Within the constantly growing Industry 4.0 environment, the spread of smart tools to aid manufacturing processes increases everyday. Specifically, the possibility to exploit simulation tools in order to monitor physical resources and extract useful data from them is tangible. One of the main technological enablers of the Industry 4.0 paradigm is, in fact, the so-called Digital Twin. In the Cyber Physical System (CPS) paradigm, the Digital Twin makes it possible to create a virtual copy of a physical asset, enabling real-time monitoring functions for disparate applications within the industrial production field. Given the fact that this tool is at an early development stage worldwide, there is still no specific or cohesive definition of it to be found in literature. However, most of the technological endeavors refer to the Digital Twin as an asset whose mere function is limited to the monitoring of its physical counterpart. As a consequence, almost the totality of the case studies found in literature can be referred to as Digital Shadows, since they only change their state when the physical side does it. This work of thesis aims at transforming the Digital Shadow into a well-rounded Digital Twin, which is able to communicate in real-time in both directions between the digital and physical side of a resource. This is possible by integrating with the digital tool certain communication protocols, which enable the sending of commands to the Manufacturing Execution System (MES) software related to the specific production facility. The implementation starts with the proposal of optimization frameworks that can be applied to any manufacturing environment. These models involve the interaction of the physical and digital sides of an asset, and work in both directions, enabling real-time optimization measures with the use of a Digital Twin. An application study of the proposed frameworks is given in the context of the Industry 4.0 Laboratory at Politecnico di Milano. Here, the development of a tailored Digital Twin for each workstation belonging to assembly line has been carried out. The Digital Twins have been created in Simulink environment; these are modelled to tackle basic shop-floor events, allowing the reduction of downtime, enhancing human-machine interaction and re-scheduling the work orders under determined conditions. Validation is also granted by the calculation of the magnitude of the optimization provided by the use of the integrated Digital Twin in the working processes of the assembly line. Possible future developments might include the integration of maintenance models with the Digital Twin, in order to allow real-time automatic planning and scheduling based on forecasting models.
Nel contesto dell’Industria 4.0, la diffusione dei cosiddetti smart tools progettati per supportare i processi produttivi cresce ogni giorno. Nello specifico, la possibilità di utilizzare strumenti di simulazione al fine di monitorare le risorse fisiche per poi ricavarne dati utili è tangibile. Una delle principali tecnologie abilitanti del paradigma Industria 4.0 è il Digital Twin. Nel panorama dei Cyber Physical System (CPS), il Digital Twin rende possibile la creazione di una copia virtuale di una risorsa fisica, abilitando funzioni di monitoraggio in real-time per le svariate applicazioni all’interno del panorama produttivo industriale. Considerando che questo strumento si trova nella sua fase iniziale di sviluppo, non è stata ancora elaborata una sua definizione precisa e totalizzante in letteratura. In ogni caso, buona parte del materiale di ricerca si riferisce al Digital Twin esclusivamente in relazione alle funzionalità di mero monitoraggio della sua controparte fisica. Di conseguenza la quasi totalità dei casi studio trovati in letteratura può essere definita Digital Shadow, in quanto il modello digitale cambia stato soltanto quando lo fa anche la parte fisica. L’obiettivo di questo lavoro di tesi è la trasformazione del Digital Shadow in un Digital Twin capace di comunicare in real-time in entrambe le direzioni fra il lato fisico e digitale di una risorsa. Ciò è reso possibile grazie all’integrazione nel tool digitale di protocolli di comunicazione che permettono di inviare comandi al Manufacturing Execution System (MES) dello specifico impianto produttivo. Il lavoro è cominciato formalizzando due framework di ottimizzazione, pensati per qualunque ambiente di produzione industriale. È in questi modelli che si concretizza l’interazione bi-laterale fra il lato fisico e quello digitale di un asset, che rende possibile l’ottimizzazione in tempo reale tramite l’uso del Digital Twin. Un caso applicativo dei framework proposti è stato implementato nel contesto dell’Industry 4.0 Laboratory, presso il Politecnico di Milano. Qui, è stato sviluppato un set di Digital Twin, modellati su misura di ciascuna stazione di lavoro della linea di assemblaggio presente. I Digital Twin sono stati creati in ambiente Simulink al fine di gestire semplici eventi a livello produttivo, permettendo la riduzione dei tempi morti, il miglioramento dell’interazione uomo-macchina e la pianificazione reattiva degli ordini di lavoro in determinate condizioni. La validazione del modello è garantita anche dal calcolo dell’entità dell’ottimizzazione apportata facendo uso del Digital Twin integrato con il MES nell’ambito dei processi di assemblaggio della linea. Possibili sviluppi futuri includono l’integrazione di modelli di manutenzione con il Digital Twin, al fine di abilitare un planning automatico basato su modelli predittivi.
Development of MES integration functionality for a digital twin in manufacturing environment
BERARDI, STEFANO
2017/2018
Abstract
Within the constantly growing Industry 4.0 environment, the spread of smart tools to aid manufacturing processes increases everyday. Specifically, the possibility to exploit simulation tools in order to monitor physical resources and extract useful data from them is tangible. One of the main technological enablers of the Industry 4.0 paradigm is, in fact, the so-called Digital Twin. In the Cyber Physical System (CPS) paradigm, the Digital Twin makes it possible to create a virtual copy of a physical asset, enabling real-time monitoring functions for disparate applications within the industrial production field. Given the fact that this tool is at an early development stage worldwide, there is still no specific or cohesive definition of it to be found in literature. However, most of the technological endeavors refer to the Digital Twin as an asset whose mere function is limited to the monitoring of its physical counterpart. As a consequence, almost the totality of the case studies found in literature can be referred to as Digital Shadows, since they only change their state when the physical side does it. This work of thesis aims at transforming the Digital Shadow into a well-rounded Digital Twin, which is able to communicate in real-time in both directions between the digital and physical side of a resource. This is possible by integrating with the digital tool certain communication protocols, which enable the sending of commands to the Manufacturing Execution System (MES) software related to the specific production facility. The implementation starts with the proposal of optimization frameworks that can be applied to any manufacturing environment. These models involve the interaction of the physical and digital sides of an asset, and work in both directions, enabling real-time optimization measures with the use of a Digital Twin. An application study of the proposed frameworks is given in the context of the Industry 4.0 Laboratory at Politecnico di Milano. Here, the development of a tailored Digital Twin for each workstation belonging to assembly line has been carried out. The Digital Twins have been created in Simulink environment; these are modelled to tackle basic shop-floor events, allowing the reduction of downtime, enhancing human-machine interaction and re-scheduling the work orders under determined conditions. Validation is also granted by the calculation of the magnitude of the optimization provided by the use of the integrated Digital Twin in the working processes of the assembly line. Possible future developments might include the integration of maintenance models with the Digital Twin, in order to allow real-time automatic planning and scheduling based on forecasting models.File | Dimensione | Formato | |
---|---|---|---|
2019_04_Berardi.pdf
non accessibile
Descrizione: Testo della tesi
Dimensione
8.62 MB
Formato
Adobe PDF
|
8.62 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/146112