This study presents a novel approach to a Sensor Allocation Problem (SAP) in Wireless Sensor Networks (WSNs), a combinatorial optimization problem focused on optimizing network topology to minimize energy consumption while ensuring connectivity. We propose a hybrid methodology that combines the Biased Random-Key Genetic Algorithm (BRKGA), with reinforcement learning-based refinement incorporated into its decoder, and Local Branching techniques for efficient sensor placement. The integration of these techniques offers a scalable and adaptable solution to complex network configurations, achieving superior performance compared to traditional exact methods. Extensive computational experiments demonstrate the robustness and effectiveness of this approach across various network topologies, including regular, semi-regular networks. Our results highlight the ability of the proposed methodology to efficiently allocate sensors in large-scale, sparse, and dynamically changing networks, addressing the challenges of energy efficiency, connectivity, and coverage.

This paper presents a hybrid optimization methodology integrating Biased Random-Key Genetic Algorithms (BRKGA), data mining, and reinforcement learning to address the Sensor Network Optimization Problem. Using statistical and clustering techniques, the method enhances BRKGA by incorporating patterns extracted from elite solutions. A reinforcement learning agent dynamically decides when and how to mine patterns, enhancing adaptivity and optimizing the process. The validated patterns are then integrated into a Quadratic Knapsack Problem (QKP) optimization model, guaranteeing compliance with budget constraints while efficiently exploring potential regions of the solution space. Experimental results demonstrate the approach's effectiveness in generating high-quality solutions for sensor allocation in Wireless Sensor Networks (WSNs), balancing exploration and exploitation, and achieving robust performance in complex optimization scenarios.

Optimizing vehicle routing is essential for logistics efficiency, but companies often operate independently due to competitive constraints, limiting potential gains from collaboration. In this work, we introduce a federated-inspired optimization approach for the Capacitated Vehicle Routing Problem (CVRP), where companies share optimized hyperparameters instead of raw data. Our method leverages Hybrid Genetic Search (HGS) for CVRP optimization while preserving data privacy. Unlike traditional federated learning, aggregation occurs outside the federated-inspired process, focusing on hyperparameter tuning rather than model updates. We evaluate different heuristic aggregation strategies, including FedAvg, Trimmed Mean, and Majority Vote. Experiments were conducted using CVRPLIB benchmark instances, varying the number of companies and problem instances to assess scalability and effectiveness. Our results indicate that federated-inspired hyperparameter aggregation can offer advantages over isolated optimization, particularly in reducing total routing costs. These findings suggest that controlled hyperparameter sharing may enhance logistics decisionmaking while maintaining operational autonomy.

Green infrastructure plays a pivotal role when combating the temperature rises caused by climate change. In addition, in the context of smart and sustainable cities, those spaces, such as urban forests, can be used as places for tours and physical activities in contact with nature, even in big urban areas. However, as the number of people using those spaces increases, so does the risk that an individual gets lost. This kind of incident must be quickly addressed in order to avoid severe health outcomes to the individual. Those rescue operations mobilize firefighters and expensive equipment, such as helicopters, making them extremely costly to public authority. In order to reduce the search time in those rescue operations, this paper considers the Data Mule Routing Problem with Limited Autonomy (DMRP-wLA). This research proposes a hybrid algorithm, composed of an Ant Colony Optimization algorithm enhanced by data mining techniques, to find high-quality solutions to the highlighted problem. Experimental results indicate that ACO-DM consistently outperforms traditional ACO approaches, achieving the best average solution costs in 11 out of 15 evaluated scenarios. In comparison with existing algorithms in the literature, the proposed ACO-DM approach provided substantial cost reductions—between 20 % and 60 % in large-scale instances—while maintaining practical UAV operational times.

This study introduces a novel approach to the Sensor Allocation Problem (SAP) in Wireless Sensor Networks (WSNs) by integrating reinforcement learning with a multi-centrality heuristic. The goal of SAP is to optimize the network topology to minimize energy consumption while maintaining connectivity. We propose a reinforcement learning agent that interacts with a multi-centrality heuristic to dynamically select sensor placement. Extensive computational experiments were performed on both regular and semi-regular grid topologies of varying sizes. Comparisons with a Mixed-Integer Linear Programming (MILP) model reveal that our approach not only yields high-quality solutions in small to medium instances but also remains effective in larger networks, where the MILP solver often fails to produce feasible solutions. Notably, the RL module consistently improves upon the baseline heuristic allocation, demonstrating significant cost reductions while preserving full network coverage. By uniting a centrality-driven heuristic with an adaptive learning mechanism, the proposed hybrid framework addresses challenges in SAP more effectively than traditional exact methods. Consequently, it provides a promising avenue for large-scale WSN deployments, maintaining robust connectivity and efficiently managing energy consumption across various network configurations.

