10081 research outputs found
Sort by
Models and algorithms for routing and scheduling optimization problems
Get PDFThis thesis encompasses three distinct yet interrelated works, each contributing to the field of optimization.
In the first work, an effective heuristic algorithm tackles the Capacitated Vehicle Routing Problem, particularly addressing large-scale instances. The algorithm employs a combination of local search and restricted Set Partitioning problem optimization, leveraging Helsgaun's LKH-3 algorithm for the local search phase. Notably, this approach consistently enhances solutions available on the CVRPLIB website.
The second work delves into the extension of the FILO framework, initially designed for the Capacitated Vehicle Routing Problem. The objective is two-fold: to be competitive with state-of-the-art algorithms for simultaneous pickup and delivery problems, and to efficiently solve very large benchmark instances with numerous customers, all while maintaining linear scalability concerning problem size. A rigorous computational study validates the success in achieving both objectives.
The third work centers on PLATiNO, a Synthetic Aperture Radar Earth observation satellite. Efficient activity planning is essential to maximize the satellite's potential while adhering to platform constraints. A genetic algorithm, combined with repair procedures and local search operators, addresses the intricacies of this planning. Additionally, Mixed Integer Linear Programming formulations are utilized to provide precise estimations of optimal solution values. Extensive testing on real-world benchmark instances demonstrates the algorithm's proficiency in computing near-optimal solutions within practical time limits
Decoding complex flow phenomena via Dynamic Mode Decomposition
Get PDFIn this thesis, the viability of the Dynamic Mode Decomposition (DMD) as a technique to analyze and model complex dynamic real-world systems is presented. This method derives, directly from data, computationally efficient reduced-order models (ROMs) which can replace too onerous or unavailable high-fidelity physics-based models. Optimizations and extensions to the standard implementation of the methodology are proposed, investigating diverse case studies related to the decoding of complex flow phenomena. The flexibility of this data-driven technique allows its application to high-fidelity fluid dynamics simulations, as well as time series of real systems observations. The resulting ROMs are tested against two tasks: (i) reduction of the storage requirements of high-fidelity simulations or observations; (ii) interpolation and extrapolation of missing data. The capabilities of DMD can also be exploited to alleviate the cost of onerous studies that require many simulations, such as uncertainty quantification analysis, especially when dealing with complex high-dimensional systems. In this context, a novel approach to address parameter variability issues when modeling systems with space and time-variant response is proposed. Specifically, DMD is merged with another model-reduction technique, namely the Polynomial Chaos Expansion, for uncertainty quantification purposes. Useful guidelines for DMD deployment result from the study, together with the demonstration of its potential to ease diagnosis and scenario analysis when complex flow processes are involved
Border archives. The effects of privatizations on the Romagna Acque Società delle Fonti S.p.A. archive.
Get PDFLa ricerca condotta intende analizzare gli effetti prodotti dal fenomeno delle privatizzazioni sugli archivi di enti pubblici divenuti società per azioni con particolare riferimento alle implicazioni per il ruolo dell’amministrazione archivistica, ponendosi l’obiettivo di indicare anche delle soluzioni per l’archivio di Romagna Acque Società delle Fonti S.p.A., società a totale capitale pubblico che gestisce tutte le fonti idropotabili della Romagna, inclusa la Diga di Ridracoli.
Ripercorrendo il contesto storico-istituzionale in cui i cc.dd archivi ‘privatizzati’ si sono sviluppati e la normativa che ne ha accompagnato nel corso del tempo la formazione e la tutela, lo studio si concentra su questa tipologia archivistica sui generis, la cui peculiarità deriva già dalla stessa natura del ‘soggetto produttore’, il quale in molti casi appartiene, congiuntamente, tanto al settore pubblico quanto a quello privato.
L’esigenza di un'analisi più completa ha suggerito di non avere un unico campione di riferimento nell’esperienza archivistica di Romagna Acque, bensì di considerare altri esempi di realtà aziendali sensibili al tema - nella fattispecie ACEA S.p.A. e Acquedotto Pugliese S.p.A. - in un'ottica di utili raffronti in chiave comparatistica tra diverse realtà e contesti di sviluppo.This research intends to analyze the effects produced by the phenomenon of privatization on the archives of public bodies that have become joint-stock companies, with particular reference to the implications for the role of archival administration, with the aim of also indicating solutions for the Romagna Acque Società delle Fonti S.p.A. archive, a fully public capital company that manages all the drinking water sources in Romagna, including the Ridracoli Dam.
Retracing the historical-institutional context in which the so-called 'privatized' archives developed and the legislation that accompanied their formation and protection over time, the study focuses on this sui generis archival typology, whose peculiarity it already derives from the very nature of the 'producing subject', which in many cases belongs, jointly, to both the public and private sectors.
