Original): Saif Ali Kheraj
Originally published in the direction of artificial intelligence.
This story is only for us for a member. Update to access all media.
The factories have their own way of speaking. If you've ever been in one, you can hear such things:
“Oee is falling, the CNC lathe has a spindle relief and we need to adjust the power speed to reduce the conversation!”
For most people (and most artificial intelligence) it sounds like a different language. Regular AI models do not understand production conditions. They are trained in the field of general internet text, such as Wikipedia and press articles. If you ask them about the “analysis of the primary cause”, they can talk about the roots of trees instead of repairing a broken machine!
That is why some researchers are working on how to train artificial intelligence especially for production. The idea is to create a model that understands the language of the factory, reads technical textbooks and helps in improving the process.
One of the approaches is the training of the AI model based on Bert (such as chatgpt, but focuses on production). This can help factories improve performance, shorten downtime and support six Sigma projects.
Six Sigma is a method of improving the quality of production. Focuses on:
Reduction of defects process programs for creating more efficient factories
It uses tools such as analysis of the main causes, statistical control and process mapping to solve problems. AI can help by analyzing factory data and suggesting ways to improve the situation.
But … read the full blog for free on the medium.
Published via AI