THANK YOU FOR SUBSCRIBING
Manufacturing Technology Insights | Tuesday, May 31, 2022
The coatings industry can evolve to predict failure modes like crack propagation, corrosion, creep, fatigue, color fading, and diffusion of one material into another.
Stay ahead of the industry with exclusive feature stories on the top companies, expert insights and the latest news delivered straight to your inbox. Subscribe today.
FREMONT, CA: When combined with the human brain and artificial intelligence, this digital cannonball can pierce problems deeper and discover more unexpected correlations than humans alone. In this sense, artificial intelligence is an extension rather than an emulation of human intelligence. As a result, AI implementation can be beneficial to the coating industry in various ways.
AI can learn to program painting robots to follow optimal trajectories on specific parts or exact quantities of a particular color needed for a certain effect, using computer vision and industrial internet of things (IIoT) analytics. The paint plant at Lamborghini was the first to use such cutting-edge technology.
Instead of manual data analysis and trial and error, AI can identify defect sources in specific parts or colors using data from pressure regulators, metering pumps, color values, rotary atomizer turbine speeds, airflows, or joint positions and torques.
Root-cause analyses will help to improve the process sequence, identify unknown correlations, and create predictive maintenance schedules for early intervention.
Curing ovens for car body paints experience changes in ambient temperature and airflow. These can result in flaws and other anomalies that the AI will notice. It will conjure up the perfect curing schedule using heat-up curve simulations, bringing the automotive industry closer to its lights-out manufacturing standard.
A machine learning model's input data must be of high quality. For example, as scientists label damaged areas in image sets with greater precision, fault detection accuracy improves. Human comprehension of chemistry, engineering, and physics is still required.
However, to achieve high levels of accuracy, quantity is more important. Continuous data input from IoT devices on the shop floor and ongoing experimentation will generate the millions of data points required to make a machine learning model truly insightful.
AI models, like humans, improve over time as new data is collected. The more information there is, the better AI learns. Furthermore, unstructured data in internal reports, literature, datasheets, and the internet has become a valuable asset. Text mining and natural language processing tools that can extract useful information, even if proprietary, pose a challenge. This can also help prevent the loss of institutional knowledge as the workforce ages.
More in News