What is AQL?
Learn about Acceptable Quality Limit and how it's used in the world of Quality Control
Learn about Acceptable Quality Limit and how it's used in the world of Quality Control
AQL sampling is a type of acceptance sampling – a statistical method of picking and inspecting a random sample of goods and ascertaining with a certain level of confidence whether it should be accepted or rejected based on the result.
Under ISO standard ISO 2859-1:1999 AQL (Acceptable Quality Limit) is defined as the “quality level that is the worst tolerable process average when a continuing series of lots is submitted for acceptance sampling”.
In a nutshell, AQL is a standard used to decide on a sample size to be randomly inspected and the number of critical, major and minor defects that are allowed for an inspection lot to be "approved". These decisions are based on the total order size as well as the inspection level required by the buyer. AQL is widely used in the garment industry as a cost-effective and reliable way to predict the defect rate of a shipment as a whole and is a methodology that is preferred to 100% final inspection due to time and cost savings.
AQL tables (or ANSI tables) are standardised tools that represent the criteria and method for determining the inspection sample size required and the number of allowable defects based on different specified AQL limits and inspection levels.
The most widely used AQL standard is a limit of:The appropriate AQL will depend on the specific industry, product end use and market. This decision will take into account levels of regulation, safety risks and customer tolerance of defects. The lower the AQL, the less defects are accepted.
Taking the example of an order of 5,000 pieces, based on General inspection level II (most commonly used inspection level), the sample size code is L which corresponds to a sample size for inspection of 200 pieces. The maximum acceptable number of major defects is 10 and the maximum acceptable number of minor defects is 14. If defects exceed the limit, the whole lot is rejected. Note that a maximum of one defect can be recorded per garment. If a garment has more than one defect, the most critical/major defect is recorded.
* Based on single sampling plan and Normal inspection severity
Table 1
Table 2
We’ve developed a tool that allows you to input your AQL specifications for a specific lot as well as the lot size and immediately calculate the required sample size for an inspection and number of allowable major and minor defects. A lot should be based on an individual product.
It's too time consuming and costly to inspect an entire order, particularly for larger shipments. Testing/inspection may also be destructive and/or disruptive. The AQL sample inspection method is very cost effective and results in a very clear guide regarding the state of your shipment. The inspector will randomly select garments from different cartons to ensure the best chance that the samples are representative of the whole order. Orders are never inspected based on samples pre-prepared by the suppliers. Based on the results of an AQL inspection, buyers are in a good position to make an informed decision regarding whether or not the shipment as a whole is of acceptable quality.
If a fixed percentage is inspected (say 10%), it is possible to inspect less units than is required (in the case of smaller orders) or more units than is required (in the case of larger orders) for a certain level of confidence in the results. AQL sampling on the other hand is a statistical tool that allows you to inspect relatively fewer units as your lot size gets bigger with a similar level of confidence.
An AQL sample inspection will not guarantee zero defects. However, inspections are structured in such a way as to provide a very good guide regarding the quality level of the order as a whole and to allow buyers to make informed decisions regarding shipment approvals without the very high cost of conducting 100% inspections.
The risk of accepting a “bad” lot (wrongly accepting a lot) is known as “consumer’s risk”. The consumer’s risk under AQL 2.5% is approximately 10% whilst the “producer’s risk” (risk of rejection of a “good” lot) is approximately 5%. Using the example of an inspection lot of 5,000 pcs under AQL 2.5% (sample size 200 pcs, maximum allowable major defects 10), the Lot Tolerance Percent Defective (LTPD) or rate of non-conforming items in lots accepted 10% of the time is 7.6%. In other words it is estimated that the sampling plan will accept a rate of 7.6% defective/ non-conforming items 10% of the time.
AQL inspections should be conducted as part of a proactive, preventative approach to quality management. In addition to a final confidence check, it can be used as a way to track and measure progress towards quality improvement over time.
Our defect reporting and predictive analysis features help customers to direct inspection efforts where they are most appropriate and our root cause analysis feature helps to identify specific areas for process improvement. Alongside inspections, we offer our customers the use of free, easy-to use tools, giving them access to best practice in all areas that impact on quality (from creating specifications and standards to logistics and packaging). In this way, we work towards reducing our customers’ total cost of quality which includes not only inspection costs but also prevention costs and the cost of poor quality through claims, loss of reputation and/or loss of sales.
We charge per Manday needed to execute the inspections. This will depend on the sample size requirements for your lots. We can guide you through a simple calculation process to estimate the required Mandays. You'll get an estimate back within 24 hours.
For more information about Happy Rabbit’s methods and approach when conducting AQL inspections, check out some of our FAQs:
How do you decide how many, and which, cartons to take samples from?
Who decides what is a critical defect, a major defect and a minor defect?
Stay up to date with what we're working on and what's happening in our industry.