Neighborhood Search Research Articles

Variable Neighborhood Search for Multi-Port Berth Allocation with Vessel Speed Optimization

This paper delves into the multi-port berth allocation problem (MBAP), enriching the traditional berth allocation problem (BAP) with vessel speed optimization (VSO). In the MBAP, it is assumed that there is cooperation between the port and the shipping companies, and the operation of the vessels and the ports is planned to maximize the overall benefits. Exploring this potential collaboration between ports and shipping entities has the potential to mitigate, or even resolve, the challenges plaguing maritime transportation, e.g., port congestion and suboptimal vessel schedules, to ultimately enhance the efficiency of maritime trade. In this paper, a new mixed-integer linear programming (MILP) model for the MBAP is formulated, which attempts to minimize the total cost incurred during operations, with various constraints such as vessel sailing, the vessel space–time relationship in ports, and the planning period. Meanwhile, an innovative variable neighborhood search (VNS) algorithm is presented, in which the initial solution generation method and neighborhood structures are proposed according to the MBAP characteristics. Furthermore, two sets of MBAP instances are generated to test the proposed MILP and VNS, of which the first set is based on real-world port data and the second on existing studies. The numerical experiments verify that the VNS can efficiently and reliably solve instances of all scales, with each neighborhood structure contributing uniquely to the iterative process. In addition, by analyzing the impact of varying oil prices on the MBAP, the study offers valuable management insights. Finally, a case study based on real data from a port group in the Yangtze River Basin is presented to further demonstrate the necessity of considering vessel service time window and planning period in the MBAP as well as the important role of the VSO in scheduling.

Re-Entrant Green Scheduling Problem of Bearing Production Shops Considering Job Reworking

To solve various reworking and repair problems caused by unqualified bearing product quality inspections, this paper introduces a green re-entrant scheduling optimization method for bearing production shops considering job reworking. By taking into account quality inspection constraints, this paper establishes an integrated scheduling mathematical model based on the entire processing–transportation–assembly process of bearing production shops with the goals for minimizing the makespan, total carbon emissions, and waste emissions. To solve these problems, the concepts of the set of the longest common machine routes (SLCMR) and the set of the shortest recombination machine combinations (SSRMC) were used to propose the re-entrant scheduling optimization method, based on system reconfiguration, to enhance the system stability and production scheduling efficiency. Then, a multi-objective hybrid optimization algorithm, based on a neighborhood local search (MOOA-LS), is proposed to improve the search scope and optimization ability by constructing a multi-level neighborhood search structure. Finally, this paper takes a bearing production shop as an example to carry out the case study and designs a series of experimental analyses and comparative tests. The final results show that in the bearing production process, the proposed model and algorithm can effectively realize green and energy-saving re-entrant manufacturing scheduling.

A three-phase algorithm for the pollution traveling Salesman problem

This paper studies a variant of the Pollution Traveling Salesman Problem (PTSP) focused on fuel consumption and pollution emissions (PTSPC). The PTSPC generalizes the well-known Traveling Salesman Problem (TSP), classified as NP-Hard. In the PTSPC, a vehicle must deliver a load to each customer through a Hamiltonian cycle, minimizing an objective function that considers the speed of each edge, the mass of the truck, the mass of the load pending delivery, and the distance traveled. We have proposed a three-phase algorithm for the PTSPC. The first phase solves the Traveling Salesman Problem (TSP) exactly with a time limit and heuristically using a Nearest Neighborhood Search approach. This phase considers the constraints associated with the PTSPC by using commercial software. In the second phase, both the obtained solutions and their inverse sequences from the initial phase undergo enhancement utilizing metaheuristic algorithms tailored for the PTSPC. These algorithms include Variable Neighborhood Search (VNS), Tabu Search (TS), and Simulated Annealing (SA). Subsequently, for the third phase, the best solution identified in the second phase—determined by having the minimum value by the PTSPC objective function—is subjected to resolution by a mathematical model designed for the PTSPC, considering the heuristic emphasis of commercial software. The efficiency of the former algorithm has been validated through experimentation involving the adaptation of instances from the Pollution Routing Problem (PRP) to the PTSPC. This approach demonstrates the capacity to yield high-quality solutions within acceptable computing times.

