AQL - Acceptable Quality Limit (Beginner’s Guide + Step-By-Step Tutorial)

November 17, 2022

AQL - Acceptable Quality Limit (Beginner’s Guide + Step-By-Step Tutorial)

AQL stands for Acceptable Quality Limit and is essential in quality control and product inspection.
In this guide to AQL for beginners, you will learn the following:

  • Quality inspections
  • What is AQL?
  • The need for random sampling
  • Benefits of random sampling
  • 3 steps tutorial on using the AQL table:
    • Step 1: Understanding the basics of the AQL table
    • Step 2: Getting the correct sample size code
    • Step 3: Calculating the sample size as well as the acceptable quality limits
  • Choosing the right sample size and AQL
  • Selecting AQL via InSpec
  • Fun fact: the Origin of AQL 

By the time you reach the end of this AQL essential guide, you will have a strong understanding of what AQL is, why it’s essential to your product and quality control, and how to use AQL to determine your next product inspection.


Why is AQL an essential step in Product Inspections?

In quality control, we use product inspection to ensure that the products of a given shipment meet the required standards. To evaluate the overall production lot quality, inspectors will evaluate products on their functional, visual, and material properties. However, the inspection process is limited by two significant problems: time and cost.

  • Inspecting products individually takes time. Therefore, it is not always possible to verify 100% of the products on every specification. On the other hand, inspecting one piece of the product may not represent a shipment’s overall quality. Considering this, how many pieces is it relevant to inspect to have both a reliable & cost-efficient assessment of your product quality?
  • Achieving 0 defects is not always possible economically or technically. Considering defects will occur, how many defective samples should you accept before it damages the product's usage or your brand reputation?

The AQL helps us answer both questions!


So, what is AQL?

AQL stands for Acceptable Quality Limits. It is a statistical method used to determine the number of defects allowed in a shipment, yet still acceptable from an overall quality standpoint. 

Inspectors use the AQL tables to determine the correct sample size (in simple terms, how many samples are to be picked and inspected from the whole product quantity) and the acceptance level (in simple terms, how many samples can be defective). According to the AQL table, Inspectors will perform a random sampling of your products to assess the different characteristics. If the number of defective products exceeds the AQL, your inspection report will clearly state that your product has failed the AQL level. The batch of products should be considered defective and rejected. 

At Bureau Veritas, we use the ANSI ASQ Z1.4 standard for acceptable quality level sampling when inspectors perform product inspections.

Note: ANSI ASQ Z1.4 is an acceptance sampling system. It is a statistical method for evaluating the quality of a product shipment from a batch of samples. It is the most commonly used standard for the consumer goods industry. It was developed by the American National Standards Institute (ANSO) and the American Society for Quality (ASQ).


The need for random sampling

Whether you own a retail shop, online business, or work as a Quality Control professional, you will need to evaluate the quality of goods or materials before they are shipped. A 100% check of products may be possible during or at the end of production. You can read our article on Full Inspection (or 100% inspection), which suggests that if your product is high value, high risk, or your order is a small quantity. Otherwise, a 100% inspection of the goods or material is not viable due to restrictions on time and manpower. It may not even be practical if the inspection includes a destructive test.

 

Benefits of AQL and random sampling

AQL and random sampling have several benefits that have made them a popular choice for managing product quality for most consumer product supply chains. AQL and random sampling:

  • Save time as only some units in the lot/shipment are inspected.
  • Save cost as the workload is less than a full inspection. Also, the product loss is saved if the inspection includes destructive testing (for example, fabric weight, and welding strength).
  • Set clear criteria for acceptance and rejection helps report precise inspection results.
  • Give the flexibility to choose different AQLs for different aspects of the same inspection. For example, a larger sample size for product appearance checks and a smaller one for destructive tests.
  • Ease the communication of quality expectations within the supply chain.

 

A 3-step tutorial on using the AQL table

Step 1: Understanding the basics of the AQL tables and technical terms

The ANSI ASQ Z1.4 AQL consists of 2 tables.
The first table lists all the different sample size codes (A, B, D...) based on 2 inputs: The “inspection levels” and the “lot or batch size”.
There are 7 levels, and from the left to the right, the number of samples to be pulled out for inspection keeps increasing. There are 3 general inspection levels (G1 to G3) and 4 Special inspection levels (S1 to S4). Using this table, you can determine the sample size code which will be used next step in AQL table 2.

Inspec bureau veritas-aql table 1-step1


The second table lists various sample sizes (top to bottom - 2 to 2000) and the acceptance and rejection numbers against the various AQLs (left to right - 0.065 to 6.5). Using this table, you can determine the maximum number of defective products you can accept in your lot.

