132051 research outputs found
Sort by
Learning to reschedule platforms: A graph neural network based deep reinforcement learning method for the train platforming and rescheduling problem
No full textThe train platforming schedule is the crucial plan for guiding trains to travel through a railway station without spatial and temporal conflicts. When trains are delayed in arriving at the station due to disturbances or disruptions, it raises the Train Platforming and Rescheduling Problem (TPRP), one of the hot topics in railway traffic management. It focuses on allocating platforms and time slots for trains to reduce delays and ensure operational efficiency in a station. This paper introduces a novel graph neural network based deep reinforcement learning method to address this problem, named Learning to Reschedule Platforms (L2RP). We formulate the solving process of TPRP as a customized Markov decision process. Meanwhile, we integrate a microscopic discrete-event train operation simulation model to serve as the agent exploration environment, which provides states, executes actions, and completes transitions. Then, we design a hybrid graph neural network based policy network to derive high-quality actions under each graph encoded state.The policy network is trained with the reward function designed to minimize total train knock-on delays and platform changes. The experiments on real-world instances show that the proposed L2RP method can produce high-quality solutions for instances of various scenarios within stably short solving times
Le sparizioni forzate in America Latina. Storia di un crimine transnazionale
No full textIl volume, il primo in lingua italiana interamente dedicato all'analisi storica delle sparizioni forzate in America Latrina, ricostruisce le dinamiche politiche, sociali e culturali che hanno dato origine alla pratica della sparizione forzata e alla sua diffusione, analizzandone le conseguenze nel lungo periodo sulle società latinoamericane e le sue riproposizioni nel tempo. Accanto alla storia della repressione, gli autori danno spazio alle mobilitazioni sociali e alle iniziative giuridiche che nel corso dei decenni hanno permesso di portare alla luce un crimine a lungo negato e di favorirne il riconoscimento a livello internazionale. Il testo offre chiavi di lettura essenziali per comprendere una delle forme più radicali della violenza politica contemporanea e le sue eredità nel presente.This volume, the first in Italian entirely devoted to the historical analysis of forced disappearances in Latin America, reconstructs the political, social, and cultural dynamics that gave rise to the practice of forced disappearance and its spread, analyzing its long-term consequences on Latin American societies and its recurrences over time. Alongside the history of repression, the authors explore the social mobilizations and legal initiatives that over the decades have brought to light a long-denied crime and promoted its international recognition. The text offers essential insights into one of the most radical forms of contemporary political violence and its legacy in the present
Train timetable rescheduling considering potential train operation conflicts: an enhanced deep reinforcement learning approach
No full textTrain timetable rescheduling (TTR) is the core of real-time train dispatching. It is a challenging sequential decision-making problem due to complex constraints. Leveraging deep reinforcement learning’s efficacy in sequential decision-making, we formulate TTR as a Markov Decision Process (MDP) and propose a Proximal Policy Optimization (PPO) method. To enhance exploration efficiency in large action spaces, we introduce a conflict-aware action selection rule (ASR) and a multi-dimensional discrete action space. Experiments on the Wuhan-Guangzhou high-speed railway under various disturbances demonstrate that: (1) The proposed method yields lower total train arrival delays than heuristic rules and other DRL algorithms, with an average optimality gap below 4.0%; (2) The computational time is approximately 4 s; (3) The ASR contributes to an average delay reduction of 27.2%; and (4) Sensitivity analysis indicates optimal performance with 5 trains per action dimension. The results validate the effectiveness and applicability of the proposed method in real-time TTR
Microwave Vortex Motion Characterization of Nb3Sn Coatings for Applications in High Magnetic Fields
No full textIn this work, microwave measurements carried out in dielectric-loaded resonators exposed to high magnetic fields are exploited to yield the surface impedance of Nb3 Sn superconducting coatings deposited via two different techniques: vapor tin diffusion, and DC magnetron sputtering. The obtained data lead to qualitative interpretations on both the Nb3 Sn superconducting properties, and vortex-dynamics and pinning, of each coating separately, as well as simple distinctive features when comparing those. When examining the respective surface impedances at varying field, it is expected that the studied films perform at substantially diverse magnitudes of flux-flow resistivity, but also in well-differentiated pinning regimes, yet the obtained surface resistances of both samples are comparable, thus demonstrating that there is room for film optimization at the expense of certain compromise between the parameters involved
Il coordinamento europeo delle norme nazionali sugli investimenti esteri e il golden power italiano
No full textL’Autore individua nella nuova formulazione dell’art. 207 TFUE, introdotta con il Trattato di Lisbona, l’origine della riforma italiana del 2012 sugli investimenti esteri, poi riproposta in altri Stati membri dell’Unione europea. Il coordinamento europeo di queste discipline definito con il Regolamento europeo 452 del 2019 è ora oggetto di una proposta di modifica.
