Neighborhood Strategy Research Articles

Boosting crayfish algorithm based on halton adaptive quadratic interpolation and piecewise neighborhood for complex optimization problems

The Crayfish Optimization Algorithm (CFish) is an innovative meta-heuristic approach that draws inspiration from the movements and behaviors of crayfish. CFish exhibits strong performance across many test sets and optimization issues, but it faces challenges with sluggish convergence, an uneven distribution between exploration and exploitation, and inadequate accuracy due to high-dimensional tasks. To tackle these concerns, this study presents an evolved version named QICFish, which integrates multiple sophisticated strategies to optimize performance. The CFish algorithm improves its population initialization by using the Halton sequence, resulting in a substantial enhancement of the exploration phase and an increase in population variety. Furthermore, the implementation of the adaptive quadratic interpolation approach enhances the algorithm's capacity to exploit its surroundings by eliminating individuals of worse quality and expediting the creation of solutions of superior quality. Furthermore, an adaptive mutation method is employed to improve the process of exploration and effectively locate more favorable areas inside the search space. Finally, the adaptive piecewise neighborhood strategy, enhances the pace of convergence and maintains a balanced transition between exploration and exploitation. Then, QICFish outperforms other cutting-edge algorithms across several dimensions of the CEC’17 and CEC’20 test sets, as shown by experimental comparisons. Furthermore, the efficacy and feasibility of the solution are confirmed by successfully addressing six intricate engineering problems and two truss topology optimization problems. The simulation findings demonstrate that QICFish exhibits robust competitive capabilities and shows great potential for engineering optimization challenges. Therefore, QICFish is a very efficient meta-heuristic approach for solving engineering optimization problems.

DCWPSO: particle swarm optimization with dynamic inertia weight updating and enhanced learning strategies.

Particle swarm optimization (PSO) stands as a prominent and robust meta-heuristic algorithm within swarm intelligence (SI). It originated in 1995 by simulating the foraging behavior of bird flocks. In recent years, numerous PSO variants have been proposed to address various optimization applications. However, the overall performance of these variants has not been deemed satisfactory. This article introduces a novel PSO variant, presenting three key contributions: First, a novel dynamic oscillation inertia weight is introduced to strike a balance between exploration and exploitation; Second, the utilization of cosine similarity and dynamic neighborhood strategy enhances both the quality of solution and the diversity of particle populations; Third, a unique worst-best example learning strategy is proposed to enhance the quality of the least favorable solution and consequently improving the overall population. The algorithm's validation is conducted using a test suite comprised of benchmarks from the CEC2014 and CEC2022 test suites on real-parameter single-objective optimization. The experimental results demonstrate the competitiveness of our algorithm against recently proposed state-of-the-art PSO variants and well-known algorithms.

A bi-level model and heuristic techniques with various neighborhood strategies for covering interdiction problem with fortification

A bi-level model and heuristic techniques with various neighborhood strategies for covering interdiction problem with fortification

Multi-Objective Optimization of Resilient, Sustainable, and Safe Urban Bus Routes for Tourism Promotion Using a Hybrid Reinforcement Learning Algorithm