The need for a more complete analysis suggested not having a single reference sample in the archival experience of Romagna Acque, but rather considering other examples of corporate entities sensitive to the topic - in this case ACEA S.p.A. and Acquedotto Pugliese S.p.A. - with a view to useful comparative comparisons between different realities and development contexts
Designing an Innovetive Modular Platform for Sports Car Using the Generative Design Method
Get PDFTraditional methods, where chassis components are tailored for each vehicle type, lack flexibility and efficiency. The concept of current modular platforms, allows the reuse of components across different models, reducing production costs and enhancing adaptability. But, in current situation these solutions are not common in sports cars segment. The research delves into the challenges and opportunities posed by modular platforms in the context of sports cars, highlighting their potential impact on driving dynamics, design aesthetics, and future innovations. The project focuses on a modular platform approach, providing diversity while maintaining a standardized design sections, emphasizing interchangeability of components besides flexibility, using cutting-edge design methods. This study addresses to create a modular platform suitable for different drivetrain and powertrain configurations, with iterative sprints targeting lightweight and high-stiffness designs by using generative design method. In addition to improving design outcomes, efforts have been made to enhance creativity by employing the steps of the generative design method within the existing workflow (IDeS), and collaboration with the Agile method variant, Scrum, has been established to filter the results, which is crucial for project development. Moreover, it has been applied to an alternative modular platform created with new parts obtained through the generative design application. The results obtained have been evaluated in terms of the model's mechanical properties. These new parts are not only geometrically more efficient but also capable of yielding the same mechanical results even when different materials are used. The parts defined to generate decided with crash tests on rear-mid and front modular platform layouts separately. The results have been compared and it has found that stress distributions are similar which means the parts that we have generated are sufficient in new design such as shapes, weight, and mechanical properties
Pseudo-operations and pseudo-analysis: applications to probability and risk modelling
Get PDFThe aim of this thesis is to generalize some well-known concepts in probability and statistics from the standard ring to the non-idempotent semi-ring (R+, ⊕h, ⊗h), where pseudo-sum ⊕h and ⊗h are defined by a suitable function h, called generator. In the first part of this dissertation, we introduce the notions of pseudo-independence with respect to a pseudo-additive fuzzy measure and of pseudo-moment generating functions, showing that the classical results concerning moment generating functions of a vector of independent random variables and of their sum extend to pseudo-moment generating functions if the random variables involved are pseudo-independent. Moreover, we prove that pseudo moment generating functions, and more in general pseudo-analysis, can be particularly efficient in characterizing a new class of bivariate random vectors that we call ”pseudo-Schur constant” family, which represents an extension of the well-known Schur-constant class. In the second part of this thesis, we give a generalization of strong and weak bivariate lack-of-memory properties, substituting into their associated functional equations the standard product by the pseudo one ⊗h: we call the distributions satisfying them pseudo strong and weak distributions. After characterising the pseudo weak distribution in full generality, we study the induced dependence structure of the underlying lifetimes and that of the residual ones. Moreover, we show that the distributions satisfying pseudo lack-of-memory properties coincide with the solutions of suitable generalizations of Kaminsky (1983) and Marshall and Olkin (2015) functional equations; finally, we analyse several examples of pseudo weak distributions that may be used in life insurance and we give a non-life insurance application to LOSS and ALAE modelling problem
The structure and evolution of Gamma-Ray Bursts: mapping explosive transients at high angular resolution
Get PDFGamma-Ray Bursts (GRBs) are brief flashes of gamma-rays that last from a fraction of second up to a few hundreds of seconds, during which a significant amount of isotropic equivalent energy is released (ranging from 10^48 to 10^54 erg). These explosive transients are associated with the catastrophic explosion of an isolated massive star (long-duration GRBs), or with the merger of compact objects (short-duration GRBs). Both scenarios lead to the formation of a highly magnetised neutron star or a spinning, stellar-mass black hole, which are thought to accrete material and launch two relativistic jets, causing the observed gamma-ray emission via magnetic processes or internal shocks. These jets interact with the surrounding material, producing the afterglow emission that extends from gamma-rays to radio waves. To understand GRB formation and evolution, a standard model involving an ultra-relativistic outflow is commonly employed. However, even sophisticated models face degeneracy in the multi-dimensional parameter space. To alleviate or possibly break the degeneracy, broad-band observations across the electromagnetic spectrum are crucial. In particular, the Very Long Baseline Interferometry technique (VLBI) has proven to be a unique asset, providing direct evidence of apparent superluminal expansion (for on-axis GRBs), centroid displacement of the outflow (for slightly off-axis GRBs) and the first confirmation that merger events can launch successful jets. In this Thesis, we employed radio and VLBI observations to characterise and constrain the outflow, the circum-burst medium, and the properties of the progenitors of GRBs. This included studies on individual events (GRB201015A and GRB221009A), which are important to test the predictions of current models, GRB host galaxies (GRB200716C), which are fundamental to constrain the nature of the progenitor through the characterisation of the surrounding environment, and the statistical properties of GRB afterglows, in order to verify the existence of potential GRB sub-populations
Learning with limited data
Get PDFIn recent years, Deep Learning techniques have demonstrated remarkable achievements across various Computer Vision tasks, frequently surpassing human capabilities. Nevertheless, these data-driven methodologies often demand large volumes of annotated data, necessitating laborious and costly manual annotation procedures.
