Jose Marcos Moreno-Vega
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Jose Marcos Moreno-Vega has a total of 76 co-authors. Here are the top co-authors by number of shared publications:
- Belén Melián-Batista
Universidad de La Laguna
53 shared publications - Christopher Expósito-Izquierdo
Universidad de La Laguna
27 shared publications - Eduardo Lalla-Ruiz
University of Twente
19 shared publications - José Andrés Moreno Pérez
Universidad de La Laguna, IUDR and DIIS
11 shared publications - Miguel García-Torres
Data Science and Big Data Lab, Pablo de Olavide University, ES-41013 Seville, Spain
8 shared publications - Israel López-Plata
Universidad de La Laguna
4 shared publications
Latest Publications
- Open Access
- 10 November 2026
- Published by Elsevier in Applied Soft Computing
Variable neighborhood search and strategic oscillation for the cross-dock scheduling problem
- 10 November 2026
- Published by Elsevier in Applied Soft Computing
Open AccessVariable neighborhood search and strategic oscillation for the cross-dock scheduling problem
- Open Access
- 30 September 2026
- Published by Elsevier in Computers & Industrial Engineering
- Vol. 223
Formulations and efficient algorithms for the Cross-Dock Door Assignment Problem with makespan minimization
- 30 September 2026
- Published by Elsevier in Computers & Industrial Engineering
- Vol. 223
Open AccessFormulations and efficient algorithms for the Cross-Dock Door Assignment Problem with makespan minimization
- Open Access
- 30 September 2026
- Published by Elsevier in SSRN Electronic Journal
Memory-guided Adaptive Large Neighborhood Search for the Two-Echelon Capacitated Vehicle Routing Problem with Intermediate Facilities
This paper studies a two-echelon variant of the Capacitated Vehicle Routing Problem with Intermediate Facilities applied to urban solid waste management, wherein low-capacity collection vehicles and high-capacity transport vehicles must be coordinated to provide efficient service. The exact synchronization constraint at transfer stations without intermediate storage capacity, combined with multiple trips and visits within the operational horizon, constitutes an NP-hard problem that has been scarcely studied in the literature. We propose a hybrid memory-guided ALNS algorithm that integrates reactive GRASP, VND, and an extended ALNS with operator selection via Q-learning and in-line memory, which integrates both levels into the algorithm’s decision-making process. Experiments are conducted on 120 instances generated from real georeferenced data from the city of Saltillo, Mexico, and the results show that Gurobi obtains high-quality solutions in small instances, while the memory-guided ALNS outperforms it in terms of quality and computational efficiency in larger instances.
- 30 September 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessMemory-guided Adaptive Large Neighborhood Search for the Two-Echelon Capacitated Vehicle Routing Problem with Intermediate Facilities
This paper studies a two-echelon variant of the Capacitated Vehicle Routing Problem with Intermediate Facilities applied to urban solid waste management, wherein low-capacity collection vehicles and high-capacity transport vehicles must be coordinated to provide efficient service. The exact synchronization constraint at transfer stations without intermediate storage capacity, combined with multiple trips and visits within the operational horizon, constitutes an NP-hard problem that has been scarcely studied in the literature. We propose a hybrid memory-guided ALNS algorithm that integrates reactive GRASP, VND, and an extended ALNS with operator selection via Q-learning and in-line memory, which integrates both levels into the algorithm’s decision-making process. Experiments are conducted on 120 instances generated from real georeferenced data from the city of Saltillo, Mexico, and the results show that Gurobi obtains high-quality solutions in small instances, while the memory-guided ALNS outperforms it in terms of quality and computational efficiency in larger instances.
