The manufacturing sector is a significant energy consumer and key contributor to greenhouse gas emissions. Improving energy efficiency is crucial, yet energy waste is challenging to identify due to its invisible nature and the lack of structured methodologies. Existing tools, such as Energy Value Stream Mapping (EVSM) and Energy Benchmarking, struggle to clearly distinguish energy waste, making it difficult to set feasible and practical energy-saving targets. This study focuses on two key types of energy waste in manufacturing plants recognized in the literature: energy consumption during planned non-productive times, such as weekends, holidays, and idle shifts, which is often overlooked, and energy inefficiencies in production equipment caused by variations in operators' behavior and skills. To address these challenges, this thesis presents a scalable, data-driven tool that follows a top-down approach, integrating graphical load profile analysis, internal energy benchmarking, and statistical analysis. Unlike existing methods, it relies solely on load profiles, shift plans, and production rates, making it highly scalable and applicable even in plants with limited metering infrastructure. Validated in a world-leading automotive company’s Body-in-White (BIW) manufacturing cell, the tool identified around 20% of total energy consumption occurring during planned non-productive times. Addressing variability in these times could reduce energy use by 2–3%, while optimizing idle periods within working shifts could save 1%. Increasing the cell's loading, shifting from 60% utilization to full utilization while maintaining the same production output, could further reduce energy consumption by 7%, leveraging economies of scale. In total, these measures could achieve up to 10% energy savings without productivity loss, with additional reductions possible by fully switching off machines during idle times. Even partial implementation of these optimizations would support Net Zero goals, as a 6% annual CO2 reduction is needed to keep the light industry on track for 2030 sustainability targets.
Il settore manifatturiero è tra i principali consumatori di energia e una delle maggiori fonti di emissioni di gas serra. Migliorare l’efficienza energetica è fondamentale, ma gli sprechi energetici sono difficili da individuare per la loro natura invisibile e la mancanza di metodologie strutturate. Gli strumenti esistenti, come Energy Value Stream Mapping (EVSM) e Energy Benchmarking, faticano a distinguere chiaramente gli sprechi, rendendo difficile definire obiettivi di risparmio raggiungibili. Questo studio analizza due principali tipologie di spreco energetico negli impianti manifatturieri: i consumi nei periodi non produttivi, come fine settimana, festività e turni inattivi, spesso trascurati, e le inefficienze operative delle attrezzature di produzione, dovute a variazioni nel comportamento e nelle competenze degli operatori. Per affrontare queste criticità, questa tesi propone uno strumento scalabile e basato sui dati, che segue un approccio top-down, integrando analisi dei profili di carico, benchmarking energetico interno e analisi statistica. A differenza dei metodi esistenti, si basa esclusivamente su profili di carico, organizzazione dei turni e capacità produttiva, risultando applicabile anche in impianti con infrastrutture di misurazione limitate. Validato nella cella di produzione Body-in-White di una azienda automobilistica, il tool ha rilevato che circa il 20% del consumo energetico totale avviene nei periodi non produttivi programmati. Ridurne la variabilità potrebbe portare a un risparmio del 2–3%, mentre ottimizzare le inattività nei turni di lavoro ridurrebbe i consumi di un ulteriore 1%. Aumentare il carico della cella dal 60% alla piena capacità senza modificare la produzione potrebbe ridurre i consumi del 7% grazie alle economie di scala. Complessivamente, queste misure permetterebbero un risparmio fino al 10%, con ulteriori riduzioni spegnendo i macchinari nei periodi di inattività. Anche una parziale implementazione di queste ottimizzazioni supporterebbe gli obiettivi Net Zero, poiché è necessaria una riduzione annuale del 6% delle emissioni di CO2 per mantenere l'industria leggera allineata agli obiettivi di sostenibilità del 2030.
A scalable data-driven tool for energy waste identification in a manufacturing plant: application in the automotive sector
CURIONI, GIULIO
2023/2024
Abstract
The manufacturing sector is a significant energy consumer and key contributor to greenhouse gas emissions. Improving energy efficiency is crucial, yet energy waste is challenging to identify due to its invisible nature and the lack of structured methodologies. Existing tools, such as Energy Value Stream Mapping (EVSM) and Energy Benchmarking, struggle to clearly distinguish energy waste, making it difficult to set feasible and practical energy-saving targets. This study focuses on two key types of energy waste in manufacturing plants recognized in the literature: energy consumption during planned non-productive times, such as weekends, holidays, and idle shifts, which is often overlooked, and energy inefficiencies in production equipment caused by variations in operators' behavior and skills. To address these challenges, this thesis presents a scalable, data-driven tool that follows a top-down approach, integrating graphical load profile analysis, internal energy benchmarking, and statistical analysis. Unlike existing methods, it relies solely on load profiles, shift plans, and production rates, making it highly scalable and applicable even in plants with limited metering infrastructure. Validated in a world-leading automotive company’s Body-in-White (BIW) manufacturing cell, the tool identified around 20% of total energy consumption occurring during planned non-productive times. Addressing variability in these times could reduce energy use by 2–3%, while optimizing idle periods within working shifts could save 1%. Increasing the cell's loading, shifting from 60% utilization to full utilization while maintaining the same production output, could further reduce energy consumption by 7%, leveraging economies of scale. In total, these measures could achieve up to 10% energy savings without productivity loss, with additional reductions possible by fully switching off machines during idle times. Even partial implementation of these optimizations would support Net Zero goals, as a 6% annual CO2 reduction is needed to keep the light industry on track for 2030 sustainability targets.File | Dimensione | Formato | |
---|---|---|---|
2025_04_Curioni_Executive Summary.pdf
non accessibile
Descrizione: Executive Summary
Dimensione
1.26 MB
Formato
Adobe PDF
|
1.26 MB | Adobe PDF | Visualizza/Apri |
2025_04_Curioni_Tesi.pdf
non accessibile
Descrizione: Tesi
Dimensione
30.28 MB
Formato
Adobe PDF
|
30.28 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/10589/234456