Politecnio die Bari - Catalogo di prodotti della Ricerca
Not a member yet
36616 research outputs found
Sort by
A Set of Criteria for the Assessment of the Suitability of a Building to be Converted for Use as a Healthcare Facility
No full textThe COVID-19 pandemic highlighted the structural criticality of health systems at the international level and acted as a catalyst for their transformation. In Italy, under Mission 6 - Health, the NRRP envisages the strengthening of territorial healthcare by constructing two types of facilities: Community Homes Facilities (CHFs) and Community Hospitals (CHFs). The regulations allow these healthcare facilities to be built from scratch or by reusing existing buildings, an option that fits into a circular economy framework that favors urban regeneration processes and limits land consumption. The selection of buildings suitable for conversion is a complex decision-making process that requires a rigorous methodological approach capable of integrating both the intrinsic characteristics of the buildings and the extrinsic characteristics of the territorial context. This research, through a systematic review of the literature and regulatory frameworks, proposes five evaluation criteria to assess the degree of suitability of a building to undergo a re-functionalization process in favor of CHFs. The adoption of this methodological framework provides evidence-based support to policymakers and urban planners, facilitating informed site selection and contributing to an effective and sustainable reorganization of territorial healthcare
Exploring Explainability in Federated Learning: A Comparative Study on Brain Age Prediction
No full textPredicting brain age from neuroimaging data is increasingly used to study aging trajectories and detect deviations linked to neurological conditions. Machine learning models trained on large datasets have shown promising results, but data privacy regulations and the challenge of sharing medical data across institutions limit the feasibility of centralized training. Federated Learning (FL) offers a solution by allowing multiple sites to collaboratively train a model without sharing raw data. However, it remains unclear how FL affects the explainability of these models, raising concerns about the consistency and reliability of their predictions.
In this study, we analyze the consistency of model explanations between centralized and federated training paradigms. Using DeepSHAP we compare feature attributions in brain age prediction models trained on the multi-site, publicly available OpenBHB dataset. We examine the impact of how data is distributed across sites (IID vs. non-IID), the number of sites participating per training round (sampling rate), and different FL aggregation methods (FedAVG, FedProx).
Our findings show that federated models provide different explanations compared to centralized models, even when trained on the same data and task. Non-IID data distributions reduce the consistency of explanations, while including a larger number of sites per training round improves stability. Interestingly, some federated models trained on non-IID data capture biologically meaningful patterns of brain aging even more effectively than centralized models. These results suggest that careful choices in how data is distributed and how training is conducted in FL can impact model accuracy and interpretability
Earth Observation Big Data for Soil Moisture Estimation Techniques in Precision Viticulture
No full textGrapevine cultivation is one of the most relevant economic drivers of the Mediterranean basin, benefiting from favorable climate conditions, characterized by dry summers and wet winters. However, climate change poses new challenges that must be addressed and, thus, traditional farming expertise is no longer sufficient to ensure optimal wine productivity and quality. Among the key factors influencing grapevines, soil moisture plays a crucial role since it affects plant health, grape composition, and overall wine quality. Optimizing water management through advanced monitoring techniques is therefore essential for enhancing vineyard sustainability. The geomatic techniques, leveraging geospatial big data and information and communications technology, have emerged as powerful tools for monitoring vineyard health and optimizing wine productivity, quality, and sustainability. This paper presents the latest technological advancements in vineyard monitoring for soil moisture assessment, providing a comprehensive analysis of their strengths and weaknesses. The research methodology is structured into two key sections: the former focuses on monitoring technologies, while the latter describes the effectiveness and efficiency of these approaches, with particulat emphasis on machine learning applications and data fusion strategies to improve accuracy and decision-making in vineyard management. Up to now, most approaches have prioritized data acquisition and dissemination, leaving their full potential underexplored. This study outlines that integrating advanced earth observation techniques can lead to more data-driven, efficient, and sustainable viticulture
Exploring the Influence of Operator Features on the Performance of Maintenance Tasks: Insight from Industry Experts
