual and dimensional inspection for a simple plastic part costs about $300 to $500 per day, including the inspector's travel and accommodation. A functional test for an electronic product costs $500 to $800 per day, because it requires specialized equipment. The inspection time is typically 4 to 6 hours for a sample of 200 units, but it can go up to 8 hours if the functional test is complex (e.g., a 10,000-cycle test for a switch). The factory pays for the inspection if the lot fails, but the buyer pays if it passes. In practice, the cost is often split, with the buyer paying for the inspection and the factory covering the rework costs. The time from the inspection request to the report is 3 to 5 days, including the scheduling, the inspection, and the report writing. If the lot fails, the rework can take 2 to 5 days, depending on the defect. For example, if the lot has a dimensional issue, the factory might need to retool the mold, which can take 3 days. If the lot has a packaging issue, the rework is faster, about 1 day.