Inspec bureau veritas-aql table 2-step1

In addition, it’s also good to learn four terms in the AQL table:

  • Lot or batch size in the table: this is the total quantity in a product batch. 
  • The Inspection Levels: The Inspection Level determines the sample size to inspect by your lot size. There are three commonly used levels: General Level I, II, and II (or GI, GII, and GIII). Level I require fewer samples to be checked, while level III requires more. 
  • The AQL limits: We use it in product inspection to determine how many defective products are acceptable. A product with a higher AQL limit is more tolerant of defects than a product with a lower AQL limit. AQL limits are used in quality control to ensure that products meet the customer's standards. In this tutorial, we use AQL 2.5 as an example. It is also the most widely used model.
  • Defect severity: In quality control, defects are classified into three levels: Critical, Major, and Minor. 
  • Critical – it means the defect is critical. For example:
    • It is unsafe for the customer. The inspector found a needle in the product. 
    • It is unlawful. A product is missing a label required by Government. 
    • If there’s one critical defect, the entire shipment is failed. 
  • Major – a defect affecting the product’s salability or usability. For example, buttons are missing from clothes. 
  • Minor defect – Minor defect refers to a defect that does not affect the salability and usability—for example, a stain inside the garment. 

It is also practical to set different Acceptable Quality Levels for varying levels of defects. For example, choose AQL2.5 for major defects and AQL4.0 for minor defects. On the other hand, critical defects remain not acceptable for quality control.

The buyer and supplier still need to agree on their standard and expectation of quality to determine the definition of the defect level. Our inspectors will inspect the product according to the requirement and address the final result in the inspection report.

Need some guidance? Our consultant is here to help if you want to choose the correct AQL for your product. Book a free consultation!

Step 2: Get the correct sample size code by using the first AQL table

To understand how to use the AQL tables, let’s imagine we are about to inspect a shipment of 40,000 towels. The client inspection manual says the sampling level is General Level I and the AQL is 2.5. 

  1. The shipment quantity is 40,000 towels. Refer to the ‘Lot or batch size’ column on the left side of Table 1 and select the appropriate range. In this case, the correct lot size is 35,001 to 150,000’ (red box).
  2. Next, select the inspection level, which determines how many products the inspector should check. In this case, the inspection level is the blue highlighted ‘General Inspection Level I’. 
  3. Then, look for the intersection of the ‘Lot or batch size’ row with the ‘General Inspection level I’ column. This will give the ‘Sample size code letter’, which in this case is ‘L’. This code letter will help us in Step 3 to get the sample size and acceptance limit.

Inspec bureau veritas-aql table-step2


Step 3: Calculate the sample size and the acceptable quality limits by using the 2nd AQL table

  1. Find the ‘Sample size code letter’ on the left (blue box in the below image). The corresponding number in the ‘Sample size’ table is the sample size to be pulled for this inspection (red box). In this case, quality inspectors will need to randomly pick 200 towels out of the total shipment quantity of 40,000.

  2. Next, we decide the number of acceptable defectives. On the top row, find the AQL that must be followed. In this case, it is ‘AQL 2.5’ (pink box).

  3. Then, look for the intersection of the ‘Sample size’ row with the ‘Acceptable Quality Level’ column. It gives the acceptance and rejection numbers. The acceptance number “Ac” is the number of defectives that is the maximum allowed in an inspection. The rejection number “Re” is the number of defectives that will cause the inspection to fail.

Inspec bureau veritas-aql table-step3
In this case, if the towel samples being inspected has 10 or fewer defectives, it passes the inspection. It failed the inspection if 11 or more out of 200 samples were defective.

Note: if the AQL is set at 2.5, this does not mean that no more than 2.5% of the sample products can be defective. AQL does not represent a percentage. It is a statistically derived standard.

 

Choosing the right sample size and AQL

Everyone in the supply chain expects zero defects in their shipment. However, it is unrealistic unless the product is safety/health related.
Random sample inspections have an inherent risk. Since only a part of the shipment is inspected, there is a possibility of rejecting a good-quality batch or accepting a poor-quality lot. 
However, sampling standards are flexible. There are multiple options of sampling levels and AQLs that allow us to choose these parameters based on the budget, the market, the risk level of the product, and its supply chain. For example, smaller sample sizes and higher AQLs are used when the product is low-risk and made in a trusted facility. The Sample size can also be small if the check involves a destructive sampling. Larger sample sizes and tighter AQLs increase inspection costs but add to the confidence level in inspection results. It is best to use objective data when deciding the sample size and AQL. We hope this blog helped you broaden your understanding of AQL.

 

Selecting AQL is simple via InSpec-bv.com

Booking your product inspection via InSpec is simple. Selecting your AQL is just a few clicks away. After inputting your product quantity, you can select your inspection level and AQL. InSpec shows you the sample quantity and the acceptable level of each type of defect in your order. Our inspector will perform the inspection according to your requirements.

NA
Fun fact: the Origin of AQL 

During World War II, the US military wanted to check the quality of bullets so they could be confident that they would function properly in the field. But the question was how much they should test so that the results are dependable and still there are bullets left to be shipped to the war front. So, they developed the standard “MIL-STD-105” – a sampling standard that uses Acceptable Quality Limit (AQL). It defines the number of units to be selected from a lot that must be inspected or tested to decide the acceptance of the lot. 
Later, several other AQL standards were developed. They were based on more scientific and statistical data, making the results more dependable. The most standard used today is ANSI ASQ Z1.4.

Get in touch today!

Code of Conduct Query - We provide independent, impartial and objective services. Should you have any questions or concerns, please email us