68994 research outputs found
Sort by
Human-Robot Collaboration: From Collaborative Robotics in SMEs to AI-Driven Natural Interaction
No full textLe attività presentate in questa tesi di dottorato esaminano come la robotica collaborativa possa essere introdotta nelle piccole e medie imprese per aumentare produttività e flessibilità, mantenendo al contempo gli operatori umani al centro del sistema produttivo. Il lavoro segue una progressione in tre fasi, dalle celle collaborative fisse alle piattaforme robotiche mobili e, infine, a una pipeline di presa abilitata dall’IA e guidata da modelli visione-linguaggio. La prima parte della tesi analizza una cella robotica collaborativa implementata presso una stazione di saldatura plastica a vibrazione. Combinando piccole zone di buffer, prelievo basato sulla visione e progettazione delle traiettorie guidata dalla simulazione, la cella è in grado di sfruttare i tempi morti della macchina per un funzionamento non presidiato. Lo studio mostra che un assetto collaborativo progettato con attenzione può ridurre in modo sostanziale lo sforzo manuale richiesto all’operatore senza compromettere la sicurezza o la qualità del prodotto, e può essere successivamente adattato a nuovi prodotti con una limitata riconfigurazione. La seconda parte estende la collaborazione a una piattaforma mobile che serve più stazioni in un contesto automotive. Un robot mobile collaborativo viene progettato e valutato in simulazione, redistribuendo i periodi di attesa dell’operatore in attività produttive di movimentazione. I risultati indicano che, con una pianificazione adeguata dei percorsi e delle visite alle stazioni, una singola piattaforma mobile può ridurre in misura significativa il coinvolgimento diretto dell’operatore in scenari a piccoli e medi lotti e supportare la riallocazione del lavoro umano verso attività di supervisione e ispezione. La parte finale introduce VL-GRiP3, una pipeline di presa modulare che collega istruzioni in linguaggio naturale all’esecuzione robotica combinando modelli visione-linguaggio compatti, registrazione della nuvola di punti basata su CAD e pianificazione della presa a sei gradi di libertà. L’architettura separa percezione, registrazione, predizione della presa e sintesi dell’azione in moduli aggiornabili in modo indipendente. Esperimenti su un robot UR3 con un gripper industriale dimostrano che VL-GRiP3 è in grado di eseguire in modo affidabile compiti di presa e rilascio su componenti reali prodotti industrialmente, offrendo al contempo maggiore trasparenza ed efficienza nell’uso dei dati. Considerati nel loro insieme, questi contributi mostrano che robot collaborativi, piattaforme mobili e pipeline modulari visione-linguaggio possono essere combinati per fornire un’automazione pratica e centrata sulla persona nelle piccole e medie imprese, allineando la pratica industriale ai principi dell’Industry 5.0. The activities presented in this Ph.D. thesis investigates how collaborative robotics can be introduced into small and medium sized enterprises to increase productivity and flexibility while keeping human operators at the center of the production system. The work follows a three stage progression, from fixed collaborative cells to mobile robotic platforms and, finally, to an AI enabled grasping pipeline driven by vision language models.
The first part of the thesis examines a collaborative robotic cell deployed at a vibrating plastic welding station. By combining small buffer zones, vision based picking, and simulation guided trajectory design, the cell is able to exploit machine idle time for unattended operation. The study shows that a carefully designed collaborative setup can substantially reduce the manual effort required from the operator without compromising safety or product quality, and can later be adapted to new products with limited reconfiguration.
The second part extends collaboration to a mobile platform that serves multiple stations in an automotive context. A mobile collaborative robot is designed and evaluated in simulation, redistributing periods of operator waiting time into productive handling activities. The results indicate that, with appropriate planning of routes and station visits, a single mobile platform can significantly reduce direct operator involvement in small and medium batch scenarios and support the reallocation of human work toward supervision and inspection.
