El QA it has become another stage in the industry. And not only because of the need for manufacturers to adapt their products to safety or other norms and standards that are imposed under the different regulations. Also to satisfy users who have more and more alternatives among the competition, and are increasingly informed about the quality and characteristics of the products on the market.
Therefore, it is the manufacturer himself who must ensure that his products comply with basic standards and of sufficient quality as to have happy customers (loyalty). In addition, these quality controls also serve the industry as good feedback to improve production, and lower costs derived from failures or returns.
What is quality control?
El QA of a company can be applied not only to physical products, but also to certain services or ethereal products such as software. It is a set of tools, actions or mechanisms that are applied in order to identify possible problems or errors in development. In this way, the production process is improved to avoid or reduce future errors or defects, as well as obtaining a much more homogeneous result.
Despite the simplicity of the definition, it is not an easy process. It is made up of various stages (planning, control and improvement) and a series of standards must be met (meeting technical safety requirements, seeking productivity, and increasing profits). In addition, if it is a complex product made up of several subsystems or parts, each of these components may need its own quality control.
If we take a look back, quality control has its precursors, like the work published by Frederick Winslow Taylor for the measurement of works in 1911. Then another method of statistical control would come in 1931 by Walter A. Shewhart. But it wasn't until 1956 that Armand Feigenbaum created total quality control.
Other later works, such as the zero defect theory and Phil Crosby's 14 Steps (1979) were also helpful as a complement. William Edwards Deming would also develop Shewhart's work in 1986. To that would be added other works such as those of Joseph M. Juran and those of Kaoru Ishikawa, both in 1985.
Finally, in 1988, Shigeru Misuno would develop un new quality control applied to the entire "width" of the company, and in 1990 came the use of advanced tools to apply them to total quality control systems, such as the Six Sigma system. This minimizes the variability of the processes, as well as their optimization to reduce or eliminate defects or failures in the products obtained.
From the 90 decade of, quality control processes were increasingly common even in small industries, reaching almost all of the production chains, and in all sectors.
El target Quality control is to offer customers a more satisfactory product with guarantees. For this to be possible, it must be applied to all the processes of the company or, at least, the most critical. Only in this way can a reduced number of failures be found and meet the quality standards set by law or by the company itself.
Currently, to meet these objectives you can count on computerized tools, AI, and automation that greatly streamline the quality control process. For example, you can use siartificial vision systems like the ones we already analyzed in another article on this page.
Of course, there are also models that can be applied to improve processess, such as Learn Manufacturing, a management model to minimize losses and improve end customer satisfaction; or Monozukuri, another more generic practice to optimize chain processes that follows a series of philosophies invented in Japan.
Advantages of quality control
Implementing a quality control system in the company has a series of very clear benefits. The objectives themselves can give you an idea of what you can improve in the company but, in addition to that, you have the following advantages:
- It can improve the productivity of the company, as well as the way in which each process is carried out and the interrelation between them.
- It allows a more detailed monitoring of production.
- It makes it possible to detect problems before the product is sold, which allows you to correct the problem and avoid losses due to returns, repairs or loss of dissatisfied customers.
- It improves the image of the company by maintaining a standard in the quality of its products or services, which can help to differentiate itself from the competition.
- Provides confidence and security to customers. Therefore, you retain customers.
- They allow feedback to improve production constantly. By analyzing the faults or defects, it is possible to determine what went wrong so that it does not happen again or that it happens to a lesser extent.
In reference to this last point, it can also be a good method to detect possible problems of machines or processes that are indicating poor maintenance or the need to implement RCM methodologies, maintenance corrective, quote, or other types of industrial maintenance.
Quality control features
Between the most important features of quality control, the following can be highlighted:
- It is not mandatory in some occasions. For certain non-critical products or services it is not mandatory to implement one, although it is recommended. It is a business decision, unless products are manufactured that must meet safety standards or adapt to certain standards. For example, if you manufacture masks, to be certified with EU standards, they must pass a series of controls.
- It is oriented to customer satisfaction and safety. Since they homogenize production so that all products are equal, and guarantee that minimum standards are met.
- It allows to implement a good improvement framework for business processes.
- It is not universal. That is, you cannot implement the same quality control system or use the same tools for several different industries. Each company needs to adapt the method to its production.
- Allows certification of products and services. If they must adhere to a standard or norm, then quality control is the key to ensuring that it is met.
All that it will rebound in the way it is produced, in how it works, and it will also be reflected in the production and economic results. However, it is not always practical or profitable to implement a production control system for certain non-critical industries, or where the quality of the result is not so dependent on production.
Apply quality control
When a company decides to apply quality control for its product or service, it has to adapt the system and tools very well to be able to carry out the control.
But you should also decide how quality control is implemented. Regardless of whether it is implemented to achieve standards or simply to improve quality, control can be carried out on large batches of products, on all final products or only on some samples.
By example, There are chain production processes in which a multitude of products are produced and it would not be profitable or possible to test each of the pieces obtained. In that case, samples are randomly selected from each lot and testing is done. In other cases, they can be done in batches, or in all the products produced for greater guarantee on certain critical products.
