Pasar al contenido principal

MARCO INTEGRADO DE PREDICCION PARA LA MOVILIDAD URBANA Y GOBERNANZA

AIA2025-163553-C44

•
Nombre agencia financiadora Agencia Estatal de Investigación
Acrónimo agencia financiadora AEI
Programa Programa Estatal para la Investigación y el Desarrollo Experimental
Subprograma Subprograma Estatal de Generación de Conocimiento Científico-Técnico y Desarrollo Experimental
Convocatoria Proyectos de Investigación en el ámbito de la Inteligencia Artificial
Año convocatoria 2025
Plan Estatal 2024-2027 Plan Estatal de Investigación Científica y Técnica y de Innovación
Centro beneficiario UNIVERSITAT POLITÈCNICA DE VALÈNCIA

Resultados relacionados

Resultados totales (Incluyendo duplicados): 9
Encontrada(s) 1 página(s)

RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

A Biased-Randomized Algorithm for the Bi-Objective Capacitated Dispersion Problem with Symmetries

  • Gómez, Juan F.
  • Calvet, Laura
  • Ammouriova, Majsa
  • Wenwen, Chen
  • Juan, Angel A.|||0000-0003-1392-1776
[EN] Given a network of nodes and a certain demand that needs to be satisfied, the capacitated dispersion problem (CDP) involves selecting a subset of nodes to maximize dispersion between them. In many practical instances, symmetry in the structure of the selected nodes (e.g., using nodes of the same type) can lead to synergies. Hence, this paper studies a bi-objective variant of the CDP to account for these symmetries. The first goal seeks to maximize the minimum distance between opened nodes, while the second goal accounts for symmetry by penalizing the use of nodes of different types (in our case, represented by different colors). We formalize the problem as a bi-objective mathematical program and address it through a classical multiobjective strategy, the & varepsilon;-constraint method. Exact methods are used when the problem size allows, even though the problem is NP-hard. To tackle larger instances, we design a biased-randomized algorithm based on a constructive heuristic. Computational experiments show that our biased-randomized algorithm provides high-quality approximations of the Pareto frontier., This work has been partially supported by the Spanish Ministry of Science, Innovation and Universities/AEI (PID2022-138860NB-I00, AIA2025-163553-C44) and the Generalitat Valenciana (2024 CIAICO 117).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

A Matheuristic Approach for the Bi-Objective and Multi-Source Team Orienteering Problem with Prioritized Nodes

  • Oltra Crespo, Sandra|||0000-0003-1995-2557
  • Agud Albesa, Lucia|||0000-0002-1222-7988
  • Garrido-Saez, Neus|||0000-0002-7903-8591
  • Juan, Angel A.|||0000-0003-1392-1776
[EN] This work proposes a matheuristic approach that combines heuristic techniques with mathematical optimization to address a bi-objective, multi-source team orienteering problem. During the exploration phase, nodes are assigned to depots using a biased-randomized round-robin heuristic, guided by a ranking criterion based on marginal distances. This step aims to generate diverse mappings and identify those that lead to promising solutions under a convex combination of the two objectives: total reward and number of priority nodes visited. In the exploitation phase, an intensive search is applied to a subset of elite mappings, further improving the quality of the solutions. The method produces a Pareto frontier, offering a range of trade-offs between the two objectives. Computational experiments on adapted benchmark instances confirm the method¿s ability to generate high-quality or near-optimal solutions, often outperforming exact methods under realistic time constraints., This work has been partially supported by the Spanish Ministry of Science, Innovation and Universities MICIU/AEI/10.13039/501100011033 (PID2025-169830NB-I00, AIA2025-163553-C44, PID2022-138860NB-I00), the Generalitat Valenciana (2024 CIAICO 117) and the Universitat Politècnica de València (PAID-06-25).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

A sim-learnheuristic algorithm for the stochastic and context-aware team orienteering problem

  • Tobar-Fernandez, C.
  • Perez-Pelo, S.
  • Lopez-Sanchez, A. d.
  • Sanchez-Oro, J.
  • Juan, A. A.
[EN] This paper addresses the Team Orienteering Problem (TOP) with stochastic, context-aware travel times. Traditionally, simheuristics have been used to address stochastic variants of TOP and related routing problems. Simheuristics combine metaheuristics and simulation to deal with stochastic inputs, but they often rely on generic, independent random sampling which ignores that travel times in real traffic networks are dynamic and depend on contextual factors, such as congestion or weather conditions. We propose to extend the traditional simheuristic concept by integrating a machine learning component to overcome this limitation. To apply this novel methodology, models are trained on historical data to predict context-aware travel times. In addition, this historical data and clustering models are used to generate context-aware probability distributions. These customized distributions provide the simulation step with more realistic inputs, ensuring that the evaluation of candidate solutions reflects actual dynamic conditions. We compare a deterministic metaheuristic, a standard simheuristic approach, and the proposed sim-learnheuristic approach on extended benchmark instances. Computational experiments show that the sim-learnheuristic consistently outperforms both the traditional simheuristic and the deterministic metaheuristic, achieving average improvements of approximately 18% and 30%, respectively, under stochastic-dynamic evaluation., This work has been partially supported by the Spanish Ministry of Science, Innovation and Universities/AEI (PID2025-169830NB-I00, PID2022-138860NB-I00, AIA2025-163553-C44) and the Generalitat Valenciana (2024 CIAICO 117), by Comunidad Autonoma de Madrid (grant ref. TEC-2024/COM-404), Ministerio de Economia y Competitividad (grant ref. PID2024-160226OB-C22), and Ministerio para la Transformacion Digital y de la Funcion Publica (Catedra ENIA AI4DDS, grant ref. TSI-100930-2023-3). Funding was also received from the "Agencia Estatal de Investigacion" through the "Ayudas para contratos para la formacion de doctores y doctoras en empresas y otras entidades (Doctorados Industriales) 2024" (grant ref. DIN2024-013339), through the company OGA (qosITconsulting, S.L).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