A hybrid genetic programming algorithm for the distributed assembly scheduling problems with transportation and sequence-dependent setup times

This paper investigates a distributed assembly permutation flow-shop scheduling problem with transportation and sequence-dependent set-up times (DAPFSP-TSDST). A hybrid genetic programming (HGP) algorithm is proposed to optimize the makespan of the assembly stage, which inherits the merits of genetic programming (GP) and neighbourhood search operators. In HGP, a hybrid problem-specific initialization heuristic is developed to make populations more diverse. Multiple neighbourhood search operators are employed as the leaf nodes, which are vital for the success of GP. A product shift strategy is proposed to strengthen its exploitability. In addition, a simulated annealing criterion is adopted to make the HGP explore more thoroughly. Finally, statistical and computational experiments are carried out on the benchmark instances. The results exhaustively identify the notable competitiveness of the HGP algorithm in coping with the DAPFSP-TSDST.

Hybrid Artificial Bee Colony Algorithm with Variable Neighborhood Search for Capacitated Vehicle Routing Problem

Aiming at the capacitated vehicle routing problem, a hybrid integer programming model with goal of lowest path cost is constructed, and a hybrid artificial bee colony algorithm with variable neighborhood search based on the model and the characteristics of the CVRP problem is proposed to solve the problem. The hybrid algorithm integrates the artificial bee colony algorithm and the variable neighborhood search algorithm, embeds a multi-variable neighborhood operator in the local search link of the artificial bee colony to carry out iteration. And the operator contains targeted transformation operations on path nodes, strings, and sub paths to ensure the diversity of the bee population. In addition, a variable neighborhood perturbation strategy is used to strengthen the algorithm's ability to escape from local optima. The comparative analysis of the literature study set and its algorithm solution shows that the designed hybrid variable neighborhood artificial bee colony algorithm has strong global search ability and high solution accuracy, especially in solution stability. HABC-VNS can obtain 47 optimal solutions in 74 examples. The average minimum deviation of the optimal solution is 0.34%, and the average deviation of the average CVRP set is 0.57%. The overall performance is better than the algorithm in the comparative literature.

Solving multi-objective green flexible job shop scheduling problem by an improved chimp optimization algorithm

As environmental contamination becomes more and more severe, enterprises need to consider optimizing environmental criteria while optimizing production criteria. In this study, a multi-objective green flexible job shop scheduling problem (MO_GFJSP) is established with two objective functions: the makespan and the carbon emission. To effectively solve the MO_GFJSP, an improved chimp optimization algorithm (IChOA) is designed. The proposed IChOA has four main innovative aspects: 1) the fast non-dominated sorting (FDS) method is introduced to compare the individuals with multiple objectives and strengthen the solution accuracy. 2) a dynamic convergence factor (DCF) is introduced to strengthen the capabilities of exploration and exploitation. 3) the position weight (PW) is used in the individual position updating to enhance the search efficiency. 4) the variable neighborhood search (VNS) is developed to strengthen the capacity to escape the local optimum. By executing abundant experiments using 20 benchmark instances, it was demonstrated that the developed IChOA is efficient to solve the MO_GFJSP and effective for reducing carbon emission in the flexible job shop.

A parallel deep adaptive large neighbourhood search algorithm for distributed heterogeneous hybrid flow shops with mixed-model assembly scheduling

ABSTRACT Nowadays, manufacturing enterprises must have fast response and flexible production capabilities to meet personalized and diversified market demands. Mixed-model production and distributed production have become the preferred production methods for enterprises. This article studies a distributed heterogeneous hybrid flow shop scheduling problem with a mixed-model assembly line (DHHFSP-MMAL), which consists of manufacturing and assembly stages. The DHHFSP-MMAL is modelled by a mixed integer linear programming (MILP) model. Three constructive heuristics and a parallel deep adaptive large neighbourhood search (PDALNS) problem are presented. A constructive heuristic with a group strategy is employed to obtain an initial solution. Several deep destroy-and-repair operators are proposed where problem-specific greedy local search methods are applied. The PDALNS assigns weights to destroy-and-repair operators to guide the selection of operators. The parallel computing technique is introduced to increase the efficiency of training. The experiments demonstrate that the PDALNS algorithm is an efficient and effective algorithm for solving the DHHFSP-MMAL problem.

