Cardiovascular disease prediction is delayed by high-dimensional clinical data and heteroginity. There is a need for decision-support system that can select relevant features. We propose a Cardio Meta Hybrid Optimizer (CMHO) framework designed to enhance feature selection and predictive accuracy in cardiac risk assessment.The CMHO framework integrates three metaheuristic algorithms-Lion Optimization (LO), Marine Predators Algorithm (MPA), and Manta Ray Foraging Optimization (MRFO)-enhanced with adaptive switching, dynamic mutation, and iterative local search (ILS). The framework was evaluated on five benchmark datasets: Cleveland, Hungarian, Statlog, Switzerland, and Long Beach VA. We uesd a CNN-LSTM architecture for classification, validated through stratified tenfold cross-validation with 10 independent repetitions. Performance was benchmarked against RFE, GA, PSO, GWO, and Lasso using ANOVA to confirm statistical significance. The CMHO-integrated CNN-LSTM model achieved a accuracy of 96.1%, outperforming traditional feature selection methods by 3%-5% (p < 0.05). The framework demonstrated stability and clinical interpretability by selecting validated biomarkers-including thalassemia, chest pain type, and maximum heart rate-with a Stability Selection Index (SSI) > 0.90.The CMHO framework provides a robust and interpretable tool for cardiovascular risk assessment. By navigating high-dimensional data across diverse populations, it offers a reliable computational approach for clinical decision support in cardiology.

ABSTRACT To improve the accuracy and speed of image segmentation, an improved whale optimization algorithm is proposed. This method introduces chaotic mapping initialization and iterative local search to balance global exploration and local exploitation. It combines Tsallis entropy-guided multi-threshold search and designs an adaptive hybrid median filter preprocessing to reduce noise and preserve details effectively. Experiments show that in a 50-dimensional test, the accuracy of the new algorithm is improved by 18%, 8% and 2% compared to the original, adaptive and chaotic whale algorithms, respectively. The improved filtering increases PSNR by 8% and structural similarity by 0.2. The final algorithm achieves a signal-to-noise ratio of 21.3 dB, a structural similarity of 0.81 and a segmentation accuracy of 93.7%, significantly outperforming the comparison algorithms. The results demonstrate that this method has significant advantages in both quality and real-time performance, providing an efficient solution for image segmentation.

Cluster Head (CH) selection is a crucial process in clustered Wireless Sensor Networks (WSNs) because it directly affects energy balance and network lifetime. However, CH selection is an NP-hard optimization problem, and many metaheuristic-based methods suffer from limited search diversity and premature convergence, leading to uneven energy dissipation. This paper formulates CH selection as a multi-criteria energy-aware optimization problem and proposes an Enhanced Secretary Bird Optimization Algorithm (ESBOA). The proposed ESBOA improves the original Secretary Bird Optimization Algorithm by integrating logistic chaotic map-based population initialization to enhance early-stage exploration and an iterative local search mechanism to strengthen solution refinement in later iterations. A multi-criteria fitness function considering residual energy, distance to the base station, and node degree explicitly guides the optimization toward energy-efficient clustering. The proposed method is implemented in a Python 3.11.9-based simulation framework using a first-order radio energy model and evaluated against standard SBOA, Crested Porcupine Optimization (CPO), and Dung Beetle Optimization (DBO). Simulation results demonstrate that ESBOA preserves more alive nodes, maintains higher residual energy, delivers more cumulative packets to the base station, and extends network lifetime, achieving approximately 3-13% improvement in last node death (LND) compared with the standard SBOA.