The final part introduces VL-GRiP3, a modular grasping pipeline that links natural language instructions to robot execution by combining compact vision language models, CAD based point cloud registration, and six degree of freedom grasp planning. The architecture separates perception, registration, grasp prediction, and action synthesis into independently updatable modules. Experiments on a UR3 robot with an industrial gripper demonstrate that VL-GRiP3 can reliably execute grasp and place tasks on real manufactured components while offering greater transparency and data efficiency.
Together, these contributions show that collaborative robots, mobile platforms, and modular vision language pipelines can be combined to deliver practical, human centred automation in small and medium sized enterprises, aligning industrial practice with the principles of Industry 5.0
Innovative business models along the industry life cycle. Does demand turbulence matter?
No full textThe industry-life cycle (ILC) literature has long emphasised that the ILC stage provides an important context for innovation and a crucial factor for firms’ innovative dynamics. Nonetheless, this framework does not fully explain the emergence of business model innovation (BMI), which is a key competitive tool in the current business scenario. In this study, we investigate the role of demand turbulence in companies’ likelihood of operating with an innovative business model by disentangling the influence of demand from the contribution of the industry life cycle. To identify innovative models, we retrieve individual business models using information from the company websites and compare them with the “prevalent” business model emerging in the company’s sector. To capture demand turbulence, we create a sectoral demand indicator for the years 2014–2019 using product-level data on consumers’ expenditures aggregated according to the ECOICOP classification. Using a sample of 1232 Italian firms observed in 2021, we show that demand turbulence is negatively associated with business model innovation even after controlling for the life-cycle stage of the industry. This suggests that BMI mainly emerges as a response to demand stability, and that companies can successfully resort to this strategy in the presence of unfavourable demand conditions, regardless of the stage of market development
Hydraulic Asymmetries for Biological and Bioinspired Valves in Tubular Channels: A Numerical Analysis
No full textBiological, biomimetic, and engineering systems make extensive use of hydraulic asymmetries to control flow inside tubular structures. Examples span physiological valves, the guided transport observed in shark intestines, and passive devices such as Tesla valves. Here we investigate the mechanisms that generate these asymmetries using the notion of diodicity, defined as the ratio between pressure drops required to drive the same flow in opposite directions. We first focus on 2D geometries, which allow us to identify and study the main contributions to hydraulic asymmetry: channel geometry and internal obstacles embedded within a channel with rigid walls. By considering both rigid and deformable obstacles, we model channels that always remain open in both directions and channels that can be completely blocked by valve-like structures. We then extend the analysis to 3D geometries, again considering rigid and elastic cases. As a general trend, we find that geometry alone establishes a baseline diodicity, while higher dimensionality and structural reconfiguration consistently amplify the effect
Blood culture practices and microbiological capacity for sepsis diagnostics in Europe (2021–2022): a cross-sectional analysis of the European Sepsis Care Survey
Full text linkBackground: Blood cultures (BCs) are key diagnostic elements for sepsis patients. Accurate preanalytical procedures are substantial, and results should be available as soon as possible to guide adequate antimicrobial treatment. This study aimed to evaluate BC collection practices and diagnostic capacity across European hospitals. Methods: This cross-sectional survey investigated BC diagnostics in acute care hospitals across 37 European countries in the years 2021 and 2022. Analyses included BC guidelines, collection sites, number of BC sets in emergency departments (EDs), wards, and intensive care units (ICUs). We also examined transfer after collection, the use of on-site vs. external laboratories, opening hours, rapid testing capacity, and turn-around times of BCs processed in microbiology laboratories with different infrastructures. Findings: Responses were collected from 907 hospitals in Europe. BC guidelines were available in 84·4% (741/878) of the hospitals. BCs were preferably collected by multiple-site sampling in EDs (62·7%, 461/735), in wards (64·0%, 513/802) and ICUs (68·5%, 518/756). One BC set was preferred in EDs in 38·4% (270/704), in wards in 40·5% (314/775), and ICUs in 34·9% (261/748). Two BC sets were preferred in EDs in 31·0% (218/704), in wards in 28·1% (218/775), and ICUs in 39·2% (293/748). 