A Bibliometric Analysis of a Genetic Algorithm for Supply Chain Agility

As a famous population-based metaheuristic algorithm, a genetic algorithm can be used to overcome optimization complexities. A genetic algorithm adopts probabilistic transition rules and is suitable for parallelism, which makes this algorithm attractive in many areas, including the logistics and supply chain sector. To obtain a comprehensive understanding of the development in this area, this paper presents a bibliometric analysis on the application of a genetic algorithm in logistics and supply chains using data from 1991 to 2024 from the Web of Science database. The authors found a growing trend in the number of publications and citations over the years. This paper serves as an important reference to researchers by highlighting important research areas, such as multi-objective optimization, metaheuristics, sustainability issues in logistics, and machine learning integration. This bibliometric analysis also underlines the importance of Non-Dominated Sorting Genetic Algorithm II (NSGA-II), sustainability, machine learning, and variable neighborhood search in the application of a genetic algorithm in logistics and supply chains in the near future. The integration of a genetic algorithm with machine learning is also a potential research gap to be filled to overcome the limitations of genetic algorithms, such as the long computational time, difficulties in obtaining optimal solutions, and convergence issues for application in logistics and supply chains.

GOGCN: an attentional network with geometry and orientation-awareness for airborne LiDAR point cloud classification

The airborne LiDAR point cloud has its own characteristics, however, the classification method always fails to capture these characteristics. In this paper, a classification method named GOGCN was designed that adopts a U-Net network structure and uses a directionally constrained nearest neighbourhood search during down-sampling to generate the directionally aware feature. The point cloud geometric structure is obtained through geometry-aware information extraction, and then a graph attention convolution is utilised to learn the most representative features. A comparative experiment on GML(B) dataset and one engineering dataset demonstrated that GOGCN network have well performance and can be widely used in classification.

The static ridesharing routing problem with flexible locations: A Norwegian case study

The municipalities in the Bergen region in Norway have recently announced a pilot project for ridesharing in the region as a means to reduce traffic congestion. As part of this project, we study the Static Ridesharing Routing Problem with Flexible Locations (SRRPFL), which aims at determining efficient routes and schedules for a set of drivers to pick up and deliver passengers at different, flexible pickup and delivery locations. We present a bi-objective mixed integer programming (MIP) model for the SRRPFL where we (lexicographically) first maximize the number of passengers serviced and then minimize the total travel times. To solve real-life instances of the SRRPFL, we propose a new Adaptive Large Neighborhood Search (ALNS) heuristic. To further improve its performance, we extend the ALNS heuristic with a local search, as well as with a set partitioning problem (denoted the Route Combination Problem) that optimally recombines the routes previously encountered in the search. The ALNS heuristic is tested on a number of test instances based on real trip data and the results demonstrate its effectiveness. The results also provide a number of insights regarding the potential benefits of ridesharing in our case study.

The pollution traveling salesman problem with refueling

This study presents the Pollution Traveling Salesman Problem with Refueling, a novel optimization problem which integrates two recently proposed variants of the Traveling Salesman Problem: the Pollution Traveling Salesman Problem and the Traveling Salesman Problem with Refueling. The proposed problem captures the operational dynamics of a real-world routing scenario involving a single vehicle originating from a central depot and delivering products to end customers. When considering the vehicle’s fuel tank capacity and fuel consumption during the routing process, the need to visit fuel stations for refueling arises. To address this complex problem, a new mixed integer linear programming model was developed, and the Gurobi solver was employed to solve smaller instances. For the effective resolution of larger practical problem cases, a two-stage double adaptive general variable neighborhood search method was proposed. The proposed methodology exhibits comparable efficiency to a commercial solver, demonstrating notably low execution time requirements. To further assess its performance, a comparative study was conducted on TSPLib instances. In comparison to various solution approaches documented in the open literature, encompassing both Variable Neighborhood Search-based and alternative methods, our proposed approach consistently yields highly competitive results within low execution times.

Rank-based Hashing for Effective and Efficient Nearest Neighbor Search for Image Retrieval

The large and growing amount of digital data creates a pressing need for approaches capable of indexing and retrieving multimedia content. A traditional and fundamental challenge consists of effectively and efficiently performing nearest-neighbor searches. After decades of research, several different methods are available, including trees, hashing, and graph-based approaches. Most of the current methods exploit learning to hash approaches based on deep learning. In spite of effective results and compact codes obtained, such methods often require a significant amount of labeled data for training. Unsupervised approaches also rely on expensive training procedures usually based on a huge amount of data. In this work, we propose an unsupervised data-independent approach for nearest neighbor searches, which can be used with different features, including deep features trained by transfer learning. The method uses a rank-based formulation and exploits a hashing approach for efficient ranked list computation at query time. A comprehensive experimental evaluation was conducted on 7 public datasets, considering deep features based on CNNs and Transformers. Both effectiveness and efficiency aspects were evaluated. The proposed approach achieves remarkable results in comparison to traditional and state-of-the-art methods. Hence, it is an attractive and innovative solution, especially when costly training procedures need to be avoided.

