Inspection in the age of smart manufacturing
Metrology – the science of measurement – is an essential aspect of manufacturing that is sometimes overlooked. Inspection helps ensure that component parts fit together accurately and that final products work properly and safely.
However, even for relatively simple manufactured products, there is more to inspection than first meets the eye. Manufacturers must consider accuracy, reliability, cost, speed and where in the process to carry out the measurement. The requirements vary across industry sectors. Volume production usually relies on a sampling inspection strategy, so automotive component producers typically measure a few parts from each batch to help ensure consistency. Safety-critical applications normally require 100% inspection, so aerospace manufacturers usually inspect each part individually and maintain records for the long term.
With the dawn of Industry 4.0 comes the rise of smart manufacturing technologies that bring new complexities to the inspection process. Yet they can’t be ignored, since they’re integral to the future development of smart factories, more efficient manufacturing and customised products – and will help boost manufacturing the world over. Indeed, according to a new study carried out by Accenture, Industry 4.0 could add as much as $14.2 trillion to the global economy by 2030, driven by smart factory solutions.
So, how can manufacturers update their inspection processes for smart manufacturing?
The challenges of Industry 4.0 and connected factories
The purpose of smart factory metrology is to give greater visibility of the whole supply chain, including suppliers. Smarter, quicker, actionable metrology is an important foundation of the future factory. A factory equipped with industrial robots and automated material handling relies on automated control systems built on timely verification and feedback.
However, inter-connectedness requires more visibility and better data flow, and this requires interoperability. Manufacturers need to adapt their inspection processes to overcome physical barriers, and to close the gaps in metrology information between different stages of the value-chain to support digital, autonomous production.
With the rise of Industry 4.0, it’s never been more pressing to adapt metrology to the needs of the factory. Here are three factors to consider:
- Inline metrology
For manufacturing operations, inline metrology (measurements carried out as part of the process) has several advantages over traditional methods, where inspection is a separate activity. Speed of decision making is crucial. Moving the parts to an inspection department requires additional transport and setup operations which can result in delays. Inline inspection saves time and allows some of the process steps to be removed. Measuring parts in-process can help reduce rework and scrap, as detecting problems sooner helps prevent problems from going unnoticed. Inline inspection also provides the opportunity to minimise the impact of any rework – it is much easier to correct a problem while the part is on the machine. Taken to its logical conclusion, in-process measurement can help prevent problems from occurring in the first place. If variation is detected early enough, it is sometimes possible to take corrective action to compensate for the variation before it becomes a problem.
- Non-contact: optical and laser systems
Non-contact metrology, including laser scanners, phase-based laser scanning and optical systems, are crucial to smart factories. Non-contact measurement allows high resolution scanning of parts and long-distance manoeuvring of measuring devices. However, these methods are not without their challenges. Non-contact metrology typically has line-of-sight limitations and requires accurate positioning to measure detailed features such as holes and sharp edges.The Metrology Hub at the University of Huddersfield is working to increase the technology’s capability for high dynamic range measurement. This aims to improve the ability of metrology to measure work pieces effectively from a distance, whilst maintaining the capacity for very high levels of accuracy.
Accuracy, reliability and continuous production are all well documented benefits of introducing robots into a factory. While robots are not usually specifically designed to measure parts, they can be adapted for use in inspection, especially larger volume components, with software such as PowerInspect®. The sweet spot for robotic inspection is the 100-micron tolerance area and above. This is typically the level at which the flexibility and speed benefits of a robotic system can outweigh its inherent accuracy limitations relative to dedicated measuring equipment. With robotics becoming smarter and more accurate in each lifecycle, the use of machine-based measurement and inspection is only set to increase.
Be ready for a rise of smarter inspection tools
The science of inspection is evolving rapidly, alongside other fields in the Industry 4.0 revolution. For example, EU-funded and multi-partner project, SYMPLEXITY, has been launched to research and improve how humans and robots collaborate, specifically in the application of industrial polishing. According to the consortium, at least 90 percent of finishing work is currently carried out by hand; robots could potentially reduce manual polishing to a level of around 20 percent, with the balance carried out by robots under human supervision.
Evaluating polished surfaces is notoriously difficult and many surface finish parameters have evolved to cover different conditions. The SIR Cell employs a surface finish sensor to measure surface quality; it can report on as many as 13 data parameters. Humans, however skilled, inevitably end up having to make subjective judgments on whether the finished part is good enough. One of SYMPLEXITY’s most exciting offshoots is developing the means to see surface-quality data using augmented reality (AR). AR could mean the end of flat 2D representations on screens. Rather than trying to visualise how a certain parameter change might affect the part, the information is presented directly on the surface.
This helps to remove some of the subjectivity so that users can make more informed decisions more quickly. A human expert will approve the results for the foreseeable future, but when measurements of properties are so quantifiable, the data can be connected directly to the quality assessment (QA) process.
Beyond research, suppliers are also developing new technologies with better measurement capabilities, and manufacturers must embrace these new methods to keep pace with competitors. Better measurement will support the evolution of more truly connected factories, helping reduce downtime, speed up production, and produce better designed, higher quality products. With smarter and more affordable measurement tools emerging rapidly, now is the time for manufacturers to step up and update their inspection processes.
Phil Hewitt is an Autodesk Product Manager specialising in metrology and manages Autodesk PowerInspect, which provides rapid inspection of complex shapes for a wide variety of measuring hardware and can be integrated into a wider manufacturing process.