48·0% (402/838) of hospitals used on-site and 52·0% (436/838) external microbiology laboratories. Around-the-clock microbiological services were available in 10⋅0% (91/907), and rapid pathogen identification in 43·7% (396/907) of hospitals. Infrastructure with around-the-clock microbiological service and rapid testing was available in 7·4% (62/840) of hospitals, and probability of a final microbiological result within two days was highest in these hospitals compared to hospitals with limited microbiology service (for BC collected on wards: 19·6% vs. 52·7%, Odds Ratio 4·59 [95% CI 2·50–7·79], p < 0·0001). Interpretation: Despite the availability of BC guidelines in many hospitals, current recommendations for BC collection were often neglected. Rapid testing capacity was limited in most microbiological laboratories, and around-the-clock service for BCs was very rare. As delay in results may have a detrimental impact on patient outcomes, strategies to improve these processes are urgently needed. Funding: The European Sepsis Alliance and a grant by Becton and Dickinson
Dengue and Aging: Challenges and Opportunities in Prevention and Care. A narrative review
Full text linkDengue infection is a global health issue with significantly increased incidence and overall burden, especially since 2024. Specifically, epidemiological trends show a rising median age of affected individuals over 65 years/old. Older individuals face increased risks of severe disease, extended hospital stays, healthcare-associated infections, and higher mortality rates, mainly due to a decline in immune function, and multimorbidity. Antibody-dependent enhancement, cytokine dysregulation, and endothelial dysfunction exacerbate disease severity. Moreover, in older patients, dengue diagnosis can be difficult, due to atypical symptoms. To date, there are no specific prognostic markers and no specific antiviral drugs. Management requires age-specific considerations. Evidence on immunomodulatory and antiviral therapies is emerging, and vaccine efficacy and safety data in older adults remain limited, despite growing interest. With an aging global population, dengue represents an urgent clinical challenge: there is an unmet and increasing need for comprehensive, practical guidelines to help clinicians in the diagnosis, treatment, prevention, and control of dengue infection in older patients.Trial registration: ClinicalTrials.gov identifier: NCT05611710..Trial registration: ClinicalTrials.gov identifier: NCT06579755.
Valutazione d'azienda, intelligenza artificiale e nuove tecnologie
No full textIl volume analizza l'impatto delle nuove tecnologie sul processo di valutazione d'azienda e affronta poi il tema della valutazione degli asset tecnologic
Refining the masonry shear modulus in masonry towers via Bayesian model updating
Full text linkThe assessment of the structural behavior of masonry towers often involves developing and identifying computational models to be used to perform static and/or time-history nonlinear analyses. Such models are frequently developed assuming isotropic masonry behavior, with model identification carried out - based on available experimental data - through deterministic tuning of a limited set of representative parameters. However, masonry exhibits complex structural textures that often deviate significantly from the commonly made assumption of isotropic behaviour. This paper investigates the effects of this assumption by focusing on the masonry shear modulus. A two-dimensional rigid body–spring model is adopted as computational approach, as its fully discrete formulation allows overcoming the limitations of the Cauchy continuum while maintaining a reduced computational cost. A Bayesian updating framework based on dynamic experimental data is employed to account for multiple sources of uncertainty, including parameter uncertainty, model inadequacy, and observation errors. Three masonry towers with different slenderness ratios are considered as representative case studies. The adopted Bayesian model updating approach allows for the estimation of their shear modulus while accounting for uncertainties, and the results show that the common assumption of isotropic behaviour in masonry numerical modelling does not hold. Using a fully discrete computational approach in combination with experimental frequency data, this behaviour has been observed for squat masonry towers
SUSTAINABILITY DISCLOSURE ACROSS OFFICIAL REPORTING AND SOCIAL MEDIA: DYNAMICS OF TRANSPARENCY, NARRATIVE CONTROL, AND STAKEHOLDER SCRUTINY Three studies on the organizational mechanisms connecting sustainability reporting, its digital communication, and the pursuit of legitimacy.