A hybrid multi-objective solution approach for a reliable truck-drone routing problem integrated with pickup and delivery services

ABSTRACT Humanitarian logistics aims to reduce the arrival time of relief goods and optimize services in disaster-affected regions. This paper introduces a multi-objective hybrid truck-drone routing problem with pickup and delivery services, considering reliability (MHTDRP-PD-R) to enhance relief operations within humanitarian logistics. The model is formulated as a mixed-integer linear programming problem with the simultaneous goals of minimizing transportation time and maximizing reliability through routing and inventory decisions made by multiple trucks and drones. An adaptive large neighborhood search (ALNS) algorithm, coupled with heuristic technique, is developed to tackle the problem. The ALNS algorithm refines initial solutions through destroy and repair operators, integrated with multi-objective optimization methods to create a unified objective function. Small and large-scale test problems are then employed to evaluate solution quality using multi-objective optimization. The calculation results show the superiority of the developed ALNS integrated with the ε-constraint method regarding the objective value and solution time.

Multi-agent deep Q-network-based metaheuristic algorithm for Nurse Rostering Problem

The Nurse Rostering Problem (NRP) aims to create an efficient and fair work schedule that balances both the needs of employees and the requirements of hospital operations. Traditional local search-based metaheuristic algorithms, such as adaptive neighborhood search (ANS) and variable neighborhood descent (VND), mainly focus on optimizing the current solution without considering potential long-term consequences, which may easily get stuck in local optima and limit the overall performance. Thus, we propose a multi-agent deep Q-network-based metaheuristic algorithm (MDQN-MA) for NRP to harness the strengths of various metaheuristics. Each agent encapsulates a metaheuristic algorithm, where its available actions represent different perspectives of the problem environment. By combining their strengths and various perspectives, these agents can work collaboratively to navigate and search for a broader range of potential solutions effectively. Furthermore, to improve the performance of an individual agent, we model its neighborhood search as a Markov Decision Process model and integrate a deep Q-network to consider long-term impacts for its neighborhood sequential decision-making. The experimental results clearly show that an individual agent in MDQN-MA can outperform ANS and VND, and multiple agents in MDQN-MA even perform better, achieving the best results among metaheuristic algorithms on the Second International Nurse Rostering Competition dataset.

Iterated variable neighborhood search for integrated scheduling of additive manufacturing and multi-trip vehicle routing problem

This study addresses a novel problem structure for integrated scheduling of additive manufacturing and delivery processes. In the manufacturing process, we consider batch production of parts, simulating Selective Laser Melting technology in parallel additive manufacturing machines with heterogeneous specifications. A batch production, referred to as a build, must consist of parts with homogenous material. During the production run, sequence-dependent material setup times occur in between builds. In the delivery process, parts are delivered in batches using a limited number of homogenous capacitated vehicles. Each vehicle might perform multi-trips. Thus, the route for each trip and departure schedule among trips must be decided. In this regard, batch delivery is independent to batch production. The problem objective function is to minimize the total completion time of all parts. For optimally solving the problem, a mixed-integer linear programming model is devised. To solve larger problems sizes under a lower computation time, an iterated variable neighborhood search algorithm with embedded rule-based heuristics is proposed. The computational experiment denotes that the proposed algorithm with route minimization rule effectively solves small-scale problems, while capacity usage maximization rule performs the best in large-scale problems and outperforms other existing metaheuristics. Further result analysis is also performed to provide managerial insights.

Adaptive Gradient-based Word Saliency for adversarial text attacks

Natural Language Processing (NLP) models are known vulnerable to adversarial text attacks. Various word-level attacks have been proposed to modify input words by a pre-calculated word saliency order or the heuristic optimization algorithm. However, the predefined fixed order usually leads to a low attack success ratio, and the heuristic algorithm is often inefficient due to many model queries. In this paper, an Adaptive Gradient-based Word Saliency (AGWS) is proposed to improve the query-efficiency and simultaneously preserve a high attack success ratio for the word-level adversarial text attack. Firstly, the AGWS calculates the word saliency with a one-time gradient to all words instead of iterative queries, so the query-efficiency is enhanced. Secondly, the AGWS adaptively updates the word saliency via the variable neighborhood search algorithm to avoid the fixed modification order, which is significant in improving the attack success ratio. Additionally, the AGWS collect substitutes with a hybrid sememe and synonym strategy to enlarge adversary options. Extensive experiments and ablation studies manifest that the AGWS achieves the highest or second highest attack success ratio with lowest word perturbation percentage and improves the query-efficiency compared with baselines. Besides, the AGWS also shows superiorities in adversarial training and transferability.