L’Articolo esamina la disciplina nazionale e quella europea, sia nelle formulazioni originarie sia nelle versioni attuali (in vigore o proposte), anche alla luce degli arresti giurisprudenziali, e individua punti di contatto e contrasti, attuali e potenziali.
Tali contrasti vengono imputati dall’Autore sul piano nazionale, alla natura ibrida della norma italiana, che aveva il duplice obiettivo di superare i profili di illegittimità delle norme sulle golden shares e di disciplinare ex novo la materia degli investimenti esteri. Sul piano unionale invece, i contrasti originano dall’esigenza di consentire agli Stati membri azioni in deroga, giustificate da esigenze di tutela degli interessi nazionali, nei limiti però stabiliti dalla giurisprudenza consolidata del mercato interno.
L’Articolo illustra infine proposte interpretative per superare i contrasti o mitigarne gli effetti.The Author traces the origin of the Italian reform on foreign investments of 2012, subsequently replicated in other European Union Member States, to the new wording of Article 207 TFEU introduced by the Treaty of Lisbon. The European coordination of these regimes, established by Regulation (EU) No. 452/2019, is currently the subject of a proposal for amendment.
The article examines both the national and EU legal frameworks in their original and current versions (whether in force or proposed), also considering relevant case law, identifying areas of convergence and divergence, both actual and potential.
At the national level, these divergences are attributed to the hybrid nature of the Italian legislation, which pursued the dual objective of overcoming the unlawfulness identified in the golden shares regime while simultaneously introducing a new regulatory framework for foreign investments. At the EU level, divergences arise from the need to permit Member States to adopt derogating measures justified by the protection of national interests, within the limits established by settled internal market case law.
The article concludes by advancing interpretative proposals aimed at resolving these divergences or mitigating their effects
Intelligenza artificiale e tutela del consumatore
No full text1. La data!cation consumeristica e l’avvento dei sistemi di intelligenza arti!ciale nelle
relazioni B2C – 2. Una legge non consumer-oriented – 3. Il targeted advertising e i dark
pattern – 4. I sistemi di raccomandazione – 5. La personalizzazione dei prezzi – 6. Il
credit scoring e il rating reputazionale – 7. Conclusion
Assessing the Effectiveness of Augmented Reality Warnings for Improving Safety at Highway Merging Zones in a Connected Environment
No full textThe merging process of vehicles moving from the on-ramp to the right lane of a highway is quite critical at interchanges for both safety and operational efficiency. Several studies have emphasized the importance of merging zones at interchanges, identifying them as major sites of conflicts between passing and entering vehicles. Incorrect assessment of gaps and precedence in these zones can result in serious collisions and traffic disruptions at interchanges. This study looks at how effective Augmented Reality (AR) solutions in a connected environment are in improving safety at merging zones, helping drivers keep a safe distance from vehicles entering the highway from on-ramps. In a driving simulation study, four different safety measures are evaluated: i) a vehicle-to-vehicle spacing measurement support, which integrates a Variable Message Signs panel (VMS) with a new type of road pavement marking; ii) a Head-Up Display (HUD) that projects a static symbol onto the vehicle windshield, advising the driver to maintain a safe distance from the entering vehicle; iii) a connected vehicle system that integrates AR technology consisting of a dynamic symbol on the road surface, which provides the driver with additional visual cues about the actual longitudinal distance from the entering vehicle; iv) the same AR-based system with an additional audible warning. The driving behavior of forty-four participants in four different configurations with the safety measures was compared with their behavior in a configuration without any measures (baseline condition). The results demonstrated a significant positive impact from all tested measures, with the most effective solution being AR systems, which demonstrated the ability to assist drivers in adjusting their distance from the entering vehicles. This study confirms the high potentialities of AR technologies and connected vehicles in enhancing the overall safety of road networks, particularly in high-risk scenarios and challenging maneuvers
“Stylish” similatives
No full textIn this contribution, we investigate two specific constructions with the lexical item ‘style’ that have developed similative uses. By looking at Russian v stile ‘in the style of’ and Italian (in) stile ‘(in) style’, we claim that the similarity mechanism that underlies these style-noun constructions does not depend on taxonomic relationships (as for type nouns) but rather relies on the context and knowledge shared by the interlocutors. In these constructions, the right-hand slot evokes a reference frame in which the left-hand element is inserted through similarity. Relying on corpus data, we analyze and compare the two constructions in terms of the mechanisms exploited to express similarity, the kind of similarity encoded, the grammaticalization pathway undergone by each construction, and the variety of functions they can perform. In addition, a comparative, contrastive analysis between the two languages allows to identify other functionally equivalent constructions that demonstrate the existence of a category of style-noun constructions