Urban transportation systems in tourism-centric cities face challenges from rapid urbanization and population growth. Efficient, resilient, and sustainable bus route optimization is essential to ensure reliable service, minimize environmental impact, and maintain safety standards. This study presents a novel Hybrid Reinforcement Learning-Variable Neighborhood Strategy Adaptive Search (H-RL-VaNSAS) algorithm for multi-objective urban bus route optimization. Our mathematical model maximizes resilience, sustainability, tourist satisfaction, and accessibility while minimizing total travel distance. H-RL-VaNSAS is evaluated against leading optimization methods, including the Crested Porcupine Optimizer (CPO), Krill Herd Algorithm (KHA), and Salp Swarm Algorithm (SSA). Using metrics such as Hypervolume and the Average Ratio of Pareto Optimal Solutions, H-RL-VaNSAS demonstrates superior performance. Specifically, H-RL-VaNSAS achieved the highest resilience index (550), sustainability index (370), safety score (480), tourist preferences score (300), and accessibility score (2300), while minimizing total travel distance to 950 km. Compared to other methods, H-RL-VaNSAS improved resilience by 12.24–17.02%, sustainability by 5.71–12.12%, safety by 4.35–9.09%, tourist preferences by 7.14–13.21%, accessibility by 4.55–9.52%, and reduced travel distance by 9.52–17.39%. This research offers a framework for designing efficient, resilient, and sustainable public transit systems that align with urban planning and transportation goals. The integration of reinforcement learning with VaNSAS significantly enhances optimization capabilities, providing a valuable tool for mathematical and urban transportation research communities.

On solving close enough orienteering problems with overlapped neighborhoods

The Close Enough Traveling Salesman Problem (CETSP) is a well-known variant of the classic Traveling Salesman Problem whereby the agent may complete its mission at any point within a target neighborhood. Heuristics based on overlapped neighborhoods, known as Steiner Zones (SZ), have gained attention in addressing CETSPs. While SZs offer effective approximations to the original graph, their inherent overlap imposes constraints on the search space, potentially conflicting with global optimization objectives. Here we show how such limitations can be converted into advantages in the Close Enough Orienteering Problem (CEOP) by aggregating prizes across overlapped neighborhoods. We further extend the classic CEOP with Non-uniform Neighborhoods (CEOP-N) by introducing non-uniform cost considerations for prize collection. To tackle CEOP (and CEOP-N), we develop a new approach featuring a Randomized Steiner Zone Discretization (RSZD) scheme coupled with a hybrid algorithm based on Particle Swarm Optimization (PSO) and Ant Colony System (ACS) — CRaSZe-AntS. The RSZD scheme identifies sub-regions for PSO exploration, and ACS determines the discrete visiting sequence. We evaluate the RSZD’s discretization performance on CEOP instances derived from established CETSP instances and compare CRaSZe-AntS against the most relevant state-of-the-art heuristic focused on single-neighborhood optimization for CEOP instances. We also compare the performance of the interior search within SZs and the boundary search on individual neighborhoods in the context of CEOP-N. Our experimental results show that CRaSZe-AntS can yield comparable solution quality with significantly reduced computation time compared to the single neighborhood strategy, where we observe an averaged 140.44% increase in prize collection and 55.18% reduction of algorithm execution time. CRaSZe-AntS is thus highly effective in solving emerging CEOP-N, examples of which include truck-and-drone delivery scenarios.

Incorporating historical information into the multi-type ant colony optimization model to optimize patch-level land use allocation

Land use optimization is a crucial approach for promoting sustainable development. However, current land use optimization models usually ignore the temporal pattern of land use evolution. Research on how to quantify and incorporate historical information into land use optimization process is scarce, and worth exploring. This study proposed a historical-information-based patch-level multi-type ant colony optimization (HI-PMACO) model to examine the effect of incorporating historical information into the optimization model from both macro and local perspectives. Specifically, we (1) constructed five machine learning models to explore the transition potentials and used the SHAP algorithm to visualize the non-linear effects of spatial variables; (2) introduced a size-adaptive neighborhood strategy to quantify the local land use evolution preference; (3) designed an evolution preference-weighted roulette wheel mechanism to incorporate the extracted historical information into the biomimetic intelligent algorithm; and (4) verified the effect of the HI-PMACO model. Results demonstrated that the proposed HI-PMACO model exhibited better performance in improving economic benefit and transition potential objectives with smaller land use change costs, and the local details were highly in line with relevant planning policies. This study contributes to spatio-temporal land use optimization modeling in methodology and land use management considering temporal evolution patterns.

