Adaptive Search Research Articles

An Adaptive Search Strategy Combination Algorithm Based on Reinforcement Learning and Neighborhood Search

Abstract Integrating multiple search operators to utilize their different characteristics in order to improve the performance of evolutionary algorithms is a challenging task. This paper proposes an adaptive combination algorithm that integrates four search operators, called RLACA. RLACA introduces a reinforcement learning-based adaptive search operator selection mechanism (RLAS) to dynamically choose the most suitable search operator based on the individual states. Additionally, a neighborhood search strategy based on differential evolution (NSDE) is incorporated to mitigate premature convergence by increasing population diversity. To verify the effectiveness of the proposed algorithm, a comprehensive testing was conducted using the CEC2017 test suite. The experimental results demonstrate that RLAS can adaptively select a suitable search operator and NSDE can enhance the algorithm’s local search capability, thereby improving the performance of RLACA. Compared with the four basic algorithms and four combination algorithms, RLACA performs better in both convergence speed and resolution accuracy.

Risk-Aware Vessel Scheduling and Routing Optimization with CVaR and LSTM-MSNet Prediction

This paper proposes an integrated optimization model for vessel scheduling and routing. The objective is to maximize shipping company profits while considering profit volatility using the Conditional Value-at-Risk metric to master risks from demand fluctuations. Simultaneously, the model balances the spot and contract container allocation by optimally adjusting shipping speeds so as to minimize carbon emissions. We account for vessel deployment, chartering costs, delay penalties, fuel expenses, and weather conditions to ensure the model’s compatibility with the practical transporting environment. In particular, a hybrid demand prediction model, combining long short-term memory and multi-scale network techniques, predicts spot and contract container volumes at ports, facilitating real-time allocation and more precise scheduling optimization. Two hybrid heuristics, one adaptive large-neighborhood search algorithm, and the Gurobi solver are devised and compared based on the efficiency and accuracy of solving the model. The results indicate that our optimization offers practical insights for shipping companies, enabling them to achieve a better trade-off between profits and risks, promoting a promising maritime transport career.

An enhanced dung beetle optimizer with adaptive node selection and dynamic step search for mobile robots path planning

An enhanced dung beetle optimizer with adaptive node selection and dynamic step search for mobile robots path planning

Designing the Liner Shipping Network of Tomorrow Powered by Alternative Fuels

The liner shipping industry is undergoing an extensive decarbonization process to reduce its 275 million tons of CO2 emissions as of 2018. In this process, the long-term solution is the introduction of new alternative maritime fuels. The introduction of alternative fuels presents a great set of unknowns. Among these are the strategic concerns regarding sourcing of alternative fuels and, operationally, how the new fuels might affect the network of shipping routes. We propose a problem formulation that integrates fuel supply/demand into the liner shipping network design problem. Here, we present a model to determine the production sites and distribution of new alternative fuels—we consider methanol and ammonia. For the network design problem, we apply an adaptive large neighborhood search combined with a delayed column generation process. In addition, we wish to test the effect of designing a robust network under uncertain demand conditions because of the problem’s strategic nature and importance. Therefore, our proposed solution method will have a deterministic and stochastic setup when we apply it to the second-largest multihub instance, WorldSmall, known from LINER-LIB. In the deterministic setting, our proposed solution method finds a new best solution to three instances from LINER-LIB. For the main considered WorldSmall instance, we even noticed a new best solution in all our tested fuel settings. In addition, we note a profit drop of 7.2% between a bunker-powered and pure alternative fuel–powered network. The selected alternative fuel production sites favor a proximity to European ports and have a heavy reliance on wind turbines. The stochastic results clearly showed that the found networks were much more resilient to the demand changes. Neglecting the perspective of uncertain demand leads to highly fluctuating profits. Supplemental Material: The online appendix is available at https://doi.org/10.1287/trsc.2023.0143 .

