logo SBA

ETD

Archivio digitale delle tesi discusse presso l’Università di Pisa

Tesi etd-07042023-094655


Tipo di tesi
Tesi di laurea magistrale
Autore
PRIFTI, KOSTANTINO
URN
etd-07042023-094655
Titolo
Enabling advanced manufacturing through AI-based Preditive Maintenance
Dipartimento
INGEGNERIA DELL'INFORMAZIONE
Corso di studi
CYBERSECURITY
Relatori
relatore Prof. Saponara, Sergio
relatore Prof.ssa Bernardeschi, Cinzia
relatore Ing. Brembilla, Fabio
Parole chiave
  • accelerometer
  • artificial intelligence
  • maintenance
  • predictive
  • semiconductors
  • testing
  • vibrations
  • z-score
Data inizio appello
21/07/2023
Consultabilità
Non consultabile
Data di rilascio
21/07/2093
Riassunto
The thesis project focuses on the implementation of a Predictive Maintenance system on semiconductor testing equipment. In the thesis project the concept of Predictive Maintenance takes two different paths: Mechanical Predictive Maintenance and Electrical Predictive Maintenance. The purpose of Mechanical Predictive Maintenance is to analyze the vibration signals produced by the testing equipment in order to implement an artificial intelligence model which, by analyzing the real-time data produced by the equipment, is able to understand when the vibrations are anomalous and therefore predict the future equipment breakdown. Electrical Predictive Maintenance, on the other hand, is based on the analysis of electrical signals produced by the testing equipment. The z-score algorithm implements a mechanism that can tell when electrical signals are out of bounds.
File