- Open Access
- 15 August 2026
- Published by Elsevier in SSRN Electronic Journal
A Hybrid Matheuristic for the Flexible Multi-Period Multi-Depot Capacitated Vehicle Routing Problem with Time Windows
This study addresses the Flexible Multi-Period Multi-Depot Capacitated Vehicle Routing Problem with Time Windows, motivated by waste collection in a circular-economy setting in which rigid collection schedules give way to flexible demand allocation, so that a customer's demand may be split across several days and visits. The problem is NP-hard and, in the variant studied here, further involves a heterogeneous fleet, multiple depots, and open-ended routes whose return leg to the depot is not charged. We propose a Mixed-Integer Programming formulation that uses continuous flow variables to enforce route connectivity and capacity, and that allows a customer's total demand to be fragmented across the horizon subject to minimum and maximum per-visit quantities and to time-window adherence. To solve larger instances, we develop a multi-start matheuristic that embeds exact subroutines within a metaheuristic frame. A reactive GRASP builds candidate visit-and-route structures without fixing its randomization parameter in advance; for any fixed structure, the collected quantities are then set exactly by solving a polynomial feasible-flow problem; and the routing is refined by a Variable Neighborhood Descent whose neighborhood ordering is learned online through a multi-armed-bandit rule. A load-aware Variable Neighborhood Search governed by strategic oscillation subsequently intensifies and diversifies the search, and a set-partitioning integer program, solved once per restart, recombines the routes accumulated over that restart into a final selection. Computational experiments over a benchmark of 60 instances compare the matheuristic with the Gurobi solver. On small instances the exact formulation provides optimal baselines, which the matheuristic reproduces. On larger instances the matheuristic scales far better and attains lower-cost solutions than the solver within a fraction of its time budget, under both the discrete and the continuous representation of demand. An ablation of the method apportions the improvement over the constructive phase between its two refinement phases, the variable neighborhood search accounting for a double-digit reduction in cost at every scale and the set-partitioning recombination for a smaller increment that grows with instance size.
- 15 August 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessA Hybrid Matheuristic for the Flexible Multi-Period Multi-Depot Capacitated Vehicle Routing Problem with Time Windows
This study addresses the Flexible Multi-Period Multi-Depot Capacitated Vehicle Routing Problem with Time Windows, motivated by waste collection in a circular-economy setting in which rigid collection schedules give way to flexible demand allocation, so that a customer's demand may be split across several days and visits. The problem is NP-hard and, in the variant studied here, further involves a heterogeneous fleet, multiple depots, and open-ended routes whose return leg to the depot is not charged. We propose a Mixed-Integer Programming formulation that uses continuous flow variables to enforce route connectivity and capacity, and that allows a customer's total demand to be fragmented across the horizon subject to minimum and maximum per-visit quantities and to time-window adherence. To solve larger instances, we develop a multi-start matheuristic that embeds exact subroutines within a metaheuristic frame. A reactive GRASP builds candidate visit-and-route structures without fixing its randomization parameter in advance; for any fixed structure, the collected quantities are then set exactly by solving a polynomial feasible-flow problem; and the routing is refined by a Variable Neighborhood Descent whose neighborhood ordering is learned online through a multi-armed-bandit rule. A load-aware Variable Neighborhood Search governed by strategic oscillation subsequently intensifies and diversifies the search, and a set-partitioning integer program, solved once per restart, recombines the routes accumulated over that restart into a final selection. Computational experiments over a benchmark of 60 instances compare the matheuristic with the Gurobi solver. On small instances the exact formulation provides optimal baselines, which the matheuristic reproduces. On larger instances the matheuristic scales far better and attains lower-cost solutions than the solver within a fraction of its time budget, under both the discrete and the continuous representation of demand. An ablation of the method apportions the improvement over the constructive phase between its two refinement phases, the variable neighborhood search accounting for a double-digit reduction in cost at every scale and the set-partitioning recombination for a smaller increment that grows with instance size.
- Open Access
- 27 May 2026
- Published by Elsevier in Expert Systems with Applications
- Vol. 329
A hybrid algorithm based on GRASP and variable neighborhood descent guided by machine learning for waste collection synchronization
- 27 May 2026
- Published by Elsevier in Expert Systems with Applications
- Vol. 329
Open AccessA hybrid algorithm based on GRASP and variable neighborhood descent guided by machine learning for waste collection synchronization
- Open Access
- 20 May 2026
- Published by Elsevier in SSRN Electronic Journal
Guided Strategic Oscillation with Reinforcement Learning for the Shared Customer Collaboration Vehicle Routing Problem
This paper addresses the Shared Customer Collaboration Vehicle Routing Problem (SCC-VRP), an NP-hard collaborative logistics problem in which competing carriers jointly serve shared customers to minimize total distribution costs. We propose Guided Strategic Oscillation with Reinforcement Learning (GSO-RL), a metaheuristic that advances the Strategic Oscillation paradigm by replacing predefined phase-transition rules with independent Multi-Armed Bandit agents that learn subregion-specific macro-move policies from accumulated regional performance history, where subregions are defined by a set of adaptive guiding solutions that evolve throughout the search. Computational experiments on benchmark instances and a new set of large-scale instances show that GSO-RL outperforms the state-of-the-art GRASPxILS by margins reaching -17.67% and a Multi-Neighborhood Tabu Search baseline by up to -14.10%, achieving the best-known solution across all evaluated instances. An ablation study confirms that GSO-RL components are essential to the scalability and effectiveness of the framework, with their contribution growing consistently with problem scale.
