Energy efficiency has become one of the most relevant issues in manufacturing above all as far as needs for a sustainable development are concerned, actually 42% of domestic energy consumption is attributable to the Italian industrial sector. Our thesis wants to be a contribution to the state of art with the development of policies aimed to energy consumption awareness that allow all the stages in the job shop system to make decisions about the energy state of the machines inside it, having the possibility of sharing and exchanging information in real time with the other stages and the system input through a proper ICT infrastructure. We have considered a job shop system because it has not been studied deeply from an energy point of view. Each stage of the system perfectly knows everything concerning its machines and its buffer, communicating with the other system components it is able to get to know what is happening in the other stages in every moment and when a job to be worked is arriving, it knows the waiting times for each job in the various buffers, the processing time and the energy state of all the machines. The stage is able to look for all the necessary information whenever it actually needs it and to decide about the energy state of the machines inside it according to the situation in that moment. Thus the stage can decide to set a machine to standby and to postpone the next job processing thus saving energy, with an impact on the makespan close to zero. The proposed energy policies help make decisions in real time according to the situation in that moment that are not based on statistical data and forecasts. Our work starts with an analysis of the literature to individuate the gaps and the research questions. The three fundamental phases are: - conceptual models definition; - models development; - experiments and analysis of the results. This has been possible using a discrete event simulation software that has allowed us to simulate productive and energy policies, implementing a personalized algorithm. These three phases have been iterated and have allowed us to clearly understand the behaviour of the system we have considered and find solutions to implement the successive cycles. Thus we have created two energy aware policies validated during the experiments and the results analysis considering both energy consumption and makespan.
Il tema dell'efficienza energetica è diventato uno dei più rilevanti nel manufacturing soprattutto in un contesto di esigenza e desiderio di uno sviluppo sostenibile. Il settore industriale in Italia, per esempio, è attualmente responsabile del 42% del consumo energetico nazionale. Con questo lavoro vogliamo dare un contributo al settore manifatturiero tramite delle politiche di efficienza energetica da poter applicare ad un sistema produttivo, in particolare ad un sistema produttivo con configurazione job shop in quanto in letteratura non è stato studiato approfonditamente da un punto di vista energetico. Grazie a queste politiche ogni reparto presente nel sistema prende decisioni riguardo allo stato energetico delle macchine al suo interno sfruttando la possibilità di condividere e scambiare informazioni in real time con gli altri componenti del sistema attraverso un'appropriata infrastruttura ICT. Ogni reparto del sistema conosce perfettamente tutto ciò che riguarda le macchine, il buffer al suo interno e comunicando con gli altri componenti del sistema può sapere in ogni momento ciò che sta accadendo negli altri reparti e quando gli arriverà un pezzo da lavorare, conosce i tempi di attesa di ogni job nei vari buffer, i tempi di lavorazione e lo stato energetico di tutte le macchine. Il reparto è in grado di cercare le informazioni che gli servono nel momento in cui ne ha effettivamente bisogno e di decidere in base alla situazione in quell'istante. Quindi il reparto può decidere di mandare una macchina in standby e di rimandare la lavorazione del job successivo ottenendo un risparmio energetico, con un impatto sul makespan prossimo allo zero. Le politiche energetiche quindi proposte permettono di prendere decisioni in tempo reale sulla base della situazione in quel momento e non basate su previsioni statistiche effettuate a partire da dati storici. Questo lavoro comincia con un'analisi della letteratura per la definizione dei gap e delle domande di ricerca. Seguono tre fasi fondamentali: - definizione dei modelli concettuali; - sviluppo dei modelli; - esperimenti e analisi dei risultati grazie all’utilizzo di un software di simulazioni a eventi discreti che permettono di simulare politiche produttive ed energetiche implementando algoritmi personalizzati. Queste tre fasi sono state iterate e questo ci ha permesso di capire a fondo il comportamento del sistema considerato e far emergere progressivamente idee e soluzioni da implementare nei cicli XIII successivi. Abbiamo quindi creato due politiche "energicamente consapevoli" testate durante gli esperimenti e l’analisi dei risultati, considerando sia il consumo energetico che il makespan.
Energy efficient policies for a self-aware job shop production system
NERVI, CECILIA;MANZONE, VALERIA
2015/2016
Abstract
Energy efficiency has become one of the most relevant issues in manufacturing above all as far as needs for a sustainable development are concerned, actually 42% of domestic energy consumption is attributable to the Italian industrial sector. Our thesis wants to be a contribution to the state of art with the development of policies aimed to energy consumption awareness that allow all the stages in the job shop system to make decisions about the energy state of the machines inside it, having the possibility of sharing and exchanging information in real time with the other stages and the system input through a proper ICT infrastructure. We have considered a job shop system because it has not been studied deeply from an energy point of view. Each stage of the system perfectly knows everything concerning its machines and its buffer, communicating with the other system components it is able to get to know what is happening in the other stages in every moment and when a job to be worked is arriving, it knows the waiting times for each job in the various buffers, the processing time and the energy state of all the machines. The stage is able to look for all the necessary information whenever it actually needs it and to decide about the energy state of the machines inside it according to the situation in that moment. Thus the stage can decide to set a machine to standby and to postpone the next job processing thus saving energy, with an impact on the makespan close to zero. The proposed energy policies help make decisions in real time according to the situation in that moment that are not based on statistical data and forecasts. Our work starts with an analysis of the literature to individuate the gaps and the research questions. The three fundamental phases are: - conceptual models definition; - models development; - experiments and analysis of the results. This has been possible using a discrete event simulation software that has allowed us to simulate productive and energy policies, implementing a personalized algorithm. These three phases have been iterated and have allowed us to clearly understand the behaviour of the system we have considered and find solutions to implement the successive cycles. Thus we have created two energy aware policies validated during the experiments and the results analysis considering both energy consumption and makespan.| File | Dimensione | Formato | |
|---|---|---|---|
|
Manzone Nervi Thesis.pdf
accessibile in internet per tutti
Descrizione: testo della tesi
Dimensione
26.14 MB
Formato
Adobe PDF
|
26.14 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/10589/122501