Two-Stage Ensemble Deep Learning Model for Precise Leaf Abnormality Detection in Centella asiatica

Leaf abnormalities pose a significant threat to agricultural productivity, particularly in medicinal plants such as Centella asiatica (Linn.) Urban (CAU), where they can severely impact both the yield and the quality of leaf-derived substances. In this study, we focus on the early detection of such leaf diseases in CAU, a critical intervention for minimizing crop damage and ensuring plant health. We propose a novel parallel-Variable Neighborhood Strategy Adaptive Search (parallel-VaNSAS) ensemble deep learning method specifically designed for this purpose. Our approach is distinguished by a two-stage ensemble model, which combines the strengths of advanced image segmentation and Convolutional Neural Networks (CNNs) to detect leaf diseases with high accuracy and efficiency. In the first stage, we employ U-net, Mask-R-CNN, and DeepNetV3++ for the precise image segmentation of leaf abnormalities. This step is crucial for accurately identifying diseased regions, thereby facilitating a focused and effective analysis in the subsequent stage. The second stage utilizes ShuffleNetV2, SqueezeNetV2, and MobileNetV3, which are robust CNN architectures, to classify the segmented images into different categories of leaf diseases. This two-stage methodology significantly improves the quality of disease detection over traditional methods. By employing a combination of ensemble segmentation and diverse CNN models, we achieve a comprehensive and nuanced analysis of leaf diseases. Our model’s efficacy is further enhanced through the integration of four decision fusion strategies: unweighted average (UWA), differential evolution (DE), particle swarm optimization (PSO), and Variable Neighborhood Strategy Adaptive Search (VaNSAS). Through extensive evaluations of the ABL-1 and ABL-2 datasets, which include a total of 14,860 images encompassing eight types of leaf abnormalities, our model demonstrates its superiority. The ensemble segmentation method outperforms single-method approaches by 7.34%, and our heterogeneous ensemble model excels by 8.43% and 14.59% compared to the homogeneous ensemble and single models, respectively. Additionally, image augmentation contributes to a 5.37% improvement in model performance, and the VaNSAS strategy enhances solution quality significantly over other decision fusion methods. Overall, our novel parallel-VaNSAS ensemble deep learning method represents a significant advancement in the detection of leaf diseases in CAU, promising a more effective approach to maintaining crop health and productivity.

Optimization of terminal area arrival flight sorting based on an improved sparrow search algorithm.

At present, airspace congestion and flight delays have become widespread concerns. This study aims to optimize the sequencing of arrival flights in the terminal area of multirunway airports. Considering the constraints of multiple runways, slant intervals and moving flight positions, this article establishes an optimization model for arrival flight sequencing in a multirunway airport terminal area. Accordingly, an improved sparrow search algorithm (ISSA) is proposed based on Chebyshev chaotic mapping, the golden sine strategy, and the variable neighborhood strategy. Through six basic test functions, the ISSA is compared with particle swarm optimization, the whale optimization algorithm, the genetic algorithm, and other algorithms to verify its superiority. Finally, two sets of instance data from Kunming Changshui Airport were used for experiments. The results show that the total delay times (TDTs) of small-scale flights (number of aircraft: 29) and large-scale flights (number of aircraft: 147) are 55.3% and 20.5% lower, respectively, than those of the first-come-first-served algorithm. The superiority of the ISSA designed in this article is verified, and it can significantly reduce the TDTs of arrival flights. It is suitable for optimizing arrival flights during peak hours at most airports. This approach provides theoretical support for optimizing the sorting of flights in terminal areas.