Abstract Efficient and robust charging infrastructure plays a key role in accelerating the adoption of electric buses (eBuses) in urban transit systems. Unlike diesel buses, eBuses depend on strategically placed fast-charging stations to maintain continuous operation while ensuring high service quality. Planners must balance infrastructure investment, network reliability, and operational feasibility when determining optimal charging locations. Providing redundant charging access prevents disruptions from station failures, energy consumption fluctuations, and scheduling uncertainties. This work introduces an Iterated Local Search (ILS) algorithm that optimises the number and placement of charging stations. The approach improves robustness by incorporating backup charging stations and flexible energy redistribution techniques. We evaluate performance using real-world public transportation data from Cork and Dublin, comparing results against the state-of-the-art Large Neighborhood Search (LNS) method. Our experiments show that ILS outperforms LNS in 81.2% of cases by requiring fewer charging stations, whereas LNS performs better in 7.8% of cases. ILS is particularly effective in high-energy-demand scenarios with strict timetable constraints and realistic discharging rates, demonstrating its effectiveness in ensuring viable and scalable charging station placement.

Depression detection using Electroencephalogram (EEG) signals obtained from wearable medical-assisted diagnostic systems has become a well-established approach in the field of affective disorders. However, despite recent advancements, on-board Artificial Intelligence (AI) models still demand substantial computational resources, presenting significant challenges for deployment on resource-constrained wearable medical devices. Embedded Multi-core Processors (MPs) offer a promising solution for accelerating these models. However, the limited computational capabilities of embedded MPs, combined with the structural diversity of AI models, complicate resource allocation and increase associated costs. To address these challenges, we propose a Memory-Aware Multi-Objective Iterative Local Search (MAMILS) algorithm to optimize task scheduling, thereby improving the efficiency of AI model deployment on wearable EEG devices. Experimental results across seven AI models demonstrate that, the MAMILS approach yields substantial improvements in key performance indicators: Total Energy Consumption (<inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\bm {TEC}$</tex-math></inline-formula>) with an average reduction of 47.57%, <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\bm {Makespan}$</tex-math></inline-formula> with an average reduction of 48.75%, and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\bm {Throughput}$</tex-math></inline-formula> with an average increase of 198.37%, all while maintaining satisfactory classification performance for both Machine Learning (ML) and Deep Learning (DL) models. Especially, on-board deployment of EEGNeX achieves an accuracy of 93.4%, sensitivity of 91.6%, and specificity of 95.8%. Further analysis indicates that, when integrated with wearable EEG sensors and executable on-board AI models, the proposed MAMILS optimization strategy shows significant promise in facilitating the widespread adoption of low-power, real-time diagnostic systems for depression detection.

Osteoporosis is a silent disease before significant fragility fractures despite its high prevalence, and its screening rate is low. In predictive healthcare analytics, the Elman recurrent neural network (ERNN) has been widely used as a learning technique. Traditional learning algorithms have some limitations, such as slow convergence rates and local minima that prevent gradient descent from finding the global minimum of the error function. The main goal is to precisely estimate each individual's risk of developing osteoporosis. These forecasts are essential for prompt diagnosis and treatment, which have a significant influence on patient outcomes. Hence, the present research focuses on making a more efficient prediction method based on an optimized Elman recurrent neural network (ERNN) for predicting osteoporosis diseases. An optimized ERNN method, IBCO-ERNN, improved bacterial colony optimization (IBCO) by optimizing the ERNN weights and biases. The IBCO approach uses an iterative local search (ILS) algorithm to enhance convergence rate and avoid the local optima problem of conventional BCO. Subsequently, the IBCO is used to optimize the ERNN's weights and biases, thereby improving convergence speed and detection rate. The effectiveness of IBCO-ERNN is evaluated using four different types of osteoporosis datasets: Femoral neck, Lumbar spine, Femoral and Spine, and BMD datasets. The proposed IBCO-ERNN produced higher accuracy at 95.61%, 96.26%, 97.26%, and 97.54 % for the Femoral neck, Lumbar spine, Femoral, and Spine datasets, respectively. The experimental findings demonstrated that, compared with other predictors, the proposed IBCO-ERNN achieved respectable accuracy and rapid convergence.