No full textLa comunicazione di sostenibilità si è rapidamente evoluta da una pratica orientata alla rendicontazione a un ambito strategico attraverso il quale le organizzazioni negoziano la propria legittimità, modellano l’identità pubblica e rispondono a una crescente pressione da parte degli stakeholder. L’introduzione della Corporate Sustainability Reporting Directive (CSRD) e la crescente pervasività delle piattaforme digitali hanno intensificato le aspettative in materia di trasparenza e accountability, trasformando la comunicazione di sostenibilità in un processo intrinsecamente cross-channel e fortemente esposto. Se i report di sostenibilità forniscono informazioni strutturate e regolamentate, i social media consentono interazioni partecipative, immediate e dialogiche che ampliano le opportunità di engagement ma, al contempo, amplificano i rischi reputazionali. Queste dinamiche rendono il coordinamento tra comunicazione istituzionale e comunicazione digitale una sfida centrale per la governance aziendale contemporanea, soprattutto in contesti di controversia, crisi ESG o scandali, in cui la visibilità pubblica agisce come un potente amplificatore del rischio. Nonostante la crescente attenzione accademica, la letteratura risulta ancora frammentata e non offre una comprensione integrata di come le organizzazioni governino la disclosure di sostenibilità attraverso canali diversi. Emergono tre principali gap teorici. In primo luogo, manca una spiegazione di come reporting formale e comunicazione sui social media si influenzino reciprocamente, né è chiaro come le organizzazioni gestiscano le tensioni tra trasparenza e controllo reputazionale. In secondo luogo, gli studi esistenti non hanno ancora sviluppato framework che descrivano come le imprese incorporano feedback, critiche e segnali digitali nei processi decisionali e nelle pratiche di rendicontazione, nonostante i social media rappresentino spazi fondamentali di co-costruzione identitaria con gli stakeholder. In terzo luogo, le strategie comunicative adottate durante controversie e scandali di sostenibilità restano poco teorizzate: è ancora limitata la conoscenza dei criteri con cui le organizzazioni scelgono tra disclosure, comunicazione selettiva, minimizzazione del problema o silenzio strategico, anche alla luce dei possibili effetti boomerang e della formazione di una memoria collettiva negativa persistente. Questa ricerca di dottorato affronta tali lacune sviluppando una comprensione integrata e multilivello della comunicazione di sostenibilità attraverso reporting ufficiale e social media. Il progetto indaga come le organizzazioni coordinano i due canali, come il feedback digitale influenzi la disclosure e come trasparenza e rischio reputazionale vengano bilanciati in condizioni di scrutinio e crisi. Per questo, la tesi è articolata in tre studi complementari. Lo Studio Uno fornisce il fondamento concettuale tramite una mappatura sistematica della letteratura, identificando le dinamiche upstream e downstream che collegano report e social media. Lo Studio Due testa empiricamente tali meccanismi attraverso un caso di studio longitudinale di un’azienda coinvolta in uno scandalo di sostenibilità, analizzando le strategie di comunicazione cross-channel prima, durante e dopo la crisi. Lo Studio Tre approfondisce ulteriormente queste evidenze tramite interviste aziendali, ricostruendo come ruoli, processi e contenuti vengano coordinati internamente per garantire coerenza cross-channel, controllo reputazionale e conformità ai requisiti normativi emergenti. Complessivamente, questa ricerca avanza la comprensione teorica della comunicazione di sostenibilità come processo multicanale, multi-attore e socio-tecnico, offrendo un quadro integrato per spiegare come le organizzazioni costruiscono e mantengono la propria legittimità in un ambiente sempre più trasparente, partecipativo e mediato digitalmente.
