Manufacturing With Machines That Learn From Defects
Increasing automation and the latest regulatory compliance requirements for fasteners have led manufacturers to reevaluate the 100% inspection and sorting concept, which in the production process is a key part to guarantee a reliable supply of high-quality products to customers.
Like any sizable or complex manufacturing project, the production of mechanical fasteners also involves such a large variety of processes that attaining zero defects is technically not possible. However, today it is possible to learn from defects and improve quality control.
Dimensional characteristics and material structure can be 100% controlled with state-of-the-art inspection and sorting machine technology, but there are still defects of aspects that often require a final visual inspection, which only the human eye can guarantee by manual sorting.
Now Artificial Intelligence (AI) can be applied. Promoting the use of Artificial Intelligence in production systems through customized solutions is the goal of the European KITT4SME project – funded by the European Union through the Horizon 2020 program – Industrial Leadership.
KITT4SME is an acronym for ‘platform-enabled KITs of arTificial intelligence FOR an easy uptake by SMEs’ and aims to ensure that even small and medium-sized manufacturing companies can take advantage of the enormous capabilities and areas of use of Artificial Intelligence.
The solution, which will be available on a digital marketplace, will allow companies to combine hardware and software modules to create ready-to-use digital kits tailored to their needs and specific processes, thus reducing the difficulties in adopting and using cutting-edge technologies.
KITT4SME involves a European network of 18 partners from 8 different countries; in this consortium. Italian company Dimac is bringing its know-how in the design and development of 100% inspection and sorting solutions, collaborating in particular with the Italian companies Holonix – provider of IoT solutions for data collection and processing from sensors – and R2M, a company expert in Artificial Intelligence and innovation.
The KITT4SME concept comes into being from the acknowledge that companies are unprepared and inadequately exposed to extremely complex and widely diverse digital solutions in the field of artificial intelligence that are often much too broad in scope and relentless in dictating their own objectives instead of focusing on the issues a company really faces. Yet, whereas big companies have capitals and resources to research and build platforms internally, SMEs struggle in finding solutions that fit their needs and, even when exposed to them, have to bear significant financial risks to embrace them. Lag in the digitisation race for them is not a system by-product it is a designed feature.
KITT4SME – Gather Shop Floor Data and Let AI Transform It
The aim of the program is that once the customised digital platform is deployed and activated in the company, it immediately starts to SENSE the shop floor to detect in real-time the insurgence of quality, performance and human-affecting issues and of any consequent behaviour of the production system divergence. With AI bringing to the surface hidden dependencies between observed parameters and undesired system effects, it is will be possible to INTERVENE and restore the ‘as-designed’ system behaviour by applying reconfigurations of single elements or entire lines aimed at solving or mitigating the issues being experienced.
For more information: www.kitt4sme.eu