The increasing adoption of Electric Vehicles (EVs) for service and goods distribution operations has led to the emergence of Electric Vehicle Routing Problems (EVRPs), a class of vehicle routing problems addressing the unique challenges posed by the limited driving range and recharging needs of EVs. While the majority of EVRP variants have considered deterministic energy consumption, this paper focuses on the Stochastic Electric Vehicle Routing Problem with a Threshold recourse policy (SEVRP-T), where the uncertainty in energy consumption is considered, and a recourse policy is employed to ensure that EVs recharge at Charging Stations (CSs) whenever their State of Charge (SoC) falls below a specified threshold. We formulate the SEVRP-T as a two-stage stochastic mixed-integer second-order cone model, where the first stage determines the sequences of customers to be visited, and the second stage incorporates charging activities. The objective is to minimize the expected total duration of the routes, composed by travel times and recharging operations. To cope with the computational complexity of the model, we propose a heuristic based on an Iterated Local Search (ILS) procedure coupled with a Set Partitioning problem. To further speed up the heuristic, we develop two lower bounds on the corresponding first-stage customer sequences. Furthermore, to handle a large number of energy consumption scenarios, we employ a scenario reduction technique. Extensive computational experiments are conducted to validate the effectiveness of the proposed solution strategy and to assess the importance of considering the stochastic nature of the energy consumption. The research presented in this paper contributes to the growing body of literature on EVRP and provides insights into managing the operational deployment of EVs in logistics activities under uncertainty.

Purpose This study develops an optimization model for operational-level aircraft maintenance, repair and overhaul (MRO) scheduling that integrates remanufacturing as a sustainable maintenance option. Design/methodology/approach A mathematical programming model is formulated to optimize MRO scheduling by considering repair, replacement and remanufacturing options for aircraft components. To address the NP-hard nature of the problem, this study introduces an iterative local search (ILS) heuristic as the primary solution methodology, enabling efficient resolution of large-scale scheduling problems while consistently delivering optimal or near-optimal results. Computational experiments on both small-scale and large-scale instances validate the effectiveness of the proposed ILS-based framework. Findings The results demonstrate that the ILS heuristic significantly reduces computational time while consistently achieving optimal or near-optimal solutions for complex MRO scheduling problems. The integration of remanufacturing options leads to cost savings and improved resource utilization, making the scheduling process more sustainable and efficient. Originality/value This study is one of the few to integrate remanufacturing into operational-level MRO scheduling while also developing an efficient heuristic approach to handle large-scale scheduling problems. The proposed ILS-based solution method offers a practical alternative to exact optimization techniques, making it applicable to real-world MRO environments.

This research tackles a crucial aspect of manufacturing system design: optimizing the Facility Layout Problem (FLP). We address a specific scenario involving multiple products with flexible processing plans on various machines in a job-shop environment. Redundant machines of each type exist, with known acquisition costs and capacities. Processing times and production volumes for each product are also pre-determined. An integer non-linear mathematical model is formulated to represent the problem. While a linearization technique is applied, the inherent NP-hardness renders exact solution methods impractical for medium to large-scale problems. To address this, three algorithms are proposed: a matheuristic, Iterated Local Search (ILS), and a Genetic Algorithm (GA). These are evaluated based on solution quality, runtime, and robustness across diverse problem instances. Results demonstrate the superiority of the ILS algorithm in terms of solution quality, robustness, and overall effectiveness. These findings offer valuable guidance for decision-makers seeking optimization tools for FLPs. The ILS’s consistent delivery of high-quality solutions with minimal variation makes it a reliable choice. Additionally, as many facility layout decisions are tactical or strategic – where computational time is less critical – the matheuristic demonstrates acceptable performance and holds promise for handling problems of varying sizes and complexities. To further validate the effectiveness and demonstrate the practical applicability of our proposed solution methodology, the ILS and matheuristic algorithms were applied to a real-world layout design case adapted from the literature. The results once again confirm the strong performance of both methods in terms of solution quality, computational efficiency, and robustness.

Abstract This paper addresses integrated production planning, multi‐quay berth allocation, and quay crane assignment scheduling problems. First, we formulate the problem as a mixed‐integer linear programming (MILP) by extending the relative position formulation presented in the literature to deal with the integrated problem, aiming to maximize the total revenue while considering the demurrage cost. Second, we propose a hybrid MILP‐adaptative large neighborhood search (MILP‐ALNS) approach that we assess using both real‐life instances and randomly generated instances in addition to reported sets from the literature. The results show that the MILP‐ALNS approach performs better than existing iterative local search procedures reported in the literature. In most cases, our approach outperforms CPLEX in terms of time to reach the best solution known while providing comparable solutions.