Decomposition-based multiobjective evolutionary algorithm with density estimation-based dynamical neighborhood strategy

Decomposition-based multiobjective evolutionary algorithm with density estimation-based dynamical neighborhood strategy

A novel MILP model and an improved genetic algorithm for disassembly line balancing and sequence planning with partial destructive mode

Collaborative optimization of disassembly line balancing and disassembly sequence planning can effectively reduce ineffective operation time. Considering the uncertainty of the connection relationship between parts of end-of-life products, a partial destructive disassembly mode based on the disassembly feasibility is introduced. In this study, a mixed integer linear programming (MILP) model of disassembly line balancing and sequence planning (DLBSP) is constructed. The aim is to minimize the number of stations, smoothness index, and energy consumption, and maximize the disassembly profit simultaneously. To obtain high-quality disassembly schemes, a multi-objective improved genetic algorithm is proposed. An encoding and decoding strategy based on problem characteristics is designed to improve the quality of the initial solutions. The feasibility of crossover and mutation operations is ensured by introducing precedence constraints, and a single-point insertion neighborhood strategy is designed to improve the local search ability of the algorithm. Then, the effectiveness of the proposed model and algorithm is verified in a small-scale case, and the performance of the proposed algorithm is better than that of nine meta-heuristics through 22 test cases. Next, the proposed model and algorithm are applied to a real CRT TV disassembly case. It is shown that the new disassembly scheme can reduce two stations, improve smoothness by 99.83%, increase profits by 89.75%, and reduce energy consumption by 24.66% compared to the original scheme. Finally, a DLBSP optimization system is developed, which can reduce the application difficulty of the proposed model and algorithm.

Improving neighborhood exploration into MOEA/D framework to solve a bi‐objective routing problem

AbstractLocal search (LS) algorithms are efficient metaheuristics to solve combinatorial problems. The performance of LS highly depends on the neighborhood exploration of solutions. Many methods have been developed over the years to improve the efficiency of LS on different problems of operations research. In particular, the exploration strategy of the neighborhood and the exclusion of irrelevant neighboring solutions are design mechanisms that have to be carefully considered when tackling NP‐hard optimization problems. An MOEA/D framework including an LS‐based mutation and knowledge discovery mechanisms is the core algorithm used to solve a bi‐objective vehicle routing problem with time windows (bVRPTW) where the total traveling cost and the total waiting time of drivers have to be minimized. We enhance the classical LS exploration strategy of the neighborhood from the literature of scheduling and propose new metrics based on customer distances and waiting times to reduce the neighborhood size. We conduct a deep analysis of the parameters to give a fine tuning of the MOEA/D framework adapted to the LS variants and to the bVRPTW. Experiments show that the proposed neighborhood strategies lead to better performance on both Solomon's and Gehring and Homberger's benchmarks.

Multistate-Constrained Multiobjective Differential Evolution Algorithm With Variable Neighborhood Strategy.

Multiobjective differential evolution (DE) algorithm (MODE) has been widely used in multiobjective optimization problems. However, due to the complex feasible regions, the optimization efficiency of MODE may decrease when solving constrained multiobjective problems. It is challenging to promote the evolution of population with few feasible solutions. In this article, a multistate-constrained MODE with variable neighborhood strategy (MSCMODE-VNS) is proposed to enhance the optimization effectiveness with complex feasible regions. First, a variable neighborhood DE strategy, based on a specially designed convergence indicator, is designed to accelerate the generation of feasible solutions. Second, a multistate population updating strategy with a comprehensive solution evaluation mechanism is devised to update the population of the next generation to improve the performance of solutions. Third, the convergence analysis, based on the probability theory, is derived to verify the effectiveness of the proposed MSCMODE-VNS algorithm. Finally, experimental results indicate that MSCMODE-VNS can achieve a satisfactory performance on three benchmark test suites and two real-world-constrained multiobjective problems.