But imagine that the quality control processes involve deteriorating or breaking the part produced. In this case it is impossible to do it on all the pieces, or you would end up destroying everything produced. It would only be possible to apply the tests to the total volume or to very representative samples when the tests carried out are not destructive. For example, it is not the same to do Vickers or Rockwell hardness tests, which are destructive tests of the part, and even some tests of tension, torsion, flexibility, brittleness, etc., to other tests that only involve visual analysis, using rays X, etc.
In the case of not being able to do the quality control inspection in 100% of the units or services, then statistical methods or standards such as the MIL-STD-105E to select samples and apply quality analysis on them. This United States military standard is not the only one, but it is one of the most sophisticated using tables and mathematical models for the samples.
On the other hand, you can choose to perform quality control in different stages of production or at the end of it. For example, if a car is being assembled, it is likely that quality checks will need to be carried out on various parts, and then a final quality check applied to the assembly after completion. More so, if some of the pieces or parts are not manufactured in the factory itself, but come from a third party. In that case, the supplier should have carried out its own quality checks before delivering the product. For example, the tire manufacturer will have to test the tires that will then be mounted on the car.
Having said that, you shouldn't confuse QA with development testing. They are two different things. Although they have similarities, development tests are carried out prior to the start of mass production. For example, following the example of tires, the tire manufacturer must have generated a computer model, with a pattern carved on the tire that guarantees good adhesion and evacuation of water to avoid hydroplaning.
The development process will involve a series of computer simulations to ensure that everything is correct, and then the so-called will be created engineering sample, or a fabricated test on which several tests will be performed under various conditions. Once approved, then the chain production will begin, and it will be in this where the quality control is applied.
Quality control for products
When quality control applies to a product, then the following points should be taken into account:
- Sample: the lot or sample chosen to carry out the inspection must be defined. You must take into account if they are destructive tests or not, and if they are not destructive you could expand the sampling range to achieve greater reliability in the results. Obviously, the selection must be random or not based on parameters that may favor the results obtained, since in that case the reality of the data would be altering. In general, one or more batches are selected from the daily production, or with a different frequency in order to be able to extrapolate the results to the rest of the production. If there has been a change in the raw material, supplier, or processes, a breakdown, etc., it is important to carry out inspections to determine if this could have affected the final product in any way.
- Standardization: you must define some parameters and variables that should be used as a certain quality standard. In case you want to certify the product, those parameters and values will be given to you by the standard under which you want to certify your products.
- Analysis: once you have the samples and standardized methods, now the appropriate tests are carried out to determine that they comply with what is sought in quality control. For example, they can range from load tests for the production of beams, to biological analysis in the food industry, etc. This analysis can also be performed during the production processes themselves, at one of the stages, or at the end of production, as required.
- Data obtained: As a result of applying quality control, favorable data may or may not be obtained. If the product meets all the standards it will be positive, but if it has defects, then other products produced on the same day or from the same batch should be reviewed to see if it is a specific failure or has spread to all products (and worse of the cases). Depending on the industry, in some cases it can be reprocessed to repair problems, and in other cases it must simply be discarded or repaired (for later sale as refurbished). For example, if polymer tubes are being produced and they do not meet the standards, they could be remelted and passed through the extruder again to create a new tube. But if a chip has been manufactured, it cannot be repaired and must be discarded or made some adjustments to sell it remarked.
- Subsequent analysis: If defects have been detected, a study can be carried out on the reason for the failure, and if they have something in common. With this information, the quality of future parts can be improved. For this, the parameters of the production processes must be adjusted, the raw material improved, etc. For example, it is likely that if PCBs are produced with solders, they will be dim. In this case, the reason may be that the tin solder was made at a low temperature and is brittle or has inadequate strength. The remedy would be to increase the temperature of the soldering iron (or maybe it is indicating that the soldering device is bad).
- Decision Making: With the above information, action could be taken to improve quality. Continuing with the previous example, the soldering temperature could be raised, or if this is not possible, replace the soldering iron with a new one.
Quality control for Portfolio
When quality control applies to a the service, then the process can be something more intangible or abstract, but in the same way it has some common steps such as:
- Sample: in this case the sample to be analyzed will not be a physical product. It can be a service or process. For example, it could be software if it is a development company. In this case, only the developed version should be tested, since the rest will be exact copies that will be distributed to customers or users.
- Standardization: for a good quality control methodology, the parameters and variables to be inspected need to be determined. You can define all the details that can be analyzed, such as software, delivery times, work procedures, communication with the customer (for example, if it is a technical service), etc.
- Take the test: according to the above, you must apply the methodology and tools necessary for your service. In these cases, the tests are not usually destructive, as they are intangible services. Therefore, it can be extended to all cases. As with products, they can be positive or not, depending on whether they meet expectations or not.
- Analysis of the results- Data obtained from service samples can be analyzed and learned from it. In this way it will be possible to determine what has gone wrong in each case.
- Decision Making- Based on this data, changes can be made to improve quality or repair the service. For example, continuing with the software case, the bug or vulnerability could be located, a patch developed and updated to correct it.
In these cases, where not a product, it is not always possible to carry out quality control prior to the delivery of the service. In these cases, it can lead to the dissatisfaction of users or customers. On the other hand, in other cases it could also be improved before said service is provided.