In this paper, we aim to reduce the number of nodes from Graph Neural Networks (GNNs), thereby simplifying models and reducing computational costs. GNNs are highly effective for various tasks, such as prediction, classification, and clustering, due to their ability to learn node and edge attributes and relationships, and they have been utilized for intelligent transportation systems recently by converting sensor networks into graph structures. Deep spatio-temporal neural networks, including Spatio-Temporal Graph Convolutional Networks (STGCNs), capture spatial and temporal dependencies, making them suitable for traffic speed forecasting, traffic demand prediction, and travel time estimation. Despite their success, GNNs face challenges in industrial applications due to significant memory usage and time consumption. In this paper, we propose a new approach to node reduction that outperforms existing methods in computational efficiency. Our experiments on two real-world traffic datasets demonstrate that using the heuristic and edge information to reduce nodes can cut computation time of optimization up to 95% and, by eliminating noise, can even enhance prediction accuracy.

Wireless sensor networks are widely valued for their effectiveness in real-time data collection. As the amount of data exchanged within such networks grows, designing a robust network topology that maximizes area coverage with minimal sensors has become a critical challenge. The choice of topology impacts key network metrics, including sensor coverage, communication range, connectivity, inference, and installation and management costsIn this paper, we address the Wireless Sensor Network Planning Problem with Multiple Sources/Destinations, presenting an optimization approach based on deep reinforcement learning. This problem is noteworthy, as sensors in various applications are often required to share data within distinct destinationsWe leverage deep reinforcement learning to effectively address the complex task of selecting optimal sensor locations. Our reinforcement learning agent dynamically learns network structure by iteratively adding and removing sensors, optimizing both sensor coverage and the total number of sensors used. Experiment across diverse scenarios demonstrate the effectiveness of our method for network planning problems of varying scales, achieving full coverage with fewer sensors than traditional approaches. Additionally, our approach also produce solutions for large instances where Mixed Integer Programming solvers were not able to. Overall, our method was able to reduce the number of sensors used by up to 22.3% compared to other methods.

In the context of smart cities where green infras-tructure is incentived, besides important benefits like regulating temperatures and absorbing pollutants among others, tour by urban forests is a way to experience closer contact with nature near of big urban centers. Eventually, visitors get lost, and helping these people with velocity is important to avoid severe incidents. Normally, rescue operations mobilize firefighters, ex-pensive equipment like helicopters and public resources. Following that idea of reducing search time in rescue operations, this paper considers the Data Mule Routing Problem with Limited Autonomy (DMRP-wLA). To find high-quality solutions, this paper proposes an Ant Colony Optimization algorithm enhanced with Reinforcement Learning to create an adaptive decision-making algorithm.

While Spatio-Temporal Graph Convolutional Networks (STGCNs) are an effective method for traffic speed fore-casting, their training and inference tend to be time-consuming. In this paper, we aim to refine these networks by strategically reducing their number of nodes, thereby boosting computational efficiency. The nodes in these graphs represent data observed for road segments, and by analyzing the interconnections and layout of the graph, we can identify nodes with minimal contribution to overall performance. Removing these nodes can potentially decrease computation time while maintaining the prediction accuracy. We employ the Biased Random-Key Genetic Algorithm (BRKGA) to identify a good set of nodes for removal, based on a predefined percentage reduction of the original graph size (e.g., retaining 95 % of the original graph). We evaluate different graph size configurations, ranging from 95 % to 70 % node retention, to determine the least impactful node set performance. Our experiments on three real-world datasets reveal that reducing nodes can decrease computation time by up to 29%, and as a byproduct of removing noise, even improve the prediction accuracy.

Hyperparameter optimization (HPO) is paragon to maximize performance when designing machine learning models. Among different HPO methods, Genetic Algorithm (GA) based optimization is considered effective because it allows a wide and diverse range of solutions to be explored. However, GA's exploratory nature makes this type of algorithm to evaluate many solutions that do not improve the overall performance. This is specially costly when the objective function to be evaluated is time-consuming, like in the HPO field. In this paper, we propose an efficient hybrid algorithm that is able to reduce computational cost by combining deep reinforcement learning with the Biased Random Key Genetic Algorithm (BRKGA), a variant of genetic algorithms. Our reinforcement learning agent has a decision-making role during the population's fitness calculation, in which it filters out chromosomes that would not improve the overall fitness of the population. The agent uses small amounts of pre-trained data to identify trends in potentially good solutions, and carry out its decision process. We conduct experiments on eight different datasets to assess the effectiveness of the proposed method, and the results show that the proposed method can significantly reduce the computation time of hyperparameter search using BRKGA (up to 44% reduction in computational time) without compromising the quality of the solution (no statistically difference in results).

This paper introduces a novel hybrid genetic algorithm combined with data mining to solve a version of the early/tardy scheduling problem in which no unforced idle time may be inserted in a sequence. The chromosome representation of the problem is based on random keys and we use the Biased Random-Key Genetic Algorithm (BRKGA) to establish an order in which jobs are scheduled. A data mining component gathers data from the evolutionary steps of BRKGA and suggests chromosomes based on past observations. In this way, we address a key challenge in BRKGA - the exploration of solutions near an individual gene - by leveraging data-driven insights to refine and enhance the search space. Comparative analysis reveals that our hybrid algorithm significantly benefits from the pattern recognition capabilities of data mining, leading to improved scheduled efficiency. The results of a series of computational experiments underscore the potential of this hybrid approach that, although requiring slightly longer computational times, is better than the previous baseline BRKGA algorithm in terms of solution quality.