Robust Optimization for Electric Vehicle Routing Problem Considering Time Windows Under Energy Consumption Uncertainty

Compared to fossil fuel-based internal combustion vehicles, electric vehicles with lower local pollution and noise are becoming more and more popular in urban logistic distribution. When electric vehicles are involved, high-quality delivery depends on energy consumption. This research proposes an electric vehicle routing problem considering time windows under energy consumption uncertainty. A mixed-integer programming model is established. The robust optimization method is adopted to deal with the uncertainty. Based on the modification of adaptive large neighborhood search algorithm, a metaheuristic procedure, called novel hybrid adaptive large neighborhood search, is designed to solve the problem, and some new operators are proposed. The numerical experiments show that the proposed metaheuristic can obtain high-performance solutions with high efficiency for large-scale instances. Furthermore, the robust solution based on the proposed model can achieve a satisfactory tradeoff between performance and risk.

Optimizing the Efficacy of Vaccine-Induced Immunotherapy in Melanomas.

Cancer therapeutic vaccines are used to strengthen a patient's own immune system by amplifying existing immune responses. Intralesional administration of the bacteria-based emm55 vaccine together with the PD1 checkpoint inhibitor produced a strong anti-tumor effect against the B16 melanoma murine model. However, it is not trivial to design an optimal order and frequency of injections for combination therapies. Here, we developed a coupled ordinary differential equations model calibrated to experimental data and used the mesh adaptive direct search method to optimize the treatment protocols of the emm55 vaccine and anti-PD1 combined therapy. This method determined that early consecutive vaccine injections combined with distributed anti-PD1 injections of decreasing separation time yielded the best tumor size reduction. The optimized protocols led to a twofold decrease in tumor area for the vaccine-alone treatment, and a fourfold decrease for the combined therapy. Our results reveal the tumor subpopulation dynamics in the optimal treatment condition, defining the path for efficacious treatment design. Similar computational frameworks can be applied to other tumors and other combination therapies to generate experimentally testable hypotheses in a fairly unrestricted and inexpensive setting.

Minimizing durations in repetitive projects through adaptive large neighborhood search

This study proposes an integer linear programming model based on line of balance, which takes the makespan as the optimization objective and characterizes various construction scenarios that may exist in different types of repetitive projects. The model is easy to solve, convenient for on-site use, and provides a solid foundation for exploring theoretical optimal solutions. Further, a novel matheuristic algorithm that integrates adaptive large neighborhood search with exact algorithms is proposed. Based on two different types of repetitive projects, the practicality of the model and the effectiveness of the algorithm are verified in ten different construction scenarios, and managerial insights are provided. A comparison with Gurobi’s results shows that in small-scale case scenarios, the matheuristic algorithm achieves solutions of the same quality with a shorter running time. In large-scale scenarios, the matheuristic algorithm outperforms Gurobi in terms of both solution quality and computational efficiency.

Applying a Modified Adaptive Large Neighborhood Search for Truck Scheduling and Pile Assignment in a Two-Stage Sorting System

Applying a Modified Adaptive Large Neighborhood Search for Truck Scheduling and Pile Assignment in a Two-Stage Sorting System

Inter-frame feature fusion enhanced spatio-temporal consistent video inpainting with sample-based techniques and adaptive local search

Inter-frame feature fusion enhanced spatio-temporal consistent video inpainting with sample-based techniques and adaptive local search

Green pickup and delivery problem with private drivers for crowd-shipping distribution considering traffic congestion