Abstract della tesi in ingleseSustainability communication has rapidly evolved from a reporting-oriented practice into a strategic domain through which organizations negotiate legitimacy, shape their public identity, and respond to growing stakeholder scrutiny. The introduction of the Corporate Sustainability Reporting Directive (CSRD) and the increasing pervasiveness of digital platforms have intensified expectations regarding transparency and accountability, transforming sustainability communication into an inherently cross-channel and highly exposed process. While official sustainability reports provide structured and regulated information, social media enable participatory, immediate, and dialogic interactions that both expand engagement opportunities and amplify reputational risks. These dynamics make the coordination between institutional communication and digital communication a central challenge for contemporary corporate governance, particularly in contexts of controversy, ESG crises, or scandals, where public visibility acts as a powerful risk multiplier. Despite growing academic attention, the literature remains fragmented and lacks an integrated understanding of how organizations govern sustainability disclosure across channels. Three main theoretical gaps emerge. First, research has not yet explained how formal reporting and social media communication influence one another, nor how organizations manage the tensions between transparency and reputational control across channels. Second, existing studies have not developed frameworks describing how firms incorporate feedback, criticism, and digital signals into internal decision-making and disclosure processes, despite social media being key spaces where corporate identity is co-constructed with stakeholders. Third, the communicative strategies adopted during sustainability controversies remain under-theorized, and little is known about how organizations choose between disclosure, selective communication, issue minimization, or strategic silence, especially considering potential backfire effects and the formation of persistent negative collective memories. This doctoral research addresses these gaps by developing an integrated, multi-level understanding of sustainability communication across official reporting and social media. The project investigates how organizations coordinate the two channels, how digital feedback influences disclosure, and how transparency and reputational risk are balanced under conditions of stakeholder scrutiny and crisis. To do so, the dissertation is structured into three complementary studies. Study One provides a conceptual foundation through a systematic mapping of the literature, identifying upstream and downstream mechanisms linking reports and social media. Study Two empirically tests these mechanisms through a longitudinal case study of a company involved in a sustainability scandal, analysing cross-channel communication strategies before, during, and after the crisis. Study Three deepens these insights through corporate interviews, reconstructing how roles, processes, and content are coordinated internally to ensure cross-channel coherence, reputational control, and compliance with emerging regulatory requirements. Overall, this research advances the theoretical understanding of sustainability communication as a multi-channel, multi-actor, and socio-technical process, offering a comprehensive framework for explaining how organizations construct and maintain legitimacy in an increasingly transparent, participatory, and digitally mediated environment
Parental bonding and attachment in the hikikomori trajectory
Full text linkAim: Hikikomori, a social withdrawal syndrome, has been suggested to be rooted in family dynamics. Early parental bonding (PB) has been linked to attachment and adulthood relationship patterns, possibly impacting the emergence of hikikomori. These outcomes have been connected to early experiences of the parents themselves, suggesting their intergenerational 'transmission'; we conducted two online cross-sectional surveys to clarify the above hypothesis. Methods: The first survey presents three groups: non-hikikomori adults (C), non-pathological hikikomori (Non-PH), and pathological hikikomori (PH); the second involved parents of individuals categorized according to the abovementioned groups. PB and attachment were evaluated through the parental bonding instrument (PBI) and Relationship Structures-Experiences in Close Relationships Scale (ECR-RS). Results: PH was associated with lower PBI 'Care', higher 'Anxious' and 'Avoidant' attachment, and the combination of 'Affectionless Control' PB and 'Fearful-Avoidant' attachment. Non-PH was linked to paternal 'Neglect', especially when combined with 'Dismissing' and 'Fearful-Avoidant' attachment. A mediation role of attachment-related 'Avoidance' between PB and hikikomori was confirmed. Parents of PH showed higher PBI 'Protection', 'Avoidant' and 'Anxious' attachment, and lower PBI 'Care': They were linked to paternal 'Affectionless Control' and 'Fearful-Avoidant' attachment. Paternal 'Neglect' was overrepresented in parents of Non-PH. Conclusions: Our results suggest that PB and attachment are involved in the appearance of hikikomori. PH may be connected to family history of 'Affectionless Control' and 'Fearful-Avoidant' attachment, whereas Non-PH may be linked to 'Neglectful' parenting, which could promote attachment-related 'Avoidance'. Specific interventions aimed at enhancing parents' sensitivity and mentalization could reduce the risk and the severity of hikikomori
Design for Environmental Sustainability through Machine Learning Based Method
No full textQuesta tesi di dottorato indaga come il Machine Learning (ML) possa supportare la progettazione per la sostenibilità ambientale integrando modelli basati sui dati con la Life Cycle Assessment (LCA). Sebbene la LCA sia una metodologia consolidata per quantificare gli impatti ambientali lungo l’intero ciclo di vita di un prodotto, dall’estrazione delle materie prime alla gestione del fine vita, la sua applicazione tradizionale risulta spesso dispendiosa in termini di tempo e poco adatta alla fase di progettazione concettuale, in cui le informazioni disponibili sono limitate ma le decisioni hanno un’influenza elevata. Il lavoro affronta questo divario sviluppando modelli surrogati basati su ML, in grado di stimare rapidamente parametri ambientali e di processo, consentendo così ai progettisti di considerare la sostenibilità insieme a prestazioni, costi e affidabilità fin dalle primissime fasi della progettazione.
La ricerca adotta e adatta la metodologia CRISP-DM (Cross-Industry Standard Process for Data Mining) come quadro di riferimento generale per strutturare l’intero flusso di lavoro: Business Understanding, Data Understanding, Data Preparation, Modeling, Evaluation e Deployment. All’interno di questo framework, viene dedicata particolare attenzione alla definizione di livelli accettabili di accuratezza dei modelli, alla selezione di metriche di errore appropriate e alla gestione dell’incertezza nei dati industriali reali. Le attività legate ai dati, quali la definizione dei confini di sistema, l’estrazione dei parametri di configurazione dai sistemi CAD/PDM, l’individuazione degli outlier tramite analisi dell’intervallo interquartile, l’aumento dei dati mediante scalatura geometrica e definizione di scenari, nonché la selezione delle caratteristiche (feature selection), sono trattate come elementi centrali per garantire la robustezza e l’interpretabilità dei modelli.
La metodologia è validata attraverso tre temi principali, ciascuno articolato in uno o più casi studio. Il primo riguarda la progettazione di veicoli elettrici a livello di assemblaggio, con confini di sistema cradle-to-grave. In questo contesto, i modelli di ML sono addestrati su dataset derivati da LCA e utilizzati per prevedere specifici impatti ambientali, supportando scelte progettuali orientate alla sostenibilità su scala di sistema. Il secondo tema si concentra su componenti manifatturieri (cradle-to-gate), tra cui un supporto per pala di turbina a vapore e un albero meccanico. Modelli analitici e diversi algoritmi di ML (ad esempio regressione lineare, alberi decisionali, random forest, alberi gradient boosted e reti neurali) sono confrontati per la previsione di parametri di processo e indicatori ambientali, evidenziando il ruolo dell’aumento dei dati e della selezione delle caratteristiche nel migliorare l’accuratezza e la sensibilità alle variazioni progettuali. Il terzo tema affronta la fase di fine vita e lo smontaggio di giunzioni meccaniche arrugginite. Campagne sperimentali supportano lo sviluppo di modelli di ML per la previsione del tempo di smontaggio e, in un’ulteriore estensione, di una Convolutional Neural Network (CNN) per la classificazione del grado di ruggine a partire da immagini, dimostrando l’applicabilità del metodo sia a problemi di regressione numerica sia di classificazione basata su immagini.