Urban logistics face complexity due to traffic congestion, fleet heterogeneity, warehouse constraints, and driver workload balancing, especially in the Heterogeneous Multi-Trip Vehicle Routing Problem with Time Windows and Time-Varying Networks (HMTVRPTW-TVN). We develop a mixed-integer linear programming (MILP) model with dual-peak time discretization and exact linearization for heterogeneous fleet coordination. Given the NP-hard nature, we propose a Hyper-Heuristic based on Cumulative Reward Q-Learning (HHCRQL), integrating reinforcement learning with heuristic operators in a Markov Decision Process (MDP). The algorithm dynamically selects operators using a four-dimensional state space and a cumulative reward function combining timestep and fitness. Experiments show that, for small instances, HHCRQL achieves solutions within 3% of Gurobi’s optimum when customer nodes exceed 15, outperforming Large Neighborhood Search (LNS) and LNS with Simulated Annealing (LNSSA) with stable, shorter runtime. For large-scale instances, HHCRQL reduces gaps by up to 9.17% versus Iterated Local Search (ILS), 6.74% versus LNS, and 5.95% versus LNSSA, while maintaining relatively stable runtime. Real-world validation using Shanghai logistics data reduces waiting times by 35.36% and total transportation times by 24.68%, confirming HHCRQL’s effectiveness, robustness, and scalability.

Efficient production scheduling is a key challenge in industrial operations and continues to attract significant interest within the field of operations research. This paper investigates a range of methodological approaches designed to solve the permutation flow shop scheduling problem (PFSP) with sequence-dependent setup times (SDST). The main objective is to minimize the total weighted flow time (TWFT) while ensuring a no-wait production environment. The proposed solution strategy is based on using algorithms with a mixed integer linear programming (MILP) formulation, heuristics, and their combination. The heuristics utilized in this paper include an advanced greedy randomized adaptive search procedure (GRASP) based on a priority rule and Hybrid-GRASP-NEH (HGRASP), where Nawaz-Enscore-Ham (NEH) takes place to initiate solutions, based on iterative global and local search methods to refine exploration capabilities and improve solution quality. These approaches were validated using a comprehensive set of experiments across diverse instance sizes that proved the efficiency of HGRASP, with the results showing a high-performance level that closely matched that of the exact MILP approach. Statistical analysis via the Friedman test (χ2 = 46.75, p = 7.04 × 10−11) confirmed significant performance differences among MILP, GRASP, and HGRASP. While MILP guarantees theoretical optimality, its practical effectiveness was limited by imposed computational time constraints, and HGRASP consistently achieved near-optimal solutions with superior computational efficiency, as demonstrated across diverse instance sizes.

The Traveling Purchaser Problem (TPP) is a generalization of the Traveling Salesman Problem (TSP) in which a list of items must be acquired by visiting a subset of markets. The objective is to minimize the total cost sustained along the route, including purchasing and traveling costs. Due to the NP-hard nature of the problem, solving the TPP in an exact manner is computationally challenging, implying the need for heuristic approaches to obtain quality solutions efficiently. This study proposes an algorithm based on the metaheuristic Iterated Local Search (ILS), complemented by a route configuration procedure that adjusts the subset of markets in the solution. The ILS is tested in benchmark instances, providing a performance comparison with other methods. The computational experiment for the asymmetric instances reveals the effectiveness and efficiency of the ILS, outperforming previously published results with statistical significance. Additional experiments are presented for the symmetric instances, pointing to the competitiveness and versatility of the ILS in relation to other heuristic approaches used in the literature. • We propose a metaheuristic approach for the unrestricted traveling purchaser problem. • We introduce a novel procedure that limits the number of neighborhood searches. • We outperform the best results in the literature for the asymmetric instances. • We solve a subset of symmetric instances to optimality. • We provide competitive results for the euclidean symmetric instances.