Crowd-shipping, employing private drivers to partially replace company-owned trucks in distribution, has emerged as a prominent trend for its cost-effectiveness and sustainability. While crowd-shipping is known as a distribution pattern that combines economic efficiency and environmental benefits, however, the frequent occurrence of traffic congestion has made this pattern less effective than it should be. In this research, the problem of vehicle routing optimization under traffic congestion is investigated from the perspective of simultaneously reducing environmental pollution and costs. Considering private drivers picking up and delivering parcels on the way, this study incorporates the objective of minimizing transport as well as particulate matter (PM) and nitrogen oxides (NOx) emission costs into route optimization for crowd-shipping and proposes a Green Pickup and Delivery Problem with Private Drivers (GPDP-PD). To be more realistic, vehicle speeds depend on the level of traffic congestion, reflecting the time-dependent nature of the proposed model. An improved adaptive large neighborhood search (ALNS) algorithm is developed, and computational experiments are conducted to demonstrate the efficiency of the improved ALNS. Case studies show that there is uncertainty about the environmental benefits of crowd-shipping under traffic congestion. Our proposed model is capable of efficiently allocating private drivers and optimizing vehicle routes according to road conditions, thus identifying the crowd-shipping operational scheme with the lowest cost and emissions. Moreover, a time limit of 0.7-0.8 h and the low cost of private drivers can achieve environmental and economic benefits simultaneously. It provides useful insights into the sustainability of logistics and distribution.

Adaptive large multiple neighborhood search for the truck-drone-collection station collaborative delivery problem

Adaptive large multiple neighborhood search for the truck-drone-collection station collaborative delivery problem

A novel Adaptive Jellyfish Search algorithm (AJFS) for maximum power optimization of photovoltaic systems: performance evaluation and analysis

ABSTRACT In solar photovoltaic (SPV) systems, optimizing output under various meteorological conditions relies on the Maximum Power Point (MPP) tracking controller. However, the presence of multiple peaks due to partial shading complicates this tracking process. While traditional and soft computing methods are commonly employed for MPP tracking, they face limitations such as the fixed step size in conventional methods and a lack of randomness in soft computing approaches once they reach a certain MPP. To address these challenges, a unique optimization technique known as the Adaptive Jellyfish Search (AJFS) method has been proposed. This method conducts both global and local searches simultaneously in a single step, enhancing the effectiveness of MPP tracking. To evaluate its performance and resilience, zero, weak, moderate, and strong shading patterns are employed in simulations, with comparisons made against traditional Jellyfish Search (JFS) and Particle Swarm Optimization (PSO) techniques. The newly suggested AJFS approach demonstrates significant improvements over traditional methods. It reduces convergence time by 49% and offers additional motivating features, including zero risk of failure, minimized oscillation in power parameters, and enhanced energy production by 1.5%. Moreover, it enables smooth tracking of the MPP, particularly in dynamically changing shading patterns. Overall, the AJFS method presents a promising solution for efficient MPP tracking in SPV systems, overcoming limitations of conventional approaches and demonstrating superior performance under various shading conditions. Its ability to simultaneously conduct global and local searches in a single step makes it well suited for optimizing energy production in real-world scenarios.

Optimizing Multi-Depot Mixed Fleet Vehicle–Drone Routing Under a Carbon Trading Mechanism

The global pursuit of carbon neutrality requires the reduction of carbon emissions in logistics and distribution. The integration of electric vehicles (EVs) and drones in a collaborative delivery model revolutionizes last-mile delivery by significantly reducing operating costs and enhancing delivery efficiency while supporting environmental objectives. This paper presents a cost-minimization model that addresses transportation, energy, and carbon trade costs within a cap-and-trade framework. We develop a multi-depot mixed fleet, including electric and fuel vehicles, and a drone collaborative delivery routing optimization model. This model incorporates key factors such as nonlinear EV charging times, time-dependent travel conditions, and energy consumption. We propose an adaptive large neighborhood search algorithm integrating spatiotemporal distance (ALNS-STD) to solve this complex model. This algorithm introduces five domain-specific operators and an adaptive adjustment mechanism to improve solution quality and efficiency. Our computational experiments demonstrate the effectiveness of the ALNS-STD, showing its ability to optimize routes by accounting for both spatial and temporal factors. Furthermore, we analyze the influence of charging station distribution and carbon trading mechanisms on overall delivery costs and route planning, underscoring the global significance of our findings.