The objective of this thesis is to introduce novel methods designed to mitigate this challenge by harnessing knowledge obtained from diverse domains or tasks, even in the presence of limited annotations. This challenge is commonly known as the Transfer Learning problem.
Our exploration will delve into the forefront of Transfer Learning, with a predominant emphasis on the advancement of techniques for Domain Adaptation in diverse computer vision tasks.
This research journey begins with a comprehensive investigation into 2D Semantic Segmentation, and we demonstrate how
other tasks such as Depth Estimation and Edge Detection can enhance the adaptability of models across different visual domains.
Subsequently, the exploration extends to the realm of 3D point cloud classification, where the challenges posed by diverse domain shifts are addressed once again exploiting auxiliary tasks such as shape reconstruction or recent Self-Supervised techniques.
The proposed works for 2D Semantic Segmentation and 3D point cloud classification lay the foundation for the development of novel frameworks aimed at tackling the challenging task of multi-modal Domain Adaptation for 3D Semantic Segmentation, where multiple sensors such as RGB cameras and LiDARs are available.
Finally, we shed some light on a new exciting and emerging topic which is solving common vision tasks on Neural Fields, which are an emerging paradigm used to represent signals such as images or 3D shapes. We will specifically focus on the 3D scenario, and in the context of Transfer Learning, show for the first time how acting directly on Neural Fields allows the possibility to transfer knowledge among different representations such as from 3D point clouds to meshes
Montessori pedagogy and methodology in relation to technological-digital development
Get PDFLa presente ricerca intende indagare la connessione di due ambiti di studi ed i possibili benefici in aree di sviluppo specifiche e trasversali, ambiti - l'uno relativo alla pedagogia e metodologia montessoriana, l'altro alle tecnologie educative - apparentemente discordanti tra loro, ma in realtà strettamente correlati dagli obiettivi formativi che entrambi si prefiggono come aiuto alla vita, a fronte della qualità immersiva che caratterizza ogni ambiente nell'odierna società complessa.
La bibliografia scientifica di riferimento, nazionale ed internazionale, evidenzia i caratteri distintivi dell'attualità dell'approccio formativo montessoriano, non solo per la crescente richiesta di aperture di scuole a metodo, ma anche per l'avvallo proveniente dagli studi psico-pedagogici e neuro-cognitivi, che aprono a possibili percorsi virtuosi in tema di educazione tecnologica.
L'orientamento della ricerca mixed methods a carattere esplorativo è stato condotto mediante l'utilizzo di strumenti quantitativi e qualitativi, procedendo alla raccolta dei dati tratti da questi ultimi con approccio botton up. Il metodo di analisi che caratterizza la Grounded Theory è stato individuato per l'analogia della circolarità osservazione-elaborazione teorica che, nella pedagogia montessoriana, permette all'insegnante, mediante un'osservazione sistematica, di procedere in una continua ricerca-azione, ponendo al centro del processo di apprendimento-insegnamento il bambino stesso.This research aims to investigate the connection of two fields of study and the possible benefits in specific and transversal areas of development, areas - one related to Montessori pedagogy and methodology, the other to educational technologies - apparently discordant, but correlated by educational objectives that both set themselves as an aid to life, for the immersive quality that characterises every environment in today's complex society.
The scientific bibliography of reference, national and international, highlights the distinctive characteristics about the actuality of the montessorian formative approach, not only due to the growing demand for opening method schools, but also for the endorsement coming from psycho-pedagogical and neuro-cognitive studies, which open up to possible virtuous paths in the field of technological education.