In the complex indoor environment, geomagnetic matching is an effective way to realize indoor positioning of mobile robots. Aiming at the problem that the application of Particle Swarm Optimization (PSO) algorithm leads to the decline of geomagnetic matching accuracy, stability and convergence speed, an Iterated Local Search-Improved Particle Swarm Optimization (ILS-IPSO) algorithm is proposed. By analyzing the timefrequency characteristics and data distribution characteristics of geomagnetic survey data, geomagnetic data preprocessing and geomagnetic reference map construction are carried out. By introducing 3σ contour Search domain constraint in the particle swarm optimization process, the weight factor, learning factors, step control factor of PSO algorithm are optimized, and finally the Local disturbance and search are implemented in combination with Iterated Local Search (ILS) algorithm. The experimental results show that the average matching accuracy error of ILSIPSO is reduced to 0.0508m, and the standard deviation of matching error is reduced to 0.0198m. Compared with PSO, Linear Dynamic time-varying Inertial Weight Particle Swarm Optimization algorithm (LDIW-PSO) and Cosine Decreasing Inertia Weight Particle Swarm Optimization algorithm (CDIWPSO) algorithms, the average matching accuracy is increased by 94.72%, 92.37% and 87.98%, the standard deviation of matching error decreased by 87.13%, 83.26% and 89.81% respectively. The optimal fitness of ILS-IPSO algorithm is increased by 79.51%, 61.81% and 57.06%, and the iteration efficiency is increased by 69.23%, 55.56% and 33.33%, respectively. This method performs well in the accuracy, stability and convergence of geomagnetic positioning, and can be widely used in the field of indoor positioning

Abstract Given an empty delivery vehicle, the backhaul profit maximization problem (BPMP) is to select a profit-maximizing subset of available pick-up-and-delivery requests to accept considering the vehicle’s capacity and a time limit for the vehicle to reach a specified destination or, equivalently, a driving-distance limit. Implemented in our computing environment, the fastest known exact algorithm for BPMP requires approximately 11 hours and 44 minutes on average to solve the largest instances in the literature, which have 70 to 80 potential pick-up/drop-off locations. The fastest available heuristic from the literature is considerably faster, and finds high quality solutions, but requires a state-of-the-art mixed-integer programming solver. We present a heuristic framework for the BPMP based on greedy construction, iterative local search, and randomization. Algorithms developed with the framework are implemented in the freely and widely available C++ language and their effectiveness is demonstrated through an extensive computational experiment on both benchmark and randomly generated problem instances. We find that our approach is competitive with approaches from the literature in solution quality as well as running time.

In this study, the ordered flow shop scheduling problem, which is in the class of NP-hard optimization problems, is considered. This problem is used especially to increase the efficiency and prevent delays in the production process. The problem was first identified in the literature during the 1970s. The main objective of this study is to develop an efficient and fast method to overcome the complexity of this problem. For this purpose, the ordered flow shop scheduling problem is explained in detail and a robust meta-heuristic method is proposed. First of all, a genetic algorithm is developed by considering Smith’s convexity criterion. While performing operations such as crossover and mutation in the genetic algorithm, the pyramid structure is integrated to ensure that the solution has certain symmetry. The developed method is compared with other methods, such as the Nawaz–Enscore–Ham (NEH), pair insert, and iterated local search (ILS) methods. In order to increase the reliability of the results, the Pyramid Structure Adapted Tabu Search (PSA-TS) algorithm is also developed. The results are validated by statistical analysis using the Wilcoxon signed-rank test and Friedman test. The proposed genetic algorithm outperforms the methods with which it is compared. To the best of the authors’ knowledge, there is no other method in the literature that preserves the pyramid structure in the ordered flow shop scheduling problem. Therefore, this study is expected to make a significant contribution to the literature in this respect.