On the Use of Biased-Randomized Transformers as Data-Driven Heuristics for Agile Optimization

  • Juan, Angel A.|||0000-0003-1392-1776
  • Escoto-Gomar, Marc
  • Guerrero, Antoni
  • Panadero, J.
  • Garcia-Sanchez, Alvaro
  • Resende, Mauricio G. C.
[EN] This paper proposes the concept of biased-randomized transformers, a novel methodology that combines biased-randomized techniques and transformer-based deep learning for 'agile' optimization (i.e., real-time optimization that is carried out iteratively in dynamic systems). On the one hand, biased-randomization techniques have been used in the past to inject controlled randomness into greedy heuristics, thus converting them into probabilistic algorithms capable of generating thousands of good-quality solutions while preserving heuristic logic. On the other hand, transformer models can capture complex patterns across thousands of variables. Once trained, these models can be seen as data-driven heuristics able to provide fast solutions to new instances and adapt to changing inputs. The combination of biased-randomization techniques with trained transformers allows for a fast exploration and selection of the high-quality solutions to NP-hard combinatorial optimization problems. The paper includes two case studies that illustrate the potential of these biased-randomized transformers., This research was funded by the Spanish Ministry of Science, Innovation and Universities/AEI (PID2022-138860NB-I00, AIA2025-163553-C44, DIN2024-013395) and the Generalitat Valenciana (2024 CIAICO 117).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Sim-Exact Methods for Stochastic Optimization: A Complementary Approach to Simheuristics

  • Juan, Angel A.|||0000-0003-1392-1776
  • Escoto-Gomar, Marc
  • Medina-Rodriguez, Veronica
  • Rodríguez Uguina, Antonio
[EN] This paper introduces a sim-exact methodology for stochastic combinatorial optimization problems. The approach combines exact optimization models with Monte Carlo or discrete-event simulation to evaluate candidate solutions under uncertainty. The method iteratively adjusts a control parameter based on simulation feedback and solves a sequence of deterministic optimization problems. Unlike scenario-based stochastic programming, the approach does not rely on explicit scenario enumeration, and unlike simheuristics, it preserves optimality with respect to each deterministic subproblem. The methodology is tested on the vehicle routing problem with stochastic demands under different levels of demand variability. Results are compared with a simheuristic approach and a sample average approximation (SAA) method. The results show that sim-exact performance is comparable to simheuristics, with no statistically significant differences in most cases, while SAA shows weaker performance under medium and high variability., This work has been partially supported by the Spanish Ministry of Science, Innovation, and Universities/AEI (PID2022-138860NB-I00, AIA2025-163553-C44, https://doi.org/10.13039/501100011033) and the Generalitat Valenciana (2024-CIAICO-117).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

The stochastic maximum diversity problem

  • Martinez-Gavara, Anna
  • Keenan, Peter
  • Pariente, Antonio
  • Marti, Rafael
  • Juan, Angel A.|||0000-0003-1392-1776
[EN] The Maximum Diversity Problem (MDP) is a well-known combinatorial optimisation model with applications ranging from facility location to social network analysis. This paper introduces a stochastic extension, the Stochastic Maximum Diversity Problem (SMDP), to better model applications based on attributes that may change, such as the travel time in location or routeing problems and affinity or preferences in social networks. The mathematical programming literature widely recognises that using stochastic data introduces an additional layer of difficulty to optimisation problems and therefore makes them more difficult to solve. The MDP is an NP-hard problem; consequently, the stochastic maximum diversity problem (SMDP) considered in this paper is difficult to solve. This paper proposes an extension to the scatter search (SS) method, the Stochastic SS, which integrates SS with simulation to solve the SMDP efficiently. Specifically, a Simheuristic SS algorithm is considered. It is based on customised search strategies to explore the solution space efficiently. The proposed approach is validated by an extensive comparison with alternative reference methods, and the results show the superior performance of the proposed method for stochastic problems., This work was supported by the Spanish Ministry of Science, Innovation and Universities (ERDF) (PID2024-160226OB-C21, PID2022-138860NB-I00, AIA2025-163553-C44, PID2021-125709OB-C21) and the Generalitat Valenciana (2021 CIAICO 224, 2024 CIAICO 117).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Generative Heuristics for Hard- and Soft-Criteria Optimization: Integrating LLMs Within Metaheuristics

