Machine Learning for Improving Quality in Advanced Manufacturing

EDT's inaugural project was based on creating a family of machine learning models that can predict the risk factors surrounding product quality in precision manufacturing.

​

Based on a real world dataset, we were able to apply deep learning techniques to predict point estimates of the quantities of interest 4 minutes ahead of time.

​

This Innovate UK project allowed us to prove the concept of an invaluable technology for the precision manufacturing sector - using commonly collected data and turning it into information that can be used on the shop floor to both improve quality and relieve bottom line pressures.

​

The material impact of this technology being deployed can be illustrated by the following scenario:

​

""

Imagine an SME manufacturer with a £10m turnover facing a 3% scrap rate – that's £300,000 put into the bin annually...

 

Our technology aims to heavily reduce this, saving up to £225,000 each year by providing an early warning to operators for up to 75% of quality failures. These savings are from direct intervention alone, there is also the consideration of time, effort, energy and prevented disruption that would have to be expended on reworking salvageable defective product.

""

​

​

​

​