CCOA‐AdaLS: Hybrid Beamforming Using Chaotic Chebyshev Aquila Optimization for mmWave Massive MIMO

ABSTRACTThis research aims to design hybrid analog and digital beamforming to improve the signal‐to‐noise ratio (SNR) and spectral efficiency (SE) of communication links. Considering the complexity and cost associated with fully connected multiple‐input multiple‐output (MIMO) communication models, a partially connected system model is adopted for the downlink millimeter‐wave (mmWave) communication model. The analog beamforming utilizes the adaptive search (AdaLS) algorithm to minimize interference and enhance user power, whereas digital beamforming is optimized using the proposed Chaotic Chebyshev Aquila Optimization (CCAO) algorithm. The CCAO algorithm integrates chaotic Chebyshev–based solution mapping with the conventional Aquila Optimization Algorithm to enhance exploration capability. The system model is illustrated, and the problem is formulated to maximize the signal‐to‐interference‐plus‐noise ratio (SINR) through the selection of the required signal at the receiver. The digital beamformer is designed using Lagrange's multiplier, and the analog beamformer is optimized using AdaLS. The proposed CCAO algorithm is detailed, incorporating chaotic dynamics to explore the solution space effectively. The research evaluates the performance of the proposed method against conventional approaches, showcasing improved normalized beam gain and SINR.

Qoolchain: A QUBO Preprocessing Toolchain for Enhancing Quantum Optimization

AbstractSolving combinatorial optimization problems is crucial in research and industry but still challenging since these problems are usually NP‐hard or NP‐complete. Classical solvers struggle with their non‐polynomial complexity. Although heuristic algorithms are widely used, they often fall short in execution time and accuracy, increasing the interest in quantum computing alternatives using Quadratic Unconstrained Binary Optimization (QUBO) formulations. However, current Noisy Intermediate‐Scale Quantum (NISQ) computers and future early fault‐tolerant quantum devices face limitations in qubit availability and circuit depth, necessitating preprocessing to reduce problem complexity. This study introduces Qoolchain, a QUBO preprocessing toolchain designed to reduce problem size and enhance solver performance. Developed in Cython, Qoolchain is compatible with major quantum frameworks and optimized for the Grover Adaptive Search (GAS) algorithm. It includes steps like persistency identification, decomposition, and probing to estimate function bounds, all with polynomial complexity. Qoolchain also proposes using the Grover Search algorithm for problem segments whose optimal value is known a priori from graph theory and Shannon decomposition to reduce QUBO problem complexity further. Evaluated against the D‐Wave preprocessing toolchain on various problems, Qoolchain demonstrates higher efficiency and accuracy. It represents a significant advancement in enabling practical quantum solvers, addressing hardware limitations, and solving complex industry‐relevant problems.

Weak compound fault identification of gearboxes based on improved symplectic geometric mode decomposition and optimized cyclic kurtosis deconvolution

Abstract Given the complex and harsh operating conditions of gear transmission systems, gearboxes are prone to compound faults. These faults, involving multiple types, often couple together, causing the weak fault pulse characteristics to be entirely masked by strong environmental noise. This significantly complicates the extraction of relevant fault information from the gearbox. To address these issues, this paper proposes a method for extracting compound fault features based on improved symplectic geometric mode decomposition (SGMD) and optimized cyclic bispectrum deconvolution (CYCBD). Firstly, considering the periodic impact characteristics of different fault types, morphological envelope cyclic bispectrum is proposed to cluster the initial components obtained from SGMD decomposition, thereby adaptively separating different fault features contained in the compound fault signals. An adaptive filter length search strategy is subsequently introduced to optimize the CYCBD by deconvolving each initial fault component, thereby eliminating the interference caused by complex transmission paths and substantial environmental noise, which, in turn, enhances weak periodic fault pulses. Following this, the enhanced signals are subjected to envelope demodulation to extract fault characteristic frequencies, enabling the identification of various types of faults. The effectiveness and feasibility of the proposed method are demonstrated through both simulation signals and actual experimental data related to gearbox compound faults. Compared with existing methods, the proposed method demonstrates superior performance in identifying weak compound faults under strong environmental noise.

