Quality assurance criteria
Key takeaways
- Quality assurance builds confidence that requirements will be met before and during production, and it starts with defining exactly what is being assured and against what criteria.
- Criteria are set against the organisation's actual products, services and processes, not generic industry assumptions, so an accurate inventory of what is produced and how comes first.
- A parameter is the specific, measurable characteristic being controlled, such as a dimension, a weight, a response time or a defect count.
- Critical limits are the maximum and/or minimum acceptable values for a parameter; they are set from the specification, not derived statistically from the process like control chart limits.
- A monitoring schedule fixes how often each parameter is checked and by whom, balancing the cost of monitoring against the risk of a non-conformance going undetected between checks.
Quality assurance builds confidence that requirements will be met before and during production (see Quality Management System, on the evolution of quality), which starts with defining exactly what is being assured and against what criteria.
Tools and their deployment
- Checklists and standard operating procedures
- Statistical process control tools (control charts, Pareto analysis)
- Quality audits (Quality Audit)
- Failure Mode and Effects Analysis (FMEA), anticipating how a process could fail before it does
Products, services and processes
Quality assurance criteria have to be set against the organisation's actual products, services and processes, not generic industry assumptions, requiring an accurate inventory of what is produced and how, the same starting point Customer Relationship Management's service charter requires.
Quality parameters
A parameter is the specific, measurable characteristic being controlled, a dimension, a weight, a response time, a defect count, chosen because it reliably indicates whether the product, service or process is conforming to requirements.
Critical limits
Critical limits are the maximum and/or minimum acceptable values for a parameter, beyond which the product or service is considered non-conforming, the same concept as a control chart's upper and lower control limits (see Statistical Quality Control, on control charts), but set from the specification rather than derived statistically from the process itself.
Monitoring schedule
Monitoring is planned on a defined schedule, how often each parameter is checked, and by whom, balancing the cost of monitoring against the risk of a non-conformance going undetected between checks.
Self-check
(a) Name four quality assurance tools and deployment methods. (4 marks)
(b) Distinguish a parameter from a critical limit, with an example of each for a bottling line filling 500ml bottles. (4 marks)
(c) Explain what FMEA does and when in the process it is applied. (2 marks)
(d) Explain how a monitoring schedule balances two competing concerns. (2 marks)
(e) A bakery sets a critical limit of 495 to 505 grams for a loaf's weight but never defines who checks it or how often. Which element from this chapter is missing? (2 marks)
Show worked answer
(a) Checklists and standard operating procedures; statistical process control tools; quality audits; Failure Mode and Effects Analysis (FMEA).
(b) A parameter is the specific, measurable characteristic being controlled, such as fill volume. A critical limit is the acceptable range for that parameter, such as 495 to 505 ml.
(c) FMEA anticipates how a process could fail before it does, systematically identifying potential failure modes and their effects. It is applied proactively, during design or planning, before the process is running.
(d) A monitoring schedule balances the cost of monitoring (more frequent checks cost more time and resources) against the risk of a non-conformance going undetected between checks.
(e) The schedule for product, service and process monitoring is missing, since the critical limit alone does not specify how often or by whom the weight is actually checked.
Exam tip
The parameter and its critical limits are always tested together, be ready to state both for a described product or process: what is measured, and the acceptable range it must fall within.