It is a well-known fact that maintenance cost inside a company can be the largest part of operational expenses, second only to energy. Usually, replacing a component or equipment just before a breakdown occurs is the best way to minimize the maintenance cost. This is the main reason why a lot of companies are struggling in collecting data from equipment, and in finding ways to exploit these data for predictive purposes. In this paper we are going to explore multiple sensors' data extracted from an injection moulding machine, with the final aim of developing a Predictive Maintenance model tailored on the specific machine utilization. After the extraction of a training set, we implemented Machine Learning algorithms in order to find the best predictive model able to discern between correct functioning and border line functioning of the machine. We are going to describe the performance reached by the developed model and to show how it deals with completely new data used for testing the model.

Using Plastic Injection Moulding Machine Process Parameters for Predictive Maintenance Purposes / Pierleoni, P.; Palma, L.; Belli, A.; Sabbatini, L.. - ELETTRONICO. - (2020), pp. 115-120. (Intervento presentato al convegno 2020 International Conference on Intelligent Engineering and Management, ICIEM 2020 tenutosi a gbr nel 2020) [10.1109/ICIEM48762.2020.9160120].

Using Plastic Injection Moulding Machine Process Parameters for Predictive Maintenance Purposes

Pierleoni P.
;
Palma L.;Belli A.;Sabbatini L.
2020-01-01

Abstract

It is a well-known fact that maintenance cost inside a company can be the largest part of operational expenses, second only to energy. Usually, replacing a component or equipment just before a breakdown occurs is the best way to minimize the maintenance cost. This is the main reason why a lot of companies are struggling in collecting data from equipment, and in finding ways to exploit these data for predictive purposes. In this paper we are going to explore multiple sensors' data extracted from an injection moulding machine, with the final aim of developing a Predictive Maintenance model tailored on the specific machine utilization. After the extraction of a training set, we implemented Machine Learning algorithms in order to find the best predictive model able to discern between correct functioning and border line functioning of the machine. We are going to describe the performance reached by the developed model and to show how it deals with completely new data used for testing the model.
2020
978-1-7281-4097-1
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11566/284875
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 12
  • ???jsp.display-item.citation.isi??? ND
social impact