- 20 May 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessGuided Strategic Oscillation with Reinforcement Learning for the Shared Customer Collaboration Vehicle Routing Problem
This paper addresses the Shared Customer Collaboration Vehicle Routing Problem (SCC-VRP), an NP-hard collaborative logistics problem in which competing carriers jointly serve shared customers to minimize total distribution costs. We propose Guided Strategic Oscillation with Reinforcement Learning (GSO-RL), a metaheuristic that advances the Strategic Oscillation paradigm by replacing predefined phase-transition rules with independent Multi-Armed Bandit agents that learn subregion-specific macro-move policies from accumulated regional performance history, where subregions are defined by a set of adaptive guiding solutions that evolve throughout the search. Computational experiments on benchmark instances and a new set of large-scale instances show that GSO-RL outperforms the state-of-the-art GRASPxILS by margins reaching -17.67% and a Multi-Neighborhood Tabu Search baseline by up to -14.10%, achieving the best-known solution across all evaluated instances. An ablation study confirms that GSO-RL components are essential to the scalability and effectiveness of the framework, with their contribution growing consistently with problem scale.
- Open Access
- 3 June 2024
- Published by Elsevier in Computers & Industrial Engineering
- Vol. 193
Optimization of the production planning process in a nuts and dried fruits industry
- 3 June 2024
- Published by Elsevier in Computers & Industrial Engineering
- Vol. 193
Open AccessOptimization of the production planning process in a nuts and dried fruits industry
- 16 April 2022
- Published by Springer Nature
- p. 133-142
Intelligent Collaborative Freight Distribution to Reduce Greenhouse Gas Emissions: A Review
- 16 April 2022
- Published by Springer Nature
- p. 133-142
Intelligent Collaborative Freight Distribution to Reduce Greenhouse Gas Emissions: A Review
- 5 June 2020
- Published by Springer Nature
- Vol. 1238, 715-727
Fuzzy Greedy Randomized Adaptive Search Procedure and Simulation Model to Solve the Team Orienteering Problem with Time Windows
Tourism is a relevant economic activity that provides important resources (income, employment,...) to countries. When a tourist visits a country or city, he/she wants to know their points of interest. To do this, he/she must select some of the places according to his/her preferences and design routes to visit them. This problem can be adequately modeled as a Team Orienteering Problem with Time Windows (TOPTW). In this paper we propose a fuzzy GRASP and a multi-agent simulation model to solve the TOPTW. Our proposal incorporates two criteria to build the restricted candidate list. The computational results obtained show the validity of the proposal.
- 5 June 2020
- Published by Springer Nature
- Vol. 1238, 715-727
Fuzzy Greedy Randomized Adaptive Search Procedure and Simulation Model to Solve the Team Orienteering Problem with Time Windows
Tourism is a relevant economic activity that provides important resources (income, employment,...) to countries. When a tourist visits a country or city, he/she wants to know their points of interest. To do this, he/she must select some of the places according to his/her preferences and design routes to visit them. This problem can be adequately modeled as a Team Orienteering Problem with Time Windows (TOPTW). In this paper we propose a fuzzy GRASP and a multi-agent simulation model to solve the TOPTW. Our proposal incorporates two criteria to build the restricted candidate list. The computational results obtained show the validity of the proposal.
- Open Access
- 12 March 2019
- Published by Elsevier in Procedia Computer Science
- Vol. 149, 529-536
Using a decision tree algorithm to predict the robustness of a transshipment schedule
- 12 March 2019
- Published by Elsevier in Procedia Computer Science
- Vol. 149, 529-536
Open AccessUsing a decision tree algorithm to predict the robustness of a transshipment schedule