Enhanced marine predator algorithm for global optimization and engineering design problems

Marine predator algorithm adopts the policy of optimal encounter rate in biological interaction between predator and prey, inspired by the Lévy and Brownian motions commonly used in oceanic predators. However, the Marine predator algorithm faces problems of premature convergence, poor search capability, and stagnation in a local optimum. In order to tolerate these shortcomings, two different strategies are applied to Marine predator algorithm in this study. (1) Taylor-based optimal neighborhood strategy (TNS) provides a more effective update in obtaining optimal individuals, thus reducing the probability of local optimum falling and attempting to overcome premature convergence, (2) Asymmetric search space with dynamic option-based learning significantly increases the probability of the population achieving the global optimum, thus improving the exploitation ability of Marine predator algorithm. The new method introduced with these two strategies added to the Marine predator algorithm is called Marine Predator Algorithm Dynamic Opposition and Taylor neighborhood search (DOTMPA). DOTMPA’s convergence performance is compared with current metaheuristic optimization and different versions of MPA in the literature on CEC2017, CEC2019, and CEC2022 benchmark suites. The results are discussed with convergence curves and box-plot analyses. DOTMPA is tested in Cantilever beam design, Tension/compression spring design, Speed reducer, Car crashworthiness, and Industrial refrigeration system engineering problems. The results are compared with the studies in the literature. Experimental results show that the proposed algorithm has strong competitiveness in terms of convergence accuracy. In addition, the stability of DOTMPA in convergence to the global optimum with trajectory analysis is also mentioned. Finally, the proposed method is applied to three truss topology optimization problems (20-truss, 24-truss, and 72-truss). The experimental results, the applicability of the proposed algorithm in practical applications, and the benchmark problem make it a promising and competitive optimization algorithm for optimization problems.

Reliability enhancement and power loss reduction in medium voltage distribution feeders using modified jellyfish optimization

The reliability of electrical systems is a critical operational characteristic for distribution companies. On the other hand, they pay great attention on reducing losses in primary and secondary distribution lines. In this article, a modified Jellyfish Search (MJFS) algorithm is presented for handling Optimal Network reconfiguration (ONR) in Medium Voltage Distribution Feeders (MVDFs). An ONR formulation is proposed for reliability enhancement and power loss reduction in a multi-objective model where three reliability measures are incorporated which are total energy not supplied (TENS), system average interruption unavailability index (SAIUI) and system average interruption frequency index (SAIFI). The proposed MJFS expands on the JFS searching benefits by introducing a quasi-oppositional-based learning and social neighborhood strategies. Also, a proposed bus-line feeding matrix (BLFM) is incorporated to develop the MJFS. Through the BLFM, each distribution line is simulated as an outage and the regarding distribution nodes which are not supplied is estimated. The simulation results are performed on a large-scale IEEE MVFD of 137-bus which derive significant improvements of the proposed MJFS over the standard JFS algorithm. Considering the proposed multi-objective model, the proposed MJFS achieves great reduction in power losses, TENS, SAIUI and SAIFI with 44.42, 30.57, 30.78 and 23.92 %, respectively compared to the initial case. Also, versus the standard JFS, grey wolf optimizer, tuna swarm optimization, equilibrium optimizer and tunicate swarm optimization, the proposed MJFS acquires great reduction in the standard deviation with improvement of 2.5, 92.51, 10.67, 91.53 and 96.51%, respectively.

An effective fruit fly optimization algorithm for the distributed permutation flowshop scheduling problem with total flowtime

Distributed permutation flowshop scheduling problem (DPFSP) has always been a hot issue. The optimization goal of minimizing the total flowtime is of great significance to the environment of multi-factory. In this paper, a discrete fruit fly optimization algorithm (DFFO) is proposed to solve the DPFSP with the total flowtime criterion. In the proposed DFFO, an initialization method considering the population quality and diversity is adopted. In the smell-based search stage, three perturbation operators, the Shift-based operator, the Exchange-based operator and the Hybrid operator are designed respectively, and each fruit fly improves its state through a specific neighborhood strategy. In addition, we propose an improved reference local search (MRLS) method to enhance the exploitation ability of fruit flies. In the vision-based search stage, fruit flies use a well-designed combination update mechanism to lead fruit flies to more potential areas. In order to enhance the exploration ability, we use random reinforcement method for the population. The parameters are evaluated using an orthogonal experimental design to determine the appropriate values of the key parameters. In addition, we test the DFFO and the state-of-art algorithms from the literature on 720 large-scale instances. The experimental results show that DFFO is a very effective metaheuristic.