A novel feature selection method based on adaptive search particle swarm optimization

A novel feature selection method based on adaptive search particle swarm optimization

Adaptive Neural Architecture Search Using Meta-Heuristics: Discovering Fine-Tuned Predictive Models for Photocatalytic CO2 Reduction

This study aims to contribute to the reduction of carbon dioxide and the production of hydrogen through an investigation of the photocatalytic reaction process. Machine learning algorithms can be used to predict the hydrogen yield in the photocatalytic carbon dioxide reduction process. Although regression-based approaches provide good results, the accuracy achieved with classification algorithms is not very high. In this context, this study presents a new method, Adaptive Neural Architecture Search (NAS) using metaheuristics, to improve the capacity of ANNs in estimating the hydrogen yield in the photocatalytic carbon dioxide reduction process through classification. The NAS process was carried out with a tool named HyperNetExplorer, which was developed with the aim of finding the ANN architecture providing the best prediction accuracy through changing ANN hyperparameters, such as the number of layers, number of neurons in each layer, and the activation functions of each layer. The nature of the NAS process in this study was adaptive, since the process was accomplished through optimization algorithms. The ANNs discovered with HyperNetExplorer demonstrated significantly higher prediction performance than the classical ML algorithms. The results indicated that the NAS helped to achieve better performance in the estimation of the hydrogen yield in the photocatalytic carbon dioxide reduction process.

Dynamically dealing with requests in a stochastic multi-period home healthcare problem with consistency constraints

This paper analyzes a Multi-Period Stochastic Vehicle Routing Problem in the healthcare sector. Patients with unknown locations and demands ask for domiciliary care services with unknown temporal distributions. Requests from patients arrive over time to a nurse agency that has to plan the activities of a fleet of nurses over several days. Based on daily information, the agency decides if a new patient can be accepted, earning the corresponding revenue, or assigned to an external provider. To guarantee high service satisfaction, the agency schedules the nurses' routing by guaranteeing consistency in nurse-patient assignments. The problem aims to plan nurses' routing to satisfy all requests of accepted patients while maximising the total profit measured as the difference between collected revenue and travelling costs. We propose different solution methodologies that either sequentially make short-sighted decisions or use a scenario-based strategy, leveraging historical data to predict future requests. All algorithms make use of an Adaptive Large Neighborhood Search and are validated on medium-sized instances. Managerial insights on the impact of consistency on the profit and its relation to date flexibility in patients' requests are provided. Interesting rules of thumb are derived from a case study conducted in the city of Brescia, Italy.

Optimizing Multi-Echelon Delivery Routes for Perishable Goods with Time Constraints

As the logistics industry modernizes, living standards improve, and consumption patterns shift, the demand for fresh food continues to grow, making cold chain logistics for perishable goods a critical component in ensuring food quality and safety. However, the presence of both soft and hard time windows among demand nodes can complicate the single-network distribution of perishable goods. In response to these challenges, this paper proposes an optimization model for multi-distribution center perishable goods delivery, considering both one-echelon and two-echelon network joint distributions. The model aims to minimize total costs, including transportation, fixed, refrigeration, goods damage, and penalty costs, while measuring customer satisfaction by the start time of service at each demand node. A two-stage heuristic algorithm is designed to solve the model. In the first stage, an initial solution is constructed using a greedy approach based on the principles of the k-medoids clustering algorithm, which considers both spatial and temporal distances. In the second stage, the initial routing solution is optimized using a linear programming approach from the Ortools solver combined with an Improved Adaptive Large Neighborhood Search (IALNS) algorithm. The effectiveness of the proposed model and algorithm is validated through a case study analysis. The results demonstrate that the initial solutions obtained through the k-medoids clustering algorithm based on spatio-temporal distance improved the overall cost optimization by 1.85% and 4.74% compared to the other two algorithms. Among the three two-stage heuristic algorithms, the Ortools-IALNS proposed here showed enhancements in the overall cost optimization over the IALNS, with improvements of 3.24%, 1.12%, and 0.41%, respectively. The two-stage heuristic algorithm designed in this study also converged faster than the other two heuristic algorithms, with overall optimization improvements of 1.55% and 1.28%, further validating the superior performance of the proposed heuristic algorithm.