This paper presents a hybrid BRKGA (Q-HBRKGA) that combines BRKGA with Q-learning and a Local Branching technique to solve the Knapsack Problem with Forfeits(KPF). The aim is to tackle this problem, a vari-ant of the 0–1 Knapsack Problem, where a penalty cost is imposed when both items of a forfeit pair are included in the solution. Notably, Q-HBRKGA achieves superior results compared to existing approaches, establishing a valid strategy for KPF problem-solving. Computational experiments, utilizing benchmark instances, demonstrate the efficacy of Q-HBRKGA by outperforming state-of-the-art methods documented in the literature.

This study addresses the Wireless Sensor Network Planning Problem with Multiple Sources/Destinations (WSNPMSD), an optimization challenge focused on reducing the sensor count within a network topology for a specified area, considering numerous sources and destinations. We introduce a hybrid strategy for tackling WSNP-MSD, particularly effective for large-scale scenarios, combining a Biased Random-key Genetic Algorithm with a Local Branching Technique. This methodology is justified by the limitations exact methods may encounter when the number of variables increases. Through computational experiments, we demonstrate the superiority of our proposed method over conventional exact methods in managing large instances of the WSNP-MSD.

With the recent upsurge in the use of deep learning and other computationally expensive machine learning models, hyperparameter optimization has become a quite important and widely researched area of study. Genetic algorithms, a subclass of evolutionary algorithms, have proven to be an effective approach and have been widely used in recent years. However, efficiently exploring the domain of possible solutions remains a challenging, and often computationally-expensive task. In this paper, we present a novel and efficient hyperparameter optimization strategy based on a genetic algorithms variant: Biased Random-key Genetic Algorithms (BRKGA). One of the main challenges of BRKGA is its limited capacity to explore the domain surrounding a particular individual. Although good genes will be preserved by its bias property, these genes are copied as they are, and even if a better solution exists in the close neighborhood of a particular gene it might never be explored. We tackle this problem by adding an exploitation component at the end of every evolutionary step, further exploring the hyperparameter domain. Several computational experiments on eight different publicly available datasets were performed to assess the effectiveness of the proposed approach and to prove it is a significant improvement over its predecessor. The results show that our proposed method outperforms, in terms of the$F_{1}$score of the resulting Artificial Neural Network, not only BRKGA but also other commonly used methods in most of the test cases.

This paper presents a hybrid BRKGA (MP-HBRKGA), that combines BRKGA with a Local Branching technique, to solve the multiproduct two-stage capacitated facility location problem (MP-TSCFLP). In this problem, a set of different products has to be transported from a set of factories, passing through a set of depots (first stage) and then transported to a set of customers (second stage). The goal in the MP-TSCFLP is to minimize the opening and transportation costs, where each kind of product has its own transportation cost per unit transported. Recent hybrid methods have been successfully applied to facility location problems, therefore, in this paper we propose adaptations of such hybrid methods and implement the MP-HBRKGA for handling the multiproduct characteristic. To the best of our knowledge, such hybrid BRKGA presented the best results for the single-product problem and have not yet been applied to solve the problem with multiple products. Computational experiments compare the obtained results to those in the literature, using four sets, with different characteristics, of large-sized instances, proposed in the literature.

Finding optimal locations for installing factories and warehouses, fulfilling customers' demand is the goal of the Two-Stage Capacitated Facility Location, which compounds a supply chain class of problems. Besides locating factories and warehouses, the problem's objective is to minimize the operational costs: opening facilities and product flow, satisfying customer demand, and without disregarding the capacity of factories and warehouses. A Hybrid Biased Random-Key Genetic Algorithm (HBRKGA), combining BRKGA with Local Search and Local Branching, was proposed to solve the problem. The computational experiments showed that the method was reliable and stood out among other approaches to the problem.

Most developing countries need to constantly work on the expansion of their electric transmission networks. This task needs to be carefully planned. Unlike several Network Design Problem, a transmission network may become more efficient after cutting-off some of its transmission lines. The version of the problem where the redesign is allowed when expanding is known in the literature as Transmission Expansion Planning Problem with Re-design, TEPr, and will be the focus of this paper. To solve TEPr, we propose a hybridization of Biased Random-Key Genetic Algorithm with Local Branching. Computational experiments showed the impact of the developed method in comparison to the straight forward application of the mathematical formulation.

Due to the constant development of society, increasing quantities of commodities have to be transported in large urban centers. Therefore, network planning problems arise as tools to support decision-making, aiming to meet the need of finding efficient ways to perform such transportations. This paper review a bi-level formulation, an one level formulation obtained by applying the complementary slackness theorem, Bellman's optimality conditions and presents an improved Relax-and-Fix heuristic, through combining a randomized constructive algorithm with a Relax-and-Fix heuristic, so high quality solutions could be found. Besides that, our computational results are compared with the results found by an one-level formulation and other heuristics found in the literature, showing the efficiency of the proposed method.