No full textThe Industry 5.0 (I5.0) paradigm promotes a human-centric approach to advanced manufacturing by integrating worker well-being with operational performance. Maintenance activities, while not directly value-adding, are crucial to ensuring system reliability and safety. However, the transition to Maintenance 4.0, characterised by the adoption of advanced technologies, has significantly increased the cognitive demands placed on operators. In this regard, it is necessary to develop models and methodologies that allow activities to be assigned to operators by jointly considering the characteristics of the activities to be performed and those of the operator, to ensure their psychophysical well-being. To this concern, this study investigates the operator features most critical for successfully executing maintenance tasks supported by advanced technologies from an I5.0 perspective. Sixteen semi-structured interviews were conducted with experts from various manufacturing sectors to identify key operator skills for technology-supported maintenance under I5.0. The findings highlight fault diagnosis and repair as the most cognitively demanding tasks, requiring the highest levels of competence. Across all maintenance categories, professional training and compliance with technical and safety standards were valued more than formal education. Tasks such as precision cleaning, component replacement, and electrical isolation required significantly more advanced skills than general cleaning or reconnection. Moreover, the importance of cognitive and manual abilities, such as memory, dexterity, and soft skills, varied across task types. These insights support the development of training and task allocation models that better align with I5.0 principles, improving both performance and operator well-being
SPH modelling of wave attenuation by an array of submerged resonators and vorticity generation mechanism
Full text linkThe present research aims to numerically analyze the attenuation of waves by a novel device designed to address coastal erosion through an innovative and environmentally friendly approach. The device consists of an array of submerged resonators inspired by
the concept of metamaterial wave control. Through their oscillatory movement induced by wave action, these resonators achieve significant wave attenuation driven by viscous dissipation mechanisms. However, the study of metamaterials in the field of water waves remains challenging due to its complexity. Further studies are required to refine the scaling and improve the correspondence to natural beach conditions and to a deeper understanding of the intrinsic (e.g. broad-banded sea) and practical (e.g. mooring, navigation, durability, local scouring processes) limitations. This numerical study shows that, with appropriate
particle resolution, the coupling between DualSPHysics and MoorDyn executed on a GPU architecture can accurately predict the motion of moored floating structures when they interact with the free surface making it a useful method for modeling these problems. We investigated the vorticity generation mechanism related to the motion of the wave and the cylinders. The vorticities near the cylinders are shown to be closely related to the motion
of both the waves and the cylinders, with the maximum vorticities being enhanced by the natural vorticity of the moving wave. Correspondences emerge between the normalized frequency spectra of the cylinder surges and the vorticities on both sides of the cylinder. Instead, the movement of the cylinders also creates a wake behind the cylinders, which has a tendency to spread downward
Enhancing Motor Function and Quality of Life Combining Advanced Robotics and Biomechatronics in an Adult with Dystonic Spastic Tetraparesis: A Case Report
Full text linkThis case report explores the innovative integration of robotic and biomechatronic technologies, including the Motore and Ultra+ devices and neuro-suits, in a 10-session rehabilitation program for a young adult with dystonic spastic tetraparesis. Notable improvements were observed in upper limb motor function, coordination, and quality of life as measured by an increase of 18 pints on the Fugl-Meyer scale and a 25% improvement in the Bartle Index. Range of motion measurements showed consistent improvements, with task execution times improving by 10 s. These findings suggest the potential of combining wearable, robotic, and biomechatronic systems to enhance neurorehabilitation. Further refinement of these technologies might support clinicians in maximizing their integration in therapeutics, despite technical issues like synchronization issues that must be overcome
Family CEO and radical innovation: A stewardship perspective
No full textThis article integrates the literature on radical innovation, the stewardship perspective, and family business research to develop and test a model examining the influence of a family CEO and the CEO's generational stage on radical innovation, considering different types of family CEOs as distinct manifestations of strategic leaders' stewardship behavior. Furthermore, building on the notion of “doing more with less”, we propose and empirically test the notion of “doing better with less”—specifically, whether the presence of a family CEO enhances the pursuit of radical innovation under resource constraints (i.e., with lower R&D intensity). Using longitudinal data over an 11-year period from 227 listed firms in the automotive and pharma/biotech industries from 29 countries, we find that firms led by a family CEO, especially those led by descendants, excel at radical innovation. Descendant-led firms are also better at radical innovation with lower R&D intensity, suggesting they do better with less. That is, our study shows that family CEOs at a later generational stage serve as catalysts for radical innovation, even under resource constraints. In addition to implications for theory and practice, our findings offer a more advanced understanding of the strategic leadership-innovation relationship in terms of distinct manifestations of stewardship behavior for radical innovation in firms with family leadership
Inflow–Outflow Behaviour of a Coastal Karst Aquifer Based on 3D Geostatistical Reconstruction of the Thermal Field
