Acceptable Quality Level (AQL)

What Is Acceptable Quality Level (AQL)?

Definition – Acceptable Quality Level or AQL defines the maximum acceptable Quality level of defects, tolerances, and deviations from the specification that can be accepted by a regulatory authority or sold as part of a standard manufactured for the production of medicine, food, or device. If these deviations are the same as those observed in the sample, they are examples of unacceptable quality. Thus, one crucial aspect of quality control is ensuring that defects do not escape to the final product and do not reach consumers. 

The definition of Acceptable Quality level is much more precise than the one used in Quality Management System (QMS) because AQL incorporates comparative fault analysis, i.e., the idea that defects do not exist only in small quantities and in larger quantities. Thus, fault-based AQL establishes the minimally acceptable number of defects that must be present in any batch within which a product or part is manufactured, handled, or sold.

How Acceptable Quality Level (AQL) Works?

Acceptable Quality Level (AQL) is a concept of quality control used in manufacturing. It involves determining the maximum number of defects allowed in a sample version of a product before rejecting it for further production. If the sample passes the AQL test, it indicates that a manufacturing process has been followed that ensures no defects occur.

Since AQL is based on defective loads, all batches above AQL must be rejected. The AQL level controls quality by demanding that in a batch of products, at least 10% of items be rejected if any product characteristics (such as size, color, or material) are defects.

The AQL standard is used in quality control for all consumer goods, from toys to cars. It ensures that the quality of a sample conforms to the standards of production. This means AQL is the minimum number of defects that must be reached to be considered safe or suitable for sale. 

Sampling Process

Sampling is the process of selecting a small number (typically of ten or fewer units) of a commodity product for evaluation. Process control establishes the conditions under which samples are taken, the criteria for rejection, and the criteria for acceptance.

Below are the parameters:

1. Inspection Level

The inspection level of a sample determines how aggressively the quality check process is carried out. A normal inspection means that the quality of workmanship and materials is assessed as proper. Tight inspections require the most careful examination and critical thinking to determine if flaws or mistakes have been made. Reduced inspections means that the manufacturer doesn’t have to worry about quality control because the item’s quality is considered excellent.

2. Acceptable Quality Level

The acceptable quality level is the number of permissible defects in a batch of goods. An acceptable quality level is required by the buyer when purchasing goods and services from a supplier. It is commonly accepted business practice to set this level at or around what is considered good practice by manufacturers. It’s usually expressed as a percentage defect rate. For example, if the AQL is 90% for a batch of paper towels, then the total number of defects in the entire batch is 1.90/1000 = 0.90%.

3. Sampling Method

Sampling is used to ensure that the sample sets used for quality control are representative and adequate. Below are the methods used for sampling:

  • Single Sampling: The single sampling method is the fastest and is used for all test cases where the data are distributed normally. The single sampling method is used most often in hypothesis testing, where one attempts to reject the null hypothesis if it is true. The main advantage is fast convergence. With single sampling, the time taken to reject each sample increases linearly with the row dimension. Therefore, it usually takes time to reject just one out of many samples in practice, and it converges quickly. This makes it very reliable.
  • Double Sampling: Double sampling is a way to find statistics with a high probability. If the first sample is indecisive in double sampling, then a second sample is taken to decide. This approach has the advantage of being unbiased since it treats both irrelevant and relevant information equally.
  • Sequential sampling: Sequential sampling makes sure you’re testing every possible combination of features — and rejecting the ones that don’t work. You’re taking a whole lot of data and making educated guesses about what might work with a specific sample. As an engineer, you don’t go running back and forth between every option you’ve considered when you run into a problem. Instead, you identify which combination of options is best and iterate from there.

OC curve

OC curves serve as a simple and powerful tool for conveying risk. The more faults on a curve, the greater the risk of acceptance. OC curve is a simple technique in statistical analysis. In it, you calculate the probability that events will occur given a certain starting condition. In this case, we are interested in the number of faults in a batch of data coming from a given source. The higher the number of faults, the less likely it is that our data estimations will be accurate. The lower the value of our estimate, the closer to the truth it gets.

What are the three AQL general inspection levels?

Three AQL levels are used to characterize the levels of detail present in a particular sample. Level I is the minimal level for identifying typical sample defects. Level II focuses on the presence of more minute defects that can be isolated and remediated. Level III includes details that would be expected of a manufacturing sample at this level.

GI inspection level: General Inspection may be the ideal choice for customers with limited time or funds. It is typically faster and less expensive than more thorough Inspection Levels.

One can use GI inspection when:

  • Budget is low
  • Suppliers have a sound quality management system
  • Products do not have a high value

GII inspection level: GII, or Good Inspection Level product inspection, is the type of quality control inspection importers select most often. With a wider coverage path and higher coverage rate than the other two AQL levels, GII inspection offers a perfect balance between cost and safety. Called the “normal” sample size, the GII Inspection level is one of the most popular inspection levels. It is chosen for its trade-off between high coverage scope and evenly distributed costs.

GIII inspection level: The GIII inspection level provides you with the best quality at the lowest risk. This can be translated to “a larger sample size,” allowing for greater accuracy of your acceptance sampling. Unlike any other inspection method, this one will enable you to realize the highest possible yield of your shipment (the percentage of defective pieces in each lot). The GIII inspection level is a pre-shipment inspection conducted to verify that received goods comply with requirements.

Why Acceptable Quality Level (AQL) Matters?

Acceptable Quality Level is essential to quality control because it helps improve quality, prevents defects, and promotes consistency across product lines. It provides consistency for management and quality assurance during manufacturing processes.

The Acceptable Quality Level can be estimated based on experience in the industry, material properties of materials used in manufacture, and similar factors, which will permit rapid determination of quality problems during product development.

Low AQL indicates that a sample of a defect-free product has been rejected or that the quality control processes have failed. On the other hand, high AQL indicates that the quality control process has been effective and accepted all the defects in the sample. AQL improves the probability that consumers will choose products meeting their requirements. It also increases the probability that products will be cost-effective since cost-conscious consumers will be more likely to search for lower quality products that offer lower prices.

Conclusion

Acceptable Quality Level (AQL) is a general term used to describe how closely a quality standard is adhered to, and it’s a key metric for measuring the level of Six Sigma success. While some might initially think that an acceptable level of quality can be assumed from industry, it is rarely possible to match or exceed that standard as a company adopting standards protecting raw materials and processes. In addition, quality never correctly matches quantity; many companies worldwide have proved AQL to be an effective tool for gauging quality practices in manufacturing industries.