The oil price shock occurred in 2014 is just one of the last events which affected the Oil and Gas industry, and that are jeopardizing enterprises operating in this field. This situation involves also Engineering, Procurement, and Construction contractors, which are forced to review their processes, in order to be more efficient and competitive. Digitalization makes new technologies available to these companies, enabling them to reach this objective. More, thanks to these technologies, EPC contractors are now ready to exploit in full benefits of a modular approach to engineering and realization of process plants. The aim of this thesis is to explore the possibility to use Big Data Analytics to improve EPC Oil and Gas contractors’ process efficiency. With this purpose, a methodology based on the use of historical data generated in past projects is proposed, in order to support design activities. The methodology exploits concepts as the ones of data lake, machine learning and modularization, and it can underlie software tools and IT solutions in different phases of the EPC process. In collaboration with the Italian EPC Saipem, the methodology has been developed based on the empirical case of the Urea-Ammonia plant. After a detailed analysis of the literature, a meticulous activity of data-gathering has been performed, extending the scope of the research beyond engineering parameters, touching all the aspects of the system in object, and adopting a life-cycle perspective. By mean of dynamic parametric design, insightful relationships between data have been found, interpreted, and validated; these relationships are essential to identify and size modules which compose the plant, and they realize the potential contained in historical data. Use cases regarding the application of the methodology have been recognized, thanks to interviews to Saipem’s engineers and managers. These use cases have been also evaluated based on their technological readiness level, business impact and required effort in terms of change-management. Moreover, additional use cases outside the boundaries of Saipem have been identified and discussed. Further expansions of the applications based on the methodology may bring to increasing improvements in the EPC process and to the complete realization of modular engineering.
Il crollo del prezzo del petrolio avvenuto nel 2014 è solo uno degli ultimi eventi che hanno interessato l’industria petrolchimica e che stanno seriamente mettendo in discussione le società che operano in questo settore. Questa situazione coinvolge anche i contraenti EPC, i quali sono obbligati a rivedere i propri processi, in modo da diventare più efficienti e competitivi. La digitalizzazione mette a disposizione di queste aziende nuove tecnologie, permettendo loro di raggiungere questo obiettivo. Inoltre, grazie a tali tecnologie, i contraenti EPC sono in grado di beneficiare a pieno di un approccio modulare alla realizzazione dei progetti. Lo scopo di questa tesi è quello di esplorare la possibilità di utilizzare Big Data Analytics per migliorare l’efficienza del processo adottato dai contraenti EPC nell’industria petrolchimica. Per perseguire tale obiettivo, viene proposta una metodologia basata sull’utilizzo di dati storici generati durante i progetti passati, al fine di supportare le attività di sviluppo dei nuovi progetti. La metodologia sfrutta i concetti di data lake, machine learning e modularizzazione, e può essere posta alla base di strumenti software e soluzioni IT in differenti fasi del processo EPC. In collaborazione con l’azienda italiana Saipem, la metodologia è stata sviluppata sulla base del caso di studio riguardante l’impianto Ammoniaca-Urea. A seguito di una dettagliata analisi della letteratura, è stata svolta una meticolosa attività di raccolta dati, estendendo lo scopo della ricerca oltre i soli parametri di ingegneria, coinvolgendo tutti gli aspetti del sistema in oggetto e abbracciando il suo completo ciclo di vita. Utilizzando logiche di progettazione parametrica dinamica, rilevanti relazioni tra i dati sono state individuate, interpretate e validate; tali relazioni sono essenziali al fine di identificare e dimensionare i moduli che compongono l’impianto, ed inoltre esse realizzano il potenziale contenuto nei dati storici. Casi d’uso relativi all’applicazione della metodologia sono stati individuati, grazie ad interviste agli ingegneri ed ai dirigenti di Saipem. Questi casi d’uso sono stati valutati sulla base del loro grado di realizzabilità tecnologica, impatto sul business e sforzo in termini di change-management. In aggiunta, ulteriori casi d’uso al di fuori del perimetro di Saipem sono stati identificati e discussi. Future espansioni delle applicazioni basate sulla metodologia possono portare a miglioramenti ulteriori nel processo EPC ed alla completa realizzazione dell’ingegneria modulare.
Big data analytics for modular plant design
MARINI, PAOLO;RADICIONI, ANDREA
2017/2018
Abstract
The oil price shock occurred in 2014 is just one of the last events which affected the Oil and Gas industry, and that are jeopardizing enterprises operating in this field. This situation involves also Engineering, Procurement, and Construction contractors, which are forced to review their processes, in order to be more efficient and competitive. Digitalization makes new technologies available to these companies, enabling them to reach this objective. More, thanks to these technologies, EPC contractors are now ready to exploit in full benefits of a modular approach to engineering and realization of process plants. The aim of this thesis is to explore the possibility to use Big Data Analytics to improve EPC Oil and Gas contractors’ process efficiency. With this purpose, a methodology based on the use of historical data generated in past projects is proposed, in order to support design activities. The methodology exploits concepts as the ones of data lake, machine learning and modularization, and it can underlie software tools and IT solutions in different phases of the EPC process. In collaboration with the Italian EPC Saipem, the methodology has been developed based on the empirical case of the Urea-Ammonia plant. After a detailed analysis of the literature, a meticulous activity of data-gathering has been performed, extending the scope of the research beyond engineering parameters, touching all the aspects of the system in object, and adopting a life-cycle perspective. By mean of dynamic parametric design, insightful relationships between data have been found, interpreted, and validated; these relationships are essential to identify and size modules which compose the plant, and they realize the potential contained in historical data. Use cases regarding the application of the methodology have been recognized, thanks to interviews to Saipem’s engineers and managers. These use cases have been also evaluated based on their technological readiness level, business impact and required effort in terms of change-management. Moreover, additional use cases outside the boundaries of Saipem have been identified and discussed. Further expansions of the applications based on the methodology may bring to increasing improvements in the EPC process and to the complete realization of modular engineering.| File | Dimensione | Formato | |
|---|---|---|---|
|
2018_10_Marini_Radicioni.pdf
non accessibile
Descrizione: Thesis text
Dimensione
13.66 MB
Formato
Adobe PDF
|
13.66 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/10589/142902