Industry 4.0 key enabling technologies such as cloud manufacturing allow for the dynamic sharing of distributed resources for efficient use at industrial network level. Interconnected users, i.e. suppliers and customers, offer and request manufacturing services over a cloud manufacturing platform, where an intelligent engine generates a number of solutions based on functional and geometrical requirements. A high number of suppliers leads to a higher number of solutions available for customers increasing the decision-making complexity from a customer perspective. Recommendation systems play a crucial role in expanding the opportunities in decision-making processes under complex information environments. In this scope, this paper proposes the conceptualization and the development of a recommendation decision support tool to be implemented in a cloud manufacturing platform to assist customers in appropriately selecting manufacturing services with reference to sheet metal cutting operations. In terms of solution selection, a Deep Neural Network (DNN) paradigm is adopted to allow for the automatic learning of optimal solution recommendation list based both on customers past experiences and new choices. In this respect, a virtual interaction environment is firstly built for system pre-training. Subsequently, users' data are inputted in the pre-trained model to predict a recommendation list. This is then subject to user interaction, i.e. selection, which will be fed back into the model to update the training parameters. This paper concludes with a simulated case study reported to exemplify the proposed methodology for a variety of decision-making scenarios.

A deep learning based-decision support tool for solution recommendation in cloud manufacturing platforms / Simeone, A.; Caggiano, A.; Deng, B.; Boun, L.. - 86:(2020), pp. 68-73. (Intervento presentato al convegno 7th CIRP Global Web Conference, CIRPe 2019 nel 2019) [10.1016/j.procir.2020.01.019].

A deep learning based-decision support tool for solution recommendation in cloud manufacturing platforms

Caggiano A.;
2020

Abstract

Industry 4.0 key enabling technologies such as cloud manufacturing allow for the dynamic sharing of distributed resources for efficient use at industrial network level. Interconnected users, i.e. suppliers and customers, offer and request manufacturing services over a cloud manufacturing platform, where an intelligent engine generates a number of solutions based on functional and geometrical requirements. A high number of suppliers leads to a higher number of solutions available for customers increasing the decision-making complexity from a customer perspective. Recommendation systems play a crucial role in expanding the opportunities in decision-making processes under complex information environments. In this scope, this paper proposes the conceptualization and the development of a recommendation decision support tool to be implemented in a cloud manufacturing platform to assist customers in appropriately selecting manufacturing services with reference to sheet metal cutting operations. In terms of solution selection, a Deep Neural Network (DNN) paradigm is adopted to allow for the automatic learning of optimal solution recommendation list based both on customers past experiences and new choices. In this respect, a virtual interaction environment is firstly built for system pre-training. Subsequently, users' data are inputted in the pre-trained model to predict a recommendation list. This is then subject to user interaction, i.e. selection, which will be fed back into the model to update the training parameters. This paper concludes with a simulated case study reported to exemplify the proposed methodology for a variety of decision-making scenarios.
2020
A deep learning based-decision support tool for solution recommendation in cloud manufacturing platforms / Simeone, A.; Caggiano, A.; Deng, B.; Boun, L.. - 86:(2020), pp. 68-73. (Intervento presentato al convegno 7th CIRP Global Web Conference, CIRPe 2019 nel 2019) [10.1016/j.procir.2020.01.019].
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/11588/831446
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 13
  • ???jsp.display-item.citation.isi??? ND
social impact