No full textThe spatiotemporal patterns of groundwater temperature may effectively delineate groundwater flow systems and help to identify aquifer recharge areas and preferential flow pathways. In coastal aquifers, they may also offer valuable insights into the spatial extent of seawater intrusion and saltwater upconing. Applying simple Kriging interpolation and variography techniques on a high-density three-dimensional temperature dataset derived from groundwater temperature–depth profiles has enabled the reconstruction of the three-dimensional thermal field for the southernmost part of the Salento coastal karst aquifer (Southern Italy). This region shows structural complexity, which poses challenges for conceptual modelling assessment. The 3D temperature model produced is a groundbreaking reconstruction derived from field data that highlights crucial insights into a shallow hydrogeological environment. Given the hydrogeological complexity and the regional scale of the aquifer, which pose challenges to straightforward groundwater flow modelling, the information on temperature distribution from maps and cross sections of the three-dimensional thermal field emerges as a pivotal tool in identifying crucial hydrogeological features. This study, bolstered by geological, geomorphological, and structural data, demonstrates that the analysis of the groundwater thermal field, which encapsulates information about aquifer permeability heterogeneity and anisotropy, is instrumental in deducing the hydraulic behaviour of faults and revealing aquifer properties. From a geostatistical perspective, this study underscores the comprehensive nature of the 3D Kriging model: it incorporates all available groundwater temperature data from all explored depths, resulting in temperature maps that show a more accurate spatial distribution than those created by Kriging within ± 2 m of selected depths
An ML-based framework for predicting prestressing force reduction in reinforced concrete box-girder bridges with unbonded tendons
No full textThe paper presents a machine learning (ML) based framework to predict the prestressing force reduction in prestressed reinforced concrete (PSC) box-girder bridges with unbonded tendons. In the field of road network safety, the reliable assessment of some bridge typologies, such as PSC box-girder bridges, depends on different aspects, among which the inaccessibility of internal unbonded tendons, the difficulty in measuring the effective prestressing force reduction over time, the design of an efficient structural health monitoring (SHM) system. To address the above issues, the proposed approach exploits the results of experimental tests on a scaled PSC box-girder to validate a nonlinear modelling strategy and, in turn, to generate a sample dataset for training different ML algorithms. To ensure generalizability of the proposed ML model, the variability of several parameters, including geometrical and mechanical properties, was accounted for. The obtained results, evaluated in terms of statistical metrics and through an eXplainability approach, revealed that the proposed surrogate model is able to predict the prestressing force reduction for this bridge typology, knowing the current prestressing force, the elastic modulus of the concrete, and the strain variation in specific cross-sections of the structure. The application of the framework on a scaled PSC box-girder experimentally tested, demonstrated its suitability for: i) estimating the prestressing force reduction without employing periodic and expensive onsite tests; and ii) providing the best strategy for employing a sensor-based SHM system
Sliding Viscoelastic Contacts: Reciprocating Adhesive Contact Mechanics and Hysteretic Loss
No full textThis study investigates the reciprocating motion of a rigid Hertzian indenter on a viscoelastic substrate with adhesion, using a finite element-based numerical model. An innovative methodology is employed to transform the sliding contact problem into an equivalent normal contact problem, enabling the accurate simulation of adhesion effects at the contact interface. The results reveal that system behaviour is governed by the interplay between viscoelasticity and adhesion, leading to notable changes in contact pressure distribution, contact area, and energy dissipation during reciprocating motion. Specifically, viscous dissipation within the substrate material dominates at intermediate sliding speeds, where the interaction between adhesion and viscoelastic relaxation processes results in pronounced hysteresis cycles. In contrast, at low and high sliding speeds (corresponding to the rubbery and glassy regions, respectively), the material behaviour is predominantly elastic, and no hysteresis is observed. Adhesion influences contact pressure distribution and contact size, particularly in the transition regime, where its effects on viscous dissipation are measurable. Moreover, the study clarifies that adhesion alone does not induce hysteresis in elastic regimes, distinguishing reciprocating contact from normal contact, where adhesive hysteresis is typically observed. New insights are also provided into how adhesion and viscoelasticity jointly impact tribological performance, offering a deeper understanding of energy dissipation mechanisms and contact mechanics during motion reversal. Interestingly, our results also show that there is a lag period after motion reversal, where friction aligns with motion direction before eventually changing direction as pressure redistribution occurs within the system. This phenomenon highlights how changes in contact mechanics affect local tribological interactions and can lead to variations in overall system response