| KEYWORD |
Early detection of Defect in Industrial Production Line Through Data Analysis and Machine Learning Applications
keywords BIG DATA, DATA ANALYSIS, MACHINE LEARNING
Reference persons ELENA MARIA BARALIS, DANILO GIORDANO, MARCO MELLIA
External reference persons Federico Perrero
Research Groups DAUIN - GR-04 - DATABASE AND DATA MINING GROUP - DBDM
Thesis type INDUSTRIAL, RESEARCH / EXPERIMENTAL
Description Production line processes for chips are complex systems composed of different starting from printing the component to testing it. Each step is subjected to different conditions and phenomena (e.g., environmental, wear and tear ageing). These differences influence the final quality of the components.
With Industry 4.0, these processes and all the produced components are constantly monitored, and the corresponding data can be used to get insights about the manufacturing process.
The thesis aims to analyse this data, describe the manufacturing process and the component result (good or defect), identify which factors co-occur when a defect is present, and develop a machine learning methodology that highlights these factors before testing the component.
The thesis will be part of an Industrial partnership with Bitron S.p.A., a leading industry in producing chips worldwide.
Required skills data science
Deadline 24/09/2022
PROPONI LA TUA CANDIDATURA