VIENID Start practicing

Home › Lessons › Food Manufacturing (SSW 2)

Quality Control

Food Manufacturing (SSW 2) · 飲食料品製造業(2号) · Written by Tokutei Pass based on the official study texts; not copied verbatim.

Basics of Quality Control (SDCA & PDCA, Design vs. Manufacturing Quality, Certifications)

Food quality control has three purposes: providing safe and hygienic food, providing delicious high-quality food, and reducing loss to increase profit. Quality Control (QC) is performed as part of company-wide Quality Assurance (QA).

Compared to the targeted 'Design Quality,' the actual result produced is called 'Manufacturing Quality.' First, run the 'SDCA cycle' to maintain operations according to the Standard, and run the 'PDCA cycle' for further improvement or problem-solving.

Subjective 'qualitative information' is replaced with 'quantitative information (numerical values)' such as color charts, sugar content (Brix), and temperatures. When revising SOPs, train workers on the reasons for change and always collect old versions to prevent misuse.

Related regulations and standards include the Food Sanitation Act (positive list system for containers/packaging and mandatory recall reporting), Measurement Act (periodic verification of scales for final products), ISO 9001 (quality), ISO 14001 (environment), and ISO 22000, FSSC 22000, and JFS standards (food safety).

Key points


  • Distinguish Design Quality (target) from Manufacturing Quality (actual result), running SDCA to maintain standards and PDCA for improvements.

  • Convert qualitative information into quantitative numbers, and when revising SOPs, train workers and collect old versions.

  • Understand the features of the Food Sanitation Act, Measurement Act, ISO 9001, ISO 22000, FSSC 22000, and JFS standards.
WordMeaning
Quality Assurance (QA), Quality Control (QC)
design quality, manufacturing quality
qualitative information, quantitative information
Positive List system, Measurement Act
FSSC 22000 standard, JFS standard

Process Control, Inspections, Traceability, and Cleaning Evaluation

Traceability includes 'Trace-forward' (tracking products and shipment destinations from a problematic ingredient lot) and 'Trace-back' (tracing ingredients and production records backward from a product).

In heat sterilization, the time required to reduce microbes to 1/10 is the 'D-value,' and the temperature change required to shorten the D-value to 1/10 is the 'Z-value' (about 10°C for spore-forming bacteria).

Inspections include 100% inspection (e.g., weight check via checkweigher, seal bite inspection) and sampling inspection (used for destructive testing), combining human five-sense 'sensory testing' with instrument-based 'physical-chemical testing' (requiring periodic calibration), while keeping 'retained samples' for root-cause investigation.

For cleaning, use strong alkaline detergent (pH ≥ 11) for burnt oil and protein, acidic detergent (pH ≤ 6) for mineral scale, and pre-rinse with warm water at 60°C or below so protein does not heat-coagulate. The ATP swab test rapidly measures organic residue.

Key points


  • Understand Trace-forward (tracking shipments) and Trace-back (tracing ingredients/processes), as well as sterilization metrics D-value and Z-value.

  • Grasp 100% vs. sampling inspection, sensory vs. physical-chemical testing (instrument calibration), and the role of retained samples.

  • Select detergents according to soil type (alkaline for oil/protein, acidic for mineral scale) and use ATP swab testing to measure organic contamination.
WordMeaning
trace-forward, trace-back
D-value, Z-value
sensory inspection, physical-chemical inspection, calibration
retained sample
ATP swab test

Statistical Quality Control, Pareto Charts, and Fishbone Diagrams

The entire group to be inspected is the 'population,' and items drawn randomly from it are 'samples.' Data includes continuous 'variables data (weight, temperature)' and discrete 'attributes data (defect counts)'; dispersion is expressed by range R (max − min) or standard deviation σ (about 95.4% falls within μ ± 2σ in a normal distribution).

A pie chart's central angle is calculated as '360° × ratio.' Among the 7 QC Tools, a 'Pareto chart' orders defect categories by frequency with a cumulative curve to find priority issues, and a 'Cause-and-Effect (Fishbone) diagram' organizes causes (such as 4M) against an effect.

Key points


  • Understand random sampling from a population, continuous vs. discrete data, range R, and standard deviation σ (approx. 95.4% within μ ± 2σ).

  • Use the 7 QC Tools (Pareto chart, Cause-and-Effect diagram, Histogram, Control chart, Check sheet, Scatter diagram, Stratification) according to purpose.
WordMeaning
population, random sampling
continuous data, discrete data, standard deviation
Pareto chart, cause-and-effect (fishbone) diagram

Histograms, Control Charts, and Complaint Handling (Why-Why Analysis)

Other tools include a 'Histogram' to view distribution center and spread against specifications (a bimodal twin-peak shape indicates two mixed groups), 'Stratification' to compare by material or machine, a 'Scatter diagram' to see correlation between two variables, and a 'Check sheet' for data collection.

A 'Control chart' sets control limits (UCL/LCL) inside product specifications; a point outside the control limits or 7 consecutive points on one side of the center line (a run) indicates a process abnormality. When complaints occur, use 'Why-Why Analysis' to pursue the root cause and prevent recurrence.

Histograms, Control Charts, and Complaint Handling (Why-Why Analysis)

Key points


  • Understand the features of Histograms (a twin-peak shape indicates two mixed groups), Stratification, Scatter diagrams, and Check sheets.

  • Detect process abnormalities on Control charts (outside control limits or 7 consecutive points on one side) and pursue root causes via Why-Why Analysis.
WordMeaning
histogram, stratification, scatter diagram
control chart, control limit lines
Why-Why Analysis

Official references