How to Build a Quality Inspection Process for LED Therapy Device Shipments
We shipped 500 LED masks to a distributor. 12% had defects: 5% had non-functioning LEDs, 4% had battery issues, 3% had cosmetic defects. The distributor returned the entire shipment. We lost $8,000 in shipping costs and the distributor’s trust. After that, we implemented a quality inspection process. Defect rate dropped to <2%. Here's how we did it.
Quality inspection is not just “checking the product.” It’s a systematic process to catch defects before they ship. Here’s how to build one.
The Quality Inspection Types
| Inspection Type | When | What It Catches | Cost |
| Incoming inspection (IQC) | When components arrive | Component defects (LEDs, batteries, PCBs) | $0.10-0.30/unit |
| In-process inspection (IPQC) | During manufacturing | Assembly defects, workmanship issues | $0.05-0.15/unit |
| Final inspection (FQC) | Before packaging | Functional defects, cosmetic defects | $0.20-0.50/unit |
| Pre-shipment inspection (PSI) | Before shipping to customer | Overall quality, packaging, labeling | $0.30-0.80/unit |
The final inspection (FQC) is the most important for LED therapy devices. This is your last chance to catch defects before the product ships. Inspect 100% of units for function (do the LEDs turn on?) and a sample for detailed inspection.
The pre-shipment inspection (PSI) is essential for large orders. For orders >100 units, inspect a sample (AQL 2.5) before shipping. This catches batch-level issues that may have been missed in FQC.
The AQL (Acceptable Quality Level) Standard
AQL is a statistical sampling standard used to determine if a batch passes or fails.
| AQL Level | Inspection Level | Sample Size (for 500 units) | Pass Criteria |
| AQL 2.5 (standard) | Level II (normal) | 50 units | ≤3 defects = Pass |
| AQL 1.5 (tightened) | Level II (normal) | 50 units | ≤2 defects = Pass |
| AQL 4.0 (loosened) | Level II (normal) | 50 units | ≤5 defects = Pass |
Use AQL 2.5 for standard inspections. This means: if you inspect 50 units and find ≤3 defects, the batch passes. If you find 4+ defects, the batch fails (inspect 100% and rework defects).
Use AQL 1.5 for critical defects (safety issues). For defects that could cause safety issues (e.g., battery overheating), use a tighter tolerance. If you find ≥3 defects in a sample of 50, fail the batch.
The Inspection Checklist
Use a checklist to ensure nothing is missed.
Functional Tests
| Test | How to Test | Pass Criteria |
| Power on/off | Press power button, hold 3 seconds | Device turns on, LED indicators light up |
| LED function | Turn on device, observe all LEDs | All LEDs light up, no dark spots |
| Light output | Use light meter (optional) | Irradiance within spec (e.g., 50-70 mW/cm²) |
| Battery charge | Charge for 1 hour, unplug | Battery indicator shows charging |
| Battery run time | Charge fully, run device | Run time ≥ spec (e.g., 60 minutes) |
| Bluetooth (if applicable) | Pair with phone | Pair within 30 seconds, app connects |
| Auto-off timer | Turn on device, wait | Device turns off after 10 minutes (or spec) |
Cosmetic Tests
| Test | How to Test | Pass Criteria |
| Enclosure | Visual inspection | No cracks, scratches, discoloration |
| Straps | Visual + manual test | Straps intact, adjustable, not frayed |
| Labeling | Visual inspection | Label present, legible, correct info |
| Packaging | Visual inspection | Box intact, foam insert in place, manual included |
The LED function test is the most critical. A dark LED (not lighting up) is a functional defect. For LED therapy devices, this is a primary failure mode. Test every unit.
The battery run time test takes time. Charge the device fully, then run it until the battery dies. Time it. If the run time is < spec (e.g., <60 minutes), fail the unit. For efficiency, you can sample-test 10% of units for run time.
The Inspection Process Workflow
| Step | Action | Responsibility |
| 1. Receive production batch | Production delivers finished goods to QC | Production |
| 2. Random sampling | Select sample based on batch size (AQL table) | QC inspector |
| 3. Functional testing | Test each sample unit per checklist | QC inspector |
| 4. Cosmetic inspection | Inspect each sample unit per checklist | QC inspector |
| 5. Record results | Log pass/fail for each test | QC inspector |
| 6. Calculate pass/fail | If defects ≤ AQL limit, batch passes. If > limit, batch fails. | QC inspector |
| 7. Rework (if failed) | Inspect 100%, rework defects, re-submit | Production |
| 8. Release to warehouse | If passed, release to warehouse for shipping | QC |
The random sampling is critical. Don’t let production select which units to inspect. They may select the best units. Use a random number generator or select every Nth unit.
The rework process: If a batch fails, inspect 100% of units. Separate the defects (rework, scrap, return to supplier). Re-inspect the reworked units. Then re-submit for PSI.
The Defect Categorization
Not all defects are equal. Categorize them.
| Category | Definition | Examples | Action |
| Critical | Safety hazard | Battery overheating, exposed wires | Fail batch, 100% inspection, quarantine |
| Major | Functional defect | LED not working, won’t turn on, battery won’t charge | Fail batch if > AQL limit |
| Minor | Cosmetic defect | Small scratch, minor discoloration | May pass if within limit |
| Incidental | Trivial defect | Dust on surface (easily removed) | Pass, note in report |
The critical defects are non-negotiable. If you find 1 critical defect, fail the entire batch. Investigate the root cause. Correct it. This is a safety issue.
The major defects affect AQL calculation. If major defects exceed the AQL limit (e.g., >3 defects in 50 units at AQL 2.5), fail the batch.
What We’ve Learned
1. The 12% defect rate cost us $8,000 and a distributor relationship. We had no inspection process. Defects shipped to customers. After implementing FQC and PSI, defect rate dropped to <2%. The $0.50/unit inspection cost was worth it.
2. The LED function test catches the #1 defect. 5% of our early units had non-functioning LEDs (soldering issue). The LED function test catches this. We test 100% of units for LED function. Dark LED rate: <0.1%.
3. The random sampling prevents production bias. We used to let production select which units to inspect. They always selected the best ones. Now we use random sampling. Defect detection rate increased 3x.
4. The AQL 2.5 standard is accepted by most buyers. B2B buyers often require inspection at AQL 2.5. If you pass, they trust the batch. If you fail, they expect rework. It’s a common language.
5. The inspection documentation is essential for traceability. We log every inspection: date, batch number, sample size, defects found, pass/fail. If a customer reports a defect later, we can trace it to the batch and inspection record. This is required for medical devices (ISO 13485) and recommended for all products.
Building a quality inspection process for LED therapy device shipments requires understanding inspection types (IQC, IPQC, FQC, PSI), using AQL standard for sampling, creating a detailed inspection checklist (functional + cosmetic tests), following a structured workflow (sample → test → record → pass/fail → rework), and categorizing defects (critical, major, minor). The 12% to <2% defect rate we achieved with an inspection process saved $8,000 in one shipment. Quality inspection is not optional — it's essential for customer satisfaction, brand reputation, and cost control. Implement it before you ship your next batch.