2. Context & scale of operations

The plant supplies retail chains and export markets, where quality and traceability are contractual requirements. QC activities are widespread and highly repetitive:

Existing infrastructure that could be reused: in-line metal detectors, QC forms in Excel, and temperature – dimension – weight measurements already standardized in the process. This was an important foundation for automating without replacing all the equipment.

3. QC before the implementation

The QC process was fully manual and paper-based. A typical inspection round went through five steps:

1. Inspection time — QC staff walk to each stage at fixed times.
2. Manual measurement — one person reads the parameters and another writes them down.
3. Paper/notebook records — figures are copied by hand into a notebook on the floor.
4. Re-entry into Excel — at the end of the shift/day, data is typed from the notebook into a file.
5. Paper record storage — filed by lot and by day.

One parameter could pass through up to 3 handwritten steps before reaching the system — consuming time and accumulating errors at every copy.

4. Problems & pain points

Interviews with operations staff and on-site observation revealed six groups of problems that directly eroded productivity, accuracy and traceability:

The key point: the problem is not people's capability, but that people are doing the machines' job — reading, copying, entering — steps that can be fully automated

5. The solution — automated QC stations at each stage

Instead of making QC staff carry notebooks around the plant, SmartBiz brings the products to be inspected to a QC station right on the line. The station automatically records weight, dimensions, temperature – humidity and metal detector signals directly into the QC form in the system — no paper, no re-entry.

Four automated data collection units

QC station = integrated hardware + automatic recording software. People move from transcribing numbers to confirming data and handling exceptions.

📌 Khám phá ngay: How to run a smart factory

6. How the automation works — from sensors to data

The architecture follows the principles of open architecture and unified data, from the measuring devices at the station to the central data platform:

• Device layer (sensors): electronic scales, AI cameras, temperature/humidity sensors and metal detectors — located at the QC station with an operator screen for QC.
• Edge gateway: collects and pre-processes data at the edge before pushing it to the system, ensuring low latency and stability when the network fluctuates.
• QC software system: backend server, QC database, monitoring dashboard and history/traceability module — a single data source for the entire operation.

The data flow of a QC sample after automation:

1. The product is brought to the QC station on the line.
2. The scale + AI camera + sensors automatically record weight, dimensions and temperature.
3. The metal detector signal is automatically recorded.
4. QC confirms or adjusts on the station screen — every edit is logged.
5. Reports update in real time on the dashboard, with a full audit trail by lot and by person.

Platform features that make the difference

• Automatic recording — scales, cameras, sensors and metal detectors push figures directly into the QC form.
• Controlled edits by QC — staff can still adjust values when needed, and every change is recorded.
• Full audit trail — records who changed what, when and why.
• Threshold alerts — alarms as soon as temperature/weight/dimensions go out of the permitted range.
• Real-time dashboard — supervisors/managers monitor all 12 lines on one screen.
• No more paper records — export PDF/Excel reports by lot, shift or stage.

7. Before & after — a day in the life of a QC employee

📌 Learn more: Understand MES – the smart manufacturing execution system – in 10 minutes

8. Phased implementation roadmap

The whole program was rolled out in a controlled way across 12 lines over about 12–20 weeks (3–5 months), with the POC as the decisive milestone before scaling up:

Running in parallel with the paper process during UAT made it possible to compare old and new data and build trust before paper was retired completely.

9. Results & benefits

The solution has an impact on two fronts: quantitative (efficiency, accuracy) and qualitative (compliance, analytical capability).

Quantitative benefits

Operational effort fell markedly across all five steps — recording, data entry, reporting, traceability and storage — as handwritten steps were replaced by automatic collection and instant synchronization.

Qualitative benefits

• The QC team is freed from manual transcription and can focus on handling exceptions and improving quality.
• Data ready for BRC/HACCP audits — traceable by lot, shift and person in seconds.
• Instant anomaly detection thanks to real-time threshold alerts.
• Consistent data across 12 lines — eliminating discrepancies caused by different recording habits.
• A data foundation for trend analysis and continuous improvement (CI).

10. Success factors & lessons learned

• Reuse existing infrastructure. Make use of the existing metal detectors and QC forms instead of replacing everything — reducing costs and shortening the timeline.
• POC before scaling up. A pilot line to calibrate the AI camera and validate real data, serving as the decisive milestone for the expansion investment.
• Keep people in control. Allow QC to make logged edits — automation does not replace professional judgment, it supports it.
• Run in parallel to build trust. Operate the paper process and the new system side by side during UAT before retiring paper completely.
• Open architecture. APIs ready to connect existing systems and to add advanced QC stages later.

11. Conclusion

The plant's problem was not a lack of people or processes, but processes that forced people to do repetitive work that machines do better: reading, copying and entering numbers. By placing an automated QC station at each stage and unifying data on a real-time platform, SmartBiz turned QC from a manual record-keeping activity into a living data system — more accurate, faster and audit-ready.

More important than the immediate savings, the solution creates a quality data foundation that lets the plant analyze trends, detect anomalies early and improve continuously — turning quality control from a compliance cost into an operational advantage.