Product quality assurance is more important than ever, in part because consumers have become so quality-conscious.
At the same time, manufacturers face several challenges, including products that get more complicated year after year and pressure to cut production lead times, as data scientist Yash Mehta describes.
Modernization is essential if manufacturers are to cope. That’s where technology comes in. The right tech can transform quality assurance processes to improve their speed, reliability and effectiveness.
Gathering Data via the Internet of Things
The future of quality assurance processes will be defined by data. How much data your professionals can gather and analyze will directly correlate with the effectiveness of your quality assurance processes.
When it comes to gathering data, few things will be more important than devices networked by the Internet of Things (IoT). IoT sensors, smart beacons and RFID tags are already widespread in the manufacturing industry, and their use will only grow.
Crucially, these devices won’t just let quality assurance professionals capture data from within their own factories. One of the biggest benefits that IoT brings to manufacturing quality assurance processes is the ability to track the quality of supplier products.
Previously, quality professionals could only check what was in their factory. IoT devices let technicians capture data in real-time from suppliers, as Industrial Intelligence’s Darren Tessitore describes. This “makes for much faster data-processing and better quality control,” he writes.
It doesn’t matter where those suppliers are, either. The connected nature of these devices means that the world of manufacturing will become significantly smaller, says Losant’s Todd Henderson. QA professionals in one country will be able to use IoT devices to monitor — from their offices — the quality of parts coming from a factory on the other side of the world.
Predicting Quality Issues with Data Analytics
Most manufacturers are already drowning in data, and they don’t know how to use it. Schneider’s Nathalie Marcotte believes that the sector generates more valuable data than any other industry in the economy. The trouble is very few companies are using it to fuel business growth.
It’s not enough to capture data in real-time, writes IQMS’s Ed Potoczak. Manufacturers need to be able to analyze the data locally as soon as it is captured if they want to create the kind of closed-loop operation that can point to problems as soon as they arise or even anticipate issues before they occur.
This is called predictive data analytics, and it is one of the most important capabilities a manufacturer can have. With machine learning capabilities in place, quality professionals no longer have to wait for issues to occur. They can take a proactive approach and fix a problem before it happens.
A survey by BCG, in partnership with ASQ and Deutsche Gesellschaft für Qualität (DGQ), found that more than 60 percent of companies believe predictive analytics will “significantly affect quality performance and the bottom line within five years.” In the manufacturing sector specifically, respondents believed that predictive quality and machine vision quality control would be some of the more important uses for predictive analytics.
Case in point: Siemens uses machine learning algorithms to carry out proactive maintenance on machinery. Data from milling machines is fed into an algorithm that notifies technicians when the machine’s spindles are close to requiring replacement, Siemens’ Klaus Helmrich explains. The result is decreased downtime and cost savings that total €10,000 per machine per year.
Continuous Monitoring After Production
With new technology, quality assurance processes don’t have to stop when a product leaves the factory.
There are two ways that AI and machine learning can be used in the quality assurance process, explains Philip Kushmaro, a partner at Gemba Finance. The first is to identify and fix production faults that are likely to cause issues with quality if left unattended. The second is to collect and analyze data about the performance of products being used in the field. These insights can then be used by development teams to ensure the quality of future products.
Stu Johnson, a director of product marketing at Plex Systems, believes that monitoring the quality of goods in market offers even greater value than monitoring the manufacturing process. “Equipment manufacturers are beginning to monitor the machines they sell to identify patterns of normal and abnormal behaviors that they can then use to improve their understanding of failure modes,” Johnson writes.
“By identifying the root cause of the product failure with traceability back to the individual operations, operators, and machines involved in its manufacture, companies can identify production flaws that can be corrected to reduce or eliminate future warranty claims.”
It’s not just about creating the best product. Continuous quality checks can help protect a company’s financial and reputational future, notes IBM’s Dan Bigos. When trends and issues are identified, companies can take proactive steps to address or fix the issue before it becomes unmanageable.
In other words, manufacturers can fight PR issues on the front foot.
Testing Quality With Digital Twins
Predictive analytics is one way to use the data collected by IoT devices. Another is to create a fully digital replica of the product being created. This is called a digital twin.
A digital twin is more than a model, explains Georg Kube, global vice president of industrial machinery & components and automotive at SAP: “Thanks to IoT sensors, the twin can receive continuous, real-time data from the object. This unique one-to-one correspondence makes it possible to virtually monitor the object.”
Being able to create a digital twin opens up a world of possibilities. Tom Leeson, an industry marketing strategist at OpenText, notes that because the digital twin can monitor and match every part of the manufacturing process, quality assurance professionals will be able to identify exactly where issues will arise.
It would also let quality professionals run tests with the product, the team at i-SCOOP adds. For instance, they would be able to stress-test the product in a new environment or analyze what would happen if the product were made with different materials or in a different way.
Smart Devices For Smarter Work
No more clipboards on the factory floor. With the right device, quality technicians can access their work on the move, engineer and writer Poornima Apte notes. This also means quality control reports can now be filled in on the move. The result: faster reports and a more streamlined process.
The software that QA professionals use is also important. Digitized and automated quality management systems hold several advantages over traditional paper-based approaches, notes the team at TrackVia. For one, paper-based QMSs require a significant amount of manpower to maintain. It’s also incredibly difficult to analyze data in them. Automated systems are much more streamlined. Data is collected in real-time, meaning that the system is always up-to-date. There’s also only one version, meaning there’s no danger of using an outdated form and no time wasted trying to find the most recent version.
This kind of quality management system is also essential if you are serious about generating insights from the data you have collected, says content strategist Rachel Beavins Tracy. A good QMS can integrate quality-related data like equipment calibration, non-conforming materials and corrective action with data from across your organization. From this data, you can spot trends at a broader level and make the kinds of improvements that significantly improve the quality of products.
Smart devices are also becoming the norm on the factory floor. Augmented reality can be used to help quality assurance professionals quickly spot defects, writes Allerin’s Naveen Joshi. One such application is the use of smart AR glasses, which use embedded IoT sensors to automatically detect imperfections and other quality issues. These issues will be displayed on the glass as soon as the QA professional looks at a product. It’s a technology that could significantly reduce inspection time.
Virtually every facet of a quality assurance professional’s job is going to be transformed by technology in the near future. But these teams need executive help. How well they do in their job will depend on how quickly an organization can modernize.