Minimizing Energy Usage and Makespan of Elevator Operation in Rush Hour Using Multi-Objective Variable Neighborhood Strategy Adaptive Search with a Mobile Application

The purpose of this study is to address two major issues: (1) the spread of epidemics such as COVID-19 due to long waiting times caused by a large number of waiting for customers, and (2) excessive energy consumption resulting from the elevator patterns used by various customers. The first issue is addressed through the development of a mobile application, while the second issue is tackled by implementing two strategies: (1) determining optimal stopping strategies for elevators based on registered passengers and (2) assigning passengers to elevators in a way that minimizes the number of floors the elevators need to stop at. The mobile application serves as an input parameter for the optimization toolbox, which employs the exact method and multi-objective variable neighborhood strategy adaptive search (M-VaNSAS) to find the optimal plan for passenger assignment and elevator scheduling. The proposed method, which adopts an even-odd floor strategy, outperforms the currently practiced procedure and leads to a 42.44% reduction in waiting time and a 29.61% reduction in energy consumption. Computational results confirmed the effectiveness of the proposed approach.

Community-based mental health prevention strategy with older adults

The purpose of this study was to characterize the components required for the implementation of a community-based intervention strategy for mental health prevention in older adults in a poor neighborhood in the city of Temuco. The sample of the qualitative study was composed of 15 older adults, 2 neighborhood leaders, and 2 psychosocial interveners who have been working in the sector. The design corresponds to a case study, and semi-structured interviews guided by a guideline of topics and participant observation in activities carried out with older adults in a mental health and self-care workshop were used in the production of data. The data were interpreted through content analysis. The results show that the form of aging is characterized by loneliness and feelings of sadness, including in some cases tensions derived from family dynamics. There is also economic precariousness and difficulty in accessing health services. During the pandemic, they had to face various situations that increased their feelings of loneliness and helplessness. On the other hand, the results show that neighborhood strategies to support the elderly operate on a contingency basis, and are of an assistance type. On the other hand, there is little presence of services and interventions by professional teams of public policy programs focused on this population group. Among the priority needs of the people interviewed are social contact, breaking the routine, participating in collective recreational activities aimed at sharing with others, and developing workshops with external professionals that allow them to learn. In relation to the strategies that can contribute to a community-based intervention, the need for workshop-type group activities is identified, with a horizontal and close treatment of the professionals who dynamize them. These should be held regularly (weekly), with a personalized call, including the use of participatory and playful strategies and the possibility of developing handicrafts. The workshops are understood by the elderly as a space for socialization, mutual support and learning. According to the findings of the research, these workshops should be complemented with interventions and resources that promote healthy lifestyles and resources to adequately meet the needs of the elderly.

SaCHBA_PDN: Modified honey badger algorithm with multi-strategy for UAV path planning