Remanufacturing has become a mainstream sustainable manufacturing paradigm for energy conservation and environmental protection. Disassembly and reprocessing operations are two main activities in remanufacturing. This work proposes multiobjective integrated scheduling of disassembly and reprocessing operations considering product structures and random processing time. First, a stochastic programming model is developed to minimize maximum completion time and total tardiness. Second, a reinforcement learning-based multiobjective evolutionary algorithm is devised considering problem-specific knowledge. Three search strategy combinations are formed: crossover and mutation, crossover and key product-based iterated local search, mutation and key product-based iterated local search. At each iteration, a Q-learning method is devised to intelligently choose a combination of premium strategies. A stochastic simulation is incorporated to evaluate the objective values of the searched solutions. Finally, the formulated model and method are compared with an exact solver, CPLEX, and three well-known metaheuristics from the literature on a set of test instances. The results confirm the excellent competitiveness of the developed model and algorithm for solving the considered problem.

Antarctica’s unique research environment necessitates innovative project selection and scheduling strategies to maximize scientific output while addressing logistical, environmental, and resource constraints. This study introduces a novel framework for the Research Project Selection and Scheduling in Multiple Antarctic Stations Problem (RPSAP), formulated as a mixed-integer programming model. The model incorporates resource-sharing constraints, station-specific capacities, transportation delays, and sustainability considerations. Given the problem’s NP-hard complexity, three metaheuristic methods—Iterated Local Search (ILS), Variable Neighborhood Search (VNS), and Simulated Annealing (SA)—were developed to efficiently solve large-scale instances. Metaheuristics demonstrated robust performance through extensive computational experiments involving 480 test instances across 60 classes. The ILS consistently outperformed others in solution quality and scalability, while SA offered competitive results in execution time. Results reveal that the metaheuristics are superior to exact methods in handling large problem sizes, with optimality achieved only for small instances using the proposed exact model. This research bridges the gap between theoretical optimization models and practical applications, offering decision-making tools for research managers to enhance resource utilization, minimize ecological impacts, and prioritize high-impact projects.

ABSTRACT The Vehicle Routing Problem (VRP) faces numerous challenges, including high-traffic areas, restricted access for cargo vehicles, high pollutant emissions, pedestrian-only zones, and limited parking. The VRP with Unmanned Vehicles seeks to address these issues by incorporating vehicles attached to the primary delivery vehicle. These attached vehicles encounter fewer access restrictions in delivery areas but have limited load capacity and range. This paper proposes formulating a two-echelon VRP using Unmanned Aerial Vehicles (UAVs) and Unmanned Ground Vehicles (UGVs) to compare their performance against the classical Capacitated Vehicle Routing Problem (CVRP). The objective is to identify the benefits—such as reduced pollutant emissions—of using these vehicles, and to highlight the advantages and disadvantages of each type of attached vehicle for the second echelon. The development is carried out in three phases. First, the problem is solved using the classical vehicle routing problem for the first echelon. Then, clusters are generated to reconfigure the routes from the initial solution, assign customers to the second echelon, and verify feasibility. Finally, a modified Iterated Local Search (ILS) is used to obtain solutions for the two-echelon problem. Performance and quality control are conducted using benchmark sets for the classical VRP to validate the proposed methodology.

In this study, an innovative prediction framework for intelligent management and allocation decisions of tourism resources is constructed by integrating the iterative local search strategy and the sparrow search algorithm to co-optimize the random forest model. The experimental results show that the tourism resource allocation level and the number of tourists show significant correlation (Pearson coefficient 0.73), and this indicator as a core driver directly affects the resource allocation decision; the resource utilization rate, as a secondary correlation variable (correlation coefficient 0.54), and the scenic spot operational efficiency form a two-way feedback mechanism, which together constitute the key influencing factors of the dynamic allocation of tourism resources. Through the quantitative analysis of the confusion matrix, the optimized model achieves 100% perfect fit in the training set, and maintains 95% prediction accuracy (190/200 samples are correctly classified) in the independent test set, and the performance degradation of only 5% fully proves the success of the algorithmic improvement strategy in avoiding overfitting, and its excellent generalization characteristic breaks through the limitations of the traditional prediction model in cross-scene applications.