In tutte le applicazioni, l’interpretabilità dei modelli è enfatizzata attraverso l’analisi dell’importanza delle caratteristiche, consentendo ai progettisti di comprendere quali parametri geometrici e di processo influenzino maggiormente gli impatti ambientali e le prestazioni operative. I modelli risultanti sono concepiti per l’integrazione nelle infrastrutture IT esistenti (ad esempio strumenti basati su CAD o servizi web), rendendoli accessibili ai team di progettazione senza richiedere competenze avanzate in LCA o intelligenza artificiale. Nel complesso, la tesi dimostra che la combinazione di ML e LCA all’interno di un framework basato su CRISP-DM può trasformare la valutazione ambientale da una fase di verifica tardiva a uno strumento proattivo di supporto alle decisioni. L’approccio proposto consente decisioni progettuali più rapide, informate e sostenibili ed è scalabile a diversi obiettivi, inclusa la progettazione orientata ai costi. Questo lavoro contribuisce pertanto con una metodologia generale, flessibile e interpretabile per integrare l’intelligenza artificiale nei processi di progettazione industriale, favorendo sostenibilità e circolarità.This PhD thesis investigates how Machine Learning (ML) can support design for environmental sustainability by integrating data-driven models with Life Cycle Assessment (LCA). While LCA is a consolidated methodology for quantifying environmental impacts across a product’s life cycle, from raw material extraction to end-of-life management, its traditional application is often time-consuming and poorly suited to the conceptual design phase, where information is limited but decisions are highly influential. The work addresses this gap by developing ML-based surrogate models capable of rapidly estimating environmental and process-related parameters, thereby enabling designers to consider sustainability alongside performance, cost, and reliability from the earliest stages of design.
The research adopts and adapts the CRISP-DM (Cross-Industry Standard Process for Data Mining) methodology as a general framework to structure the entire workflow: Business Understanding, Data Understanding, Data Preparation, Modeling, Evaluation, and Deployment. Within this framework, specific attention is devoted to defining acceptable model accuracy, selecting appropriate error metrics and addressing uncertainty in real industrial data. Data-related activities, such as system boundary definition, configuration parameter extraction from CAD/PDM systems, outlier detection via interquartile range analysis, data augmentation through geometric scaling and scenario definition, and feature selection, are treated as central enablers of model robustness and interpretability.
The methodology is validated through three main themes, each articulated into one or more case studies. The first concerns the design of electric vehicles at the assembly level, with cradle-to-grave system boundaries. Here, ML models are trained on LCA-derived datasets and used to predict specific environmental impact, supporting sustainability-oriented design choices at the system scale. The second theme focuses on manufactured components (cradle-to-gate), including a steam turbine blade carrier and a mechanical shaft. Analytical models and multiple ML algorithms (e.g., linear regression, decision trees, random forests, gradient boosted trees, and neural networks) are compared for predicting process parameters and environmental indicators, highlighting the role of data augmentation and feature selection in improving accuracy and sensitivity to design variations. The third theme addresses the end-of-life stage and the disassembly of rusted mechanical joints. Experimental campaigns support the development of ML models for predicting disassembly time and, in a further extension, a Convolutional Neural Network (CNN) for classifying rust degree from images, demonstrating the applicability of the method to both numerical regression and image-based classification. Across all applications, model interpretability is emphasized through feature importance analysis, enabling designers to understand which geometric and process parameters most significantly influence environmental impacts and operational performance. The resulting models are conceived for integration into existing IT infrastructures (e.g., CAD-based tools, web services), making them accessible to design teams without requiring advanced expertise in LCA or AI. Overall, the thesis shows that combining ML with LCA within a CRISP-DM-based framework can transform environmental assessment from a late verification step into a proactive decision-support tool. The proposed approach supports faster, more informed, and more sustainable design decisions, and it is scalable to different objectives, including cost-oriented design. This work therefore contributes a general, flexible, and interpretable methodology for embedding artificial intelligence into industrial design processes to foster sustainability and circularity