Quality assurance as a factor critical to your success
Business potentials of quality assurance
Achieving high product quality is a fundamental success factor for manufacturing companies. To achieve this goal, manufacturing companies are increasingly focusing on the digitalisation and automation of quality assurance processes and their integration into the ongoing production process. This can increase the quantity of the inspection scope, for example by using imaging techniques, and improve production quality. One example is the identification and complete automation of the smooth runners within the inspection process and the concentration of the inspectors on the outliers.
Significant cost savings
However, digitalisation in quality assurance (QA) also enables significant cost savings. In sectors such as the automotive industry or in plant and mechanical engineering, the cost factor for QA is often classified as crucial to business success. Here, there is an enormous savings potential by optimising and automating the inspection process. Automated QA also helps to avoid errors in production and the corresponding follow-up costs. Errors that occur during production or the use of products often cause high costs and lead to a company's loss of image with its customers. At a special brick manufacturer with incinerators and blast furnaces, for example, the product costs themselves are relatively low, whereas the follow-up costs for shutting down a furnace due to defective bricks are to be classified as extremely high. Generally speaking, follow-up damage is critical in all areas where personal safety is at stake, such as medical technology or passenger transport.
Automating and standardising
If the QA process is to be standardised and automated, process analysis and optimisation are number one priorities. At a multinational semiconductor manufacturer, for example, the relatively complex manual inspection of printed circuit boards could be significantly automated by an acoustic imaging process. As a result, the corresponding QA efforts could be brought down from 20 to one hour per week and the reject rate could be significantly reduced. However, higher-level processes can also be optimised, for example by implementing a complaints database to enable complaints to be handled centrally. The evaluation of the complaint data shows where production quality can be improved. Generally, there are still many processes in QA processes in companies that are paper-based. A conversion to digitally based workflows helps considerably to simplify and optimise them.
Advanced analytics as a key technology
But the automated detection of errors is only the first step. True to the motto "From error detection to error avoidance", we are approaching the highest state of the art. Only the continuous digitalisation of the QA process makes it possible to collect and evaluate the data in order to gain new insights. By enriching this data with information from production planning and control systems, as well as sensor data from production, complex optimisations can be derived. For example, which production conditions, environmental variables or fluctuations in the quality of raw materials have an impact on the product and where it is therefore worthwhile to start. The expertise of data scientists is required to identify the appropriate algorithms based on the given problem, the quantity and quality of the data. Large manufacturers such as AWS and Microsoft already offer a low-threshold introduction to the topic of artificial intelligence. Especially in the field of image recognition there are already very good, pre-trained algorithms, which can be trained for specific problems with own image material.
Innovative services through a transparent QA process
Automated quality assurance offers a wide range of potential benefits, such as saving on testing costs, improving quality and increasing customer satisfaction, but also developing new services and business models. Additional profits can be generated through the continuous storage, evaluation and processing of production data. For example, a manufacturing company can make QA data available to its customers through its own portal so that they themselves can track the QA process of the products manufactured for them. A service that not only provides an additional benefit to customers and a competitive advantage for manufacturing companies, but in the case of a chargeable model also generates additional revenue.
One thing is for sure: Digitalisation and automation in quality assurance as well as the use of Advanced Analytics and artificial intelligence offer great potential to production companies in many industries. This ranges from error identification to error avoidance and the associated cost savings to the possibility of new innovative services. However, to implement such measures, raise this potential and steer QA projects in the right direction, one needs an inter-disciplinary approach and the right spectrum of competences.
Thomas Renner, the author of this contribution, is a consulting expert for business analytics with more than 20 years of experience in business process management.