  • Saiz, Miguel
  • Panadero, J.
  • Juan, Angel A.|||0000-0003-1392-1776
[EN] In many real-world domains, optimization problems involve both quantitative objectives and qualitative criteria that cannot be directly formulated as mathematical functions. While recent research has integrated large language models (LLMs) into optimization algorithms, these approaches primarily use LLMs to guide the search process, generate heuristics, or tune optimization strategies, rather than to evaluate semantic objectives. This paper introduces generative heuristics, a methodology that combines traditional metaheuristics with LLM-based semantic evaluation to address 'soft optimization' problems containing both hard quantitative constraints and soft qualitative objectives. The proposed methodology first applies a metaheuristic algorithm to generate a shortlist of high-quality candidate solutions satisfying the quantitative objectives. These candidates are then evaluated by an LLM according to a user-defined qualitative rubric, and the resulting semantic scores are incorporated into the optimization process through a weighted soft objective. The methodology is illustrated through a constrained portfolio optimization problem in which financial risk is minimized while simultaneously promoting strategic objectives related to sector diversification, geographic exposure, environmental-social-governance quality, and overall portfolio coherence. Experimental results show that the proposed methodology consistently improves strategic alignment while producing solutions that remain close to those obtained from the hard optimization formulation., This work has been partially supported by the Spanish Ministry of Science, Innovation and Universities MICIU/AEI/10.13039/501100011033 (PID2022-138860NB-I00, AIA2025-163553-C44), the Barcelona City Council-La Caixa (2025 OVT 280226), and the Generalitat Valenciana (CIAICO/2024/117, CIBEST/2026/135).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Evaluating Urban Mobility Transitions: A Dual-Track Framework for City-Scale and Local Assessment

  • Cuartas-Micieces, Javier A.
  • Soriano-González, Raquel|||0000-0002-1337-9561
  • Juan, Angel A.|||0000-0003-1392-1776
  • Ammouriova, Majsa
[EN] Evaluating urban mobility transitions is essential to determine whether local transport interventions support broader sustainability goals. Cities increasingly implement initiatives to promote public transport, active mobility, and low-carbon transport systems. Still, assessing their impact on city-scale structural change remains challenging. Existing evaluation approaches often rely on project-level monitoring or fragmented indicators, which limits cross-city comparison and the assessment of long-term system transformation. This paper proposes a dual-track methodology to evaluate sustainable urban mobility interventions. The first track uses city-defined key performance indicators to capture local implementation processes, governance dynamics, and perceived outcomes. The second track relies on publicly available open data to assess city-scale changes in mobility indicators, including public transport accessibility, cycling infrastructure provision, and traffic-related air pollution. The methodology is applied to ten European cities using open data and satellite-based environmental indicators. Results indicate that while cities report progress at the project level, external indicators show limited short-term structural change in city-wide mobility systems. These findings highlight the value of open data as an independent evaluation layer that contextualises local results and supports transparent assessment of urban mobility transitions., This work has been partially supported by the European Commission (UP2030, HORIZONMISS- 2021-CIT-02-01-101096405), the Spanish Ministry of Science and Innovation (PID2022-138860NB-I00, AIA2025-163553-C44), and the Generalitat Valenciana (2024 CIAICO 117).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Solving a Multi-Period Dynamic Pricing Problem Using Learning-Augmented Exact Methods and Learnheuristics

  • Juan, Angel A.|||0000-0003-1392-1776
  • Yangchongyi, Men
  • Medina-Rodriguez, Veronica
  • Escoto-Gomar, Marc
[EN] This paper addresses a dynamic multi-period pricing problem that incorporates time-varying contextual information and inventory constraints. Sales are modeled as a function of both price and a multidimensional context vector, which may include factors such as customer location, income, loyalty, competitor prices, and promotional activity. This formulation captures complex market dynamics over a finite selling horizon. The problem is formulated as a quadratic programming model, and two alternative solution approaches are proposed. The first uses a multivariate regression model to approximate the sales function linearly, allowing an exact quadratic programming solution that serves as a benchmark. The second is a 'learnheuristic' algorithm that combines a nonlinear sales learning model with metaheuristic optimization to generate high-quality pricing strategies under realistic operational constraints. Computational experiments demonstrate the effectiveness of the proposed learnheuristic approach., This work has been partially supported by the Spanish Ministry of Science, Innovation and Universities/AEI (PID2022-138860NB-I00, AIA2025-163553-C44), the Generalitat Valenciana (2024-CIAICO-117) and Coca-Cola EPP (CE20231180).