The orientation of the mixed methods research of an exploratory nature was conducted through the use of quantitative and qualitative tools, proceeding to the collection of the data with a botton up approach. The method of analysis that characterises the Grounded Theory has been identified for the analogy of the circularity observation - theoretical processing that, in Montessori pedagogy, allows the teacher, through systematic observation, to proceed in a continuous research - action, placing the child himself at the centre of the learning - teaching process
SMEs in the digital single market
Get PDFLa tesi dottorale in tema di micro, piccole e medie imprese (PMI) nel Mercato Unico Digitale, prendendo in esame anche le più recenti disposizioni e programmi emanati per far fronte agli effetti economici e sociali della crisi pandemica da Covid-19 e della guerra di aggressione della Russia nei confronti dell’Ucraina, indaga l’evoluzione della regolamentazione giuridica della politica industriale a livello europeo, analizzando ed approfondendo come nel tempo tale politica abbia trovato una più concreta e precisa definizione all’interno dell’Unione europea. Lo studio dottorale si focalizza, in particolare, sull’analisi dell’evoluzione della regolamentazione giuridica europea definita in favore delle PMI. La tesi dottorale analizza, successivamente, la disciplina del Digital Single Market introdotta a livello europeo che mira ad innovare il Mercato Unico europeo mediante l’utilizzo di strumenti e servizi digitali. Vengono approfonditi, in particolare, i nuovi strumenti digitali di finanziamento delle PMI sviluppatesi all’interno di un quadro di riferimento più ampio riconducibile al fenomeno del FinTech. Tra le attività innovative che si propongono come alternative ai finanziamenti bancari e che operano attraverso piattaforme web vengono approfonditi strumenti quali il crowdfunding e le initial coin offering su blockchain. Da ultimo sono analizzate nel dettaglio le politiche europee a favore dei Balcani occidentali nell’ottica di una graduale integrazione di questi ultimi all’interno dell’Unione europea, nonché sono approfondite specificamente le politiche introdotte per le PMI da parte dei Paesi dei Balcani occidentali e gli strumenti digitali alternativi ai finanziamenti bancari utili anche per le PMI. Dall’analisi condotta emergono criticità relative alla regolamentazione giuridica europea in favore delle PMI, che non riesce ancora appieno a creare un ambiente in cui tali imprese possano essere sostenute ed incoraggiate verso un futuro maggiormente digitale che riesca a renderle protagoniste di una nuova era economica ove digitale e green assumeranno un ruolo sempre più centrale per le imprese.The PhD thesis on micro, small and medium-sized enterprises (SMEs) in the Digital Single Market, taking also into consideration the most recent regulations and programs enacted to deal with the economic and social effects of the Covid-19 pandemic crisis and Russia's war of aggression against Ukraine, survey the evolution of the legal regulation of industrial policy at the European level, analyzing and deepening how over time this policy has met a more concrete and precise definition within the European Union. The PhD study focuses, in detail, on the analysis of the evolution of European legal regulation drawn up in favor of SMEs. Then the PhD thesis analyzes the Digital Single Market framework introduced at the European level that aims to innovate the European Single Market through the use of digital tools and services. In particular, new digital SME financing instruments developed within a broader framework pertaining to the FinTech phenomenon are examined. Among the innovative instrument that are proposed as alternatives to bank financing and operate through web platforms, the crowdfunding and the initial coin offerings on blockchain are explored in detail. Lastly, European policies for the Western Balkans are analyzed in detail with a perspective of the gradual integration of the Western Balkans within the European Union, as well as the policies introduced for SMEs by the Western Balkan countries and the alternative digital instruments to bank financing that are also useful for SMEs are specifically examined. The researches reveals critical issues related to the European legal regulation in favor of SMEs, which is still not fully capable of creating an environment in which these businesses can be supported and encouraged toward a more digital future that succeeds in making them players in a new economic era where digital and green will take an increasingly central role for businesses
Credibility of digital health predictors of human movement
Get PDFHuman mobility is a critical indicator of health, impacted by complex physiological systems. Disruptions to these systems can lead to reduced mobility, with severe consequences including loss of independence and increased mortality. This not only affects individuals but also poses societal and healthcare challenges. To address this, innovative solutions are needed, leveraging technologies like wearable sensors and computational simulations. These tools offer deep insights into human biomechanics, allowing for long-term monitoring and analysis of mobility parameters, even in pathological conditions. Innovations in wearable sensors enable real-world mobility monitoring, while computational models, particularly musculoskeletal dynamics models, predict human body behaviour, facilitating personalised treatment plans. However, ensuring the credibility of these technologies is paramount, requiring rigorous testing against established standards before clinical use. This PhD thesis aimed to investigate the credibility assessment of two models: analytics software for wearable sensor data and musculoskeletal dynamics models for identifying the primary cause for the loss of muscle force (i.e., dynapenia), mirroring the two distinct projects (i.e., the Mobilise-D and the ForceLoss projects, respectively). Both projects targeted conditions affecting mobility, necessitating credibility assessments for drug development and clinical decision-making. The Mobilise-D project, funded by EU, aimed to qualify mobility-related parameters extracted from wearable sensors as biomarkers for drug development. Engagements with regulatory authorities provided valuable feedback, guiding the qualification process. Conversely, the ForceLoss project developed a new framework combining experimental measurements and computational simulations for the differential diagnosis of dynapenia on osteoarthritic patients. In addition to credibility assessments, the thesis aimed to lower barriers by sharing experimental data and regulatory insights