Parallel batch processing machines scheduling for SLM additive manufacturing in cloud manufacturing

For the scheduling problem of parallel batch processing machines for Selective Laser Melting (SLM) additive manufacturing in cloud manufacturing environments, an Improved Discrete Artificial Bee Colony algorithm(IABC) is proposed by considering the service time of the product as well as introducing the mathematical model of the problem. Firstly, a population initialization strategy is introduced to improve the quality of initial solutions and accelerate the convergence of the population.Secondly, through three neighborhood search operators, the transition from the original nectar source to the new nectar source is achieved for employed bees and following bees. Furthermore, a local search strategy and a modified abandonment pattern are proposed to increase population diversity and prevent the algorithm from falling into local optima.Finally, the performance of the algorithm is analyzed, and the IABC algorithm was applied to 36 examples. The effectiveness and advantages of IABC in solving the studied problem are verified through a large number of simulation experiments.

Intelligent design of sensor networks for data-driven sensor maintenance at railways

With rapid advances in digitization, many critical processes in transportation, industries, and our daily life rely on sensor measurements. With time, however, the measurements may get gradually biased and their precision deteriorates, leading to an enhanced risk of major disruptions caused by false sensor measurements. All single sensor measurements are uncertain and deviate from the true value. To detect malfunctioning sensors early on, a set of recent measurements of each sensor has to be constantly cross-checked against the measurements of a given number of other sensors, i.e., sensors should form a diagnosable network.In this article, we examine the intelligent positioning of safety-relevant sensors at railways such that the installed sensors can constantly cross-check each other and the number of the required sensors is minimized. The arising sensor positioning problem (SPP) belongs to the family of the coordinated set covering problems with two binary matrices: the choice of columns in one matrix implies the selection of specific columns and rows in the other matrix. We formulate an integer program, provide some formal analysis of the SPP and design a customized large neighborhood search metaheuristic RuM, which finds close-to-optimality solutions fast. In our computational experiments, we show that if we ignore the diagnosability requirement, the installed sensors cannot sufficiently cross-check each other in most cases. However, it costs only a few (or even no) additional sensors to ensure the diagnosability of the sensor network.

A multi-period vaccines supply chain network design with capacity expansion and different replenishment cycles under uncertain demand

The World Health Organization (WHO) proposes guidelines for vaccine distribution networks that incorporate economic order quantity and risk pooling principles, specifically through implementing different replenishment cycles between different levels of the distribution network. The substantial impact of different replenishment cycles on inventory levels is demonstrated in the existing literature. However, current optimization network design research has paid limited attention to this aspect. This article aims to address this research gap by investigating a two-stage stochastic programming for a multi-period perishable vaccine network design problem, taking into account the different replenishment cycles characteristic and capacity expansion under uncertain demand conditions. Given the inherent difficulty of solving the proposed problem, a matheuristic approach based on Variable Neighborhood Search (VNS) is developed. To illustrate the practical application and analyze the impact of uncertain demand, we apply the proposed model and its solution method using the Indonesia dataset in the COVID-19 situation. The computational results indicate that lower replenishment cycles leads to higher inventory levels, total costs, and capacity requirements for the top-level distribution center. Optimum replenishment cycles and further managerial implications and future potential research are also presented.

Heterogeneous vehicle scheduling with precedence constraints

The problem of heterogeneous vehicle scheduling with precedence constraints is inspired by the transportation service in tourism areas. Tourists must take the shuttle vehicles provided by the areas because of the long distances between the scenic spots. The scheduling of vehicles in tourism areas is complicated because the transportation requests of tourists are precedence-constrained temporally and spatially. The problem optimises both the cost of using vehicles and the waiting time of tourists. A mixed-integer linear programming model is formulated according to the description of a graph. An adaptive large neighbourhood search algorithm with several specialised operators is designed to solve the problem. Experiments based on randomly generated instances validate the mathematical model and the algorithm. A real-size instance based on Qiandao Lake in China is also analysed. The results indicate that the algorithm outperforms the model. The sensitivities of key parameters are analysed with managerial insights presented.