The honey badger algorithm (HBA) is a meta-heuristic optimization algorithm that simulates the foraging behavior of honey badgers. Since the algorithm is prone to premature convergence when solving complex optimization problems. To improve the overall optimization performance of the basic HBA, this paper develops a modified HBA named SaCHBA_PDN based on the Bernoulli shift map, piecewise optimal decreasing neighborhood, and horizontal crossing with strategy adaptation and applies it to solve the unmanned aerial vehicle (UAV) path planning problem. Firstly, the Bernoulli shift map is invoked to the HBA algorithm to change its initialization process, thus increasing the diversity of the population and speeding up the convergence speed. Secondly, a new piecewise optimal decreasing neighborhood strategy (PODNS) is proposed to address the shortcomings of unbalanced convergence of the traditional optimal neighborhood strategy. The proposed PODNS increases the optimization efficiency of HBA and enhances the local search ability to avoid falling into the local optimum. Finally, a novel horizontal crossing with strategy adaptation is introduced to balance exploration and exploitation and enhance the global optimization ability. These strategies collaborate to enhance HBA in accelerating overall performance. The superiority of SaCHBA_PDN is comprehensively verified by comparing it with the original HBA and numerous celebrated and newly developed algorithms on the well-known 23 classical benchmark functions and IEEE CEC2017 test suite, respectively. Experimental results show that SaCHBA_PDN has a better performance than other optimization algorithms. Furthermore, SaCHBA_PDN is used to solve a UAV path planning problem based on the threat source model and applied to circular and irregular obstacle scenarios as well as two-dimensional grid maps. Simulation results show that SaCHBA_PDN can obtain more feasible and efficient paths in different obstacle environments.

A hybridization of PSO and VNS to solve the machinery allocation and scheduling problem under a machinery sharing arrangement

This paper presents the hybrid Particle Swarm Optimization and the Variable Neighborhood Search (PSO-VNS) to solve the machinery and equipment allocation and scheduling to help the growers expand the production level to meet the increased demands and growing interest, and to increase profitability. This problem can be formulated as the Machinery and Equipment Allocation and Scheduling Problem (MEASP) which is special type of a hybrid flow shop scheduling problem (HFS) with sequence dependent setup times, machine eligibility, machine grouping, blocking, tooling constraint, and time windows (HFS |SDST, rcrc, Grouping, blocking, Tool, Tw | Cmax). The objective of this research is to minimize the total completion time for sugarcane cultivation. A Mixed Integer Linear Programming (MILP) model was developed to handle small-scale problems. The hybrid Particle Swarm Optimization and the Variable Neighborhood Search (PSO-VNS) was used for large-scale problems. Two new velocity update formulae and a position update formula were incorporated into the Particle Swarm Optimization (PSO), and four types of neighborhood strategies were developed for the VNS. The experimental results show that all the PSO-VNS methods outperform the traditional PSO due to the effectiveness of the newly proposed velocity and position update formulae in finding the optimal solutions, the PSO-VNS-6 methods, on the average, can reduce the computational time by 58.43% from the original PSO and improve the solution quality by 11.71%.

Community Agricultural Reservoir Construction and Water Supply Network Design in Ubon Ratchathani, Thailand, Using Adjusted Variable Neighborhood Strategy Adaptive Search

Agricultural sectors all over the world are facing water deficiencies as a result of various factors. Countries in the Greater Mekong Subregion (GMS) in particular depend on the production of agricultural products; thus, drought has become a critical problem in such countries. The average water level in the lower part of the Mekong River has been decreasing dramatically, resulting in the wider agricultural area of the Mekong watershed facing a lack of water for production. The construction of community reservoirs and associated water supply networks represents a strategy that can be used to address drought problems in the GMS. This study aims to solve the agricultural community reservoir establishment and water supply network design (CR–WSND) problem in Khong Chiam, Ubon Ratchathani, Thailand—a city located in the Mekong Basin. The CR–WSND model is formulated using mixed-integer programming (MIP) in order to minimize the cost of reservoir construction and water irrigation. An adjusted variable neighborhood strategy adaptive search (A-VaNSAS) is applied to a real-world scenario involving 218 nodes, and its performance is compared with that of the original variable neighborhood strategy adaptive search (VaNSAS), differential evolution (DE), and genetic algorithm (GA) approaches. An improved box selection formula and newly designed improvement black boxes are added to enhance the quality beyond the original VaNSAS. The results reveal that the quality of the solution from A-VaNSAS is significantly better than those of GA, DE, and VaNSAS (by 6.27%, 9.70%, and 9.65%, respectively); thus, A-VaNSAS can be used to design a community reservoir and water supply network effectively.