A Deep Learning Algorithm for the Throughput Estimation of a CONWIP Line - Inria - Institut national de recherche en sciences et technologies du numérique Accéder directement au contenu
Communication Dans Un Congrès Année : 2021

A Deep Learning Algorithm for the Throughput Estimation of a CONWIP Line

Résumé

The ability to meet increasingly personalized market demand in a short period of time and at a low cost can be regarded as a fundamental principle for industrialized countries’ competitive revival. The aim of Industry 4.0 is to resolve the long-standing conflict between the individuality of on-demand output and the savings realized through economies of scale. Significant progress has been established in the field of Industry 4.0 technologies, but there is still an open gap in the literature regarding methodologies for efficiently manage the available productive resources of a manufacturing system. The CONtrolled Work-In-Progress (CONWIP) production logic, proposed by Spearman et al., allows controlling the Work-In-Progress (WIP) in a production system while monitoring the throughput. However, an affordable estimation tool is still required to deal with the increased variability that enters the current production system. Taking advantage of recent advances in the field of machine learning, this paper contributes to the development of a performance estimation tool for a production line using a deep learning neural network. The results demonstrated that the proposed estimation tool can outperform the current best-known mathematical model by estimating the throughput of a CONWIP Flow-Shop production line with a given variability and WIP value set into the system.
Fichier principal
Vignette du fichier
509923_1_En_15_Chapter.pdf (686.79 Ko) Télécharger le fichier

Dates et versions

hal-04030367 , version 1 (15-03-2023)

Licence

Paternité

Identifiants

Citer

Silvestro Vespoli, Andrea Grassi, Guido Guizzi, Valentina Popolo. A Deep Learning Algorithm for the Throughput Estimation of a CONWIP Line. IFIP International Conference on Advances in Production Management Systems (APMS), Sep 2021, Nantes, France. pp.143-151, ⟨10.1007/978-3-030-85874-2_15⟩. ⟨hal-04030367⟩
4 Consultations
30 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More