The growing focus on environmental sustainability has led industries to adopt methodologies aimed at assessing and reducing their impact on the ecosystem, including Life Cycle Assessment (LCA). In the furniture sector, however, the integration of sustainable approaches is complex due to craftsmanship, high-quality materials and customization. Small and medium-sized enterprises in particular face difficulties in implementing complex valuation methods due to the costs and technical expertise required. To address these needs, this research proposes an integrated eco-design framework that combines simplified LCA methodologies and machine learning tools to support design decisions in the early stages. The objective is to facilitate the integration of sustainability before fundamental design choices, thanks to a modular approach that makes it easier to evaluate different materials and configurations, while maintaining the required quality standards. The framework consists of three main tools: the Qualitative Feedback Tool, which collects and analyses qualitative feedback from stakeholders; the Eco-Design Tool, which offers guidelines based on machine learning algorithms; and the Simplified LCA Tool, able to provide a rapid assessment of the environmental impact. The validation takes place through a case study at a leading company in the sector, focused on the design of a sustainable seat archetype. The results show an improvement in decision-making efficiency and a reduction of late corrective actions. Ultimately, the proposed approach demonstrates the importance of integrating sustainability at the design stage, overcoming the limitations of ex post evaluations. The combination of machine learning and eco-design provides a data-driven and easily implementable model, allowing companies to align with emerging environmental regulations and growing market expectations for sustainable products.
La crescente attenzione verso la sostenibilità ambientale ha indotto le industrie ad adottare metodologie mirate a valutare e ridurre il proprio impatto sull’ecosistema, tra cui la Valutazione del Ciclo di Vita (LCA). Nel settore dell’arredamento, tuttavia, l’integrazione di approcci sostenibili risulta complessa a causa dell’artigianalità, dei materiali di alta qualità e della personalizzazione. Le piccole e medie imprese, in particolare, incontrano difficoltà nell’implementare metodi di valutazione articolati a causa dei costi e delle competenze tecniche richiesti. Per rispondere a tali esigenze, questa ricerca propone un framework di eco-design integrato, che unisce metodologie LCA semplificate e strumenti basati sul machine learning a supporto delle decisioni progettuali nelle fasi iniziali. L’obiettivo è favorire l’integrazione della sostenibilità prima delle scelte di design fondamentali, grazie a un approccio modulare che renda più agevole valutare differenti materiali e configurazioni, mantenendo gli standard di qualità richiesti. Il framework prevede tre strumenti principali: il Qualitative Feedback Tool, che raccoglie e analizza riscontri qualitativi di stakeholder; l’Eco-Design Tool, che offre linee guida basate su algoritmi di machine learning; e il Simplified LCA Tool, in grado di fornire una valutazione rapida dell’impatto ambientale. La validazione avviene attraverso uno studio di caso presso un’azienda leader del settore, focalizzato sulla progettazione di un archetipo di seduta sostenibile. I risultati evidenziano un miglioramento nell’efficienza decisionale e una riduzione degli interventi correttivi tardivi. In definitiva, l’approccio proposto dimostra l’importanza di integrare la sostenibilità sin dalla fase di progettazione, superando le limitazioni delle valutazioni ex post. L’unione di machine learning ed eco-design fornisce un modello basato sui dati e facilmente implementabile, consentendo alle imprese di allinearsi alle normative ambientali emergenti e alle crescenti aspettative del mercato in termini di prodotti sostenibili.
An integrated Eco-Design Framework for the Furniture Sector: Machine Learning-Driven Guidelines and Tools for Sustainable Innovation
SARTINI, MIKHAILO
2025
Abstract
The growing focus on environmental sustainability has led industries to adopt methodologies aimed at assessing and reducing their impact on the ecosystem, including Life Cycle Assessment (LCA). In the furniture sector, however, the integration of sustainable approaches is complex due to craftsmanship, high-quality materials and customization. Small and medium-sized enterprises in particular face difficulties in implementing complex valuation methods due to the costs and technical expertise required. To address these needs, this research proposes an integrated eco-design framework that combines simplified LCA methodologies and machine learning tools to support design decisions in the early stages. The objective is to facilitate the integration of sustainability before fundamental design choices, thanks to a modular approach that makes it easier to evaluate different materials and configurations, while maintaining the required quality standards. The framework consists of three main tools: the Qualitative Feedback Tool, which collects and analyses qualitative feedback from stakeholders; the Eco-Design Tool, which offers guidelines based on machine learning algorithms; and the Simplified LCA Tool, able to provide a rapid assessment of the environmental impact. The validation takes place through a case study at a leading company in the sector, focused on the design of a sustainable seat archetype. The results show an improvement in decision-making efficiency and a reduction of late corrective actions. Ultimately, the proposed approach demonstrates the importance of integrating sustainability at the design stage, overcoming the limitations of ex post evaluations. The combination of machine learning and eco-design provides a data-driven and easily implementable model, allowing companies to align with emerging environmental regulations and growing market expectations for sustainable products.File | Dimensione | Formato | |
---|---|---|---|
tesi_sartini.pdf
embargo fino al 05/11/2026
Dimensione
4.2 MB
Formato
Adobe PDF
|
4.2 MB | Adobe PDF |
I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/20.500.14242/215243
URN:NBN:IT:UNIVPM-215243