Neighborhood Search Research Articles

Effective metaheuristic and rescheduling strategies for the multi-AGV scheduling problem with sudden failure

This paper investigates a novel scheduling problem, the multi-AGV scheduling problem with sudden failure (MASP-SF), which is of great importance in modern intelligent manufacturing systems. The problem considers how to assign tasks to multiple AGVs with the objective of minimizing the total cost under two events of normal production and sudden failure in the workshop. In the event of normal production, a population-based variable neighborhood search (PVNS) algorithm with effective strategies such as rule-based stochastic method in the initialization stage, pancake flipping strategy in the shaking stage, improved neighborhood structures in the variable neighborhood descent stage is presented to solve the multi-AGV scheduling problem. In the event of sudden failure, a mixed-integer linear programming model and two rescheduling strategies are proposed to solve MASP-SF. The two rescheduling strategies are the rapid repair rescheduling strategy and a rescheduling strategy based on an improved NNH. To validate the above ideas, we conduct comprehensive and in-depth experiments on a practical plant and statistical analysis of the resulting data. The results show that the scheduling solutions generated by the rescheduling strategies are feasible, and the proposed PVNS algorithm has a superior performance compared to existing algorithms in multi-AGV scheduling problem.

Bus Rescheduling for Long-Term Benefits: An Integrated Model Focusing on Service Capability and Regularity

Unplanned disruptions, such as vehicle breakdowns, in a public transportation system can lead to severe delays and even service interruptions, preventing the successful implementation of subsequent plans and the overall stability of transit services. A common solution to address such issues is implementing a bus bridging service using an experience-based response strategy, involving the deployment of spare buses to continue affected services. However, with this approach, it becomes impractical and challenging to generate a feasible and rational rescheduling scheme for the remaining transit services when spare buses are insufficient or widespread disruptions occur. In response to this challenge, we propose an innovative model that integrates service capability and regularity, aiming to minimize rescheduling costs through timetable adjustments and scheduling reassignments. We apply dynamic programming to comprehensively consider the hysteresis effects of disruptions and achieve a long-term optimal rescheduling scheme. To efficiently solve the proposed model, the large neighborhood search algorithm is improved by incorporating operational rules. Finally, several experiments are conducted under an actual transit operation scenario in Shenzhen. The results demonstrate that our method significantly reduces trip cancellations and, simultaneously, diminishes the increase in the departure interval resulting from the adjusted schedule by 23.27%.

Transactive energy trading in distribution systems via privacy-preserving distributed coordination

This paper presents an approach to the transactive energy (TE) market that aims to improve the network efficiency and fairness of electricity market in distribution systems. The TE market is designed to allow subscribers to trade electricity in real-time based-on their energy consumption and generation forecast. The proposed framework utilizes distribution locational marginal pricing (DLMP), which accurately reflects the costs of electricity transmission and distribution, to fairly allocate these costs among subscribers. The Small Neighborhood Search Optimal Points (SNSOP) algorithm is then introduced to solve the challenge of finding optimal points for all participants in the TE market, as the maximum financial gain for one participant comes at the maximum financial loss of another participant. The results of the case study show that the proposed framework leads to a more efficient and reliable electricity network, by enabling decentralized control and reducing losses through optimal trades in the TE market.

An efficient hybrid adaptive large neighborhood search method for the capacitated team orienteering problem

This paper focuses on a NP-Hard vehicle routing problem with profits, and presents a hybrid heuristic solution approach for the Capacitated Team Orienteering Problem (CTOP), which is a variant of the well-known team orienteering problem. In the CTOP, given a set of homogeneous fleet and a set of clients to each is associated a deterministic profit and volume demand, one maximizes the collected profits for serving customers by operating a set of routes while respecting vehicles’ limited time and volume capacities. Our proposed method to solve the CTOP is a Hybrid Adaptive Large Neighborhood Search (HALNS) heuristic. Several removal and insertions operators with different node selection strategies and local search procedures are developed to generate solutions in the first step. In the second step, solutions’ routes are passed to a set packing problem which is solved using an exact branch-and-cut procedure in order to obtain the best solution. Compared to all the state-of-the-art methods, extensive computational experiments show that our method outperforms in terms of solution quality and/or computational times when tested on the small and large-scale benchmark CTOP instances respectively proposed by Archetti et al. (2009) and Tarantilis et al. (2023). Furthermore, our hybrid method improves 10 of the best known solutions for large-scale benchmark instances.

A practical optimization framework for political redistricting: A case study in Arizona

Following the decennial census, each state in the U.S. redraws its congressional and state legislative district boundaries, which must satisfy various legal criteria. For example, Arizona’s Constitution describes six legal criteria, including contiguity, population balance, competitiveness, compactness, and the preservation of communities of interest, political subdivisions and majority–minority districts, each of which is to be enforced “to the extent practicable”. Optimization algorithms are well suited to draw district maps, although existing models and methods have limitations that inhibit their ability to draw legally-valid maps. Adapting existing optimization methods presents two major challenges: the complexity of modeling to achieve multiple and subjective criteria, and the computational intractability when dealing with large redistricting input graphs. In this paper, we present a multi-stage optimization framework tailored to redistricting in Arizona. This framework combines key features from existing methods, such as a multilevel algorithm that reduces graph input sizes and a larger local search neighborhood that encourages faster exploration of the solution space. This framework heuristically optimizes geographical compactness and political competitiveness while ensuring that other criteria in Arizona’s Constitution are satisfied relative to existing norms. Compared to Arizona’s enacted map (CD118) to be used until 2032, the most compact map produced by the algorithm is 41% more compact, and the most competitive map has five more competitive districts. To enable accessibility and to promote future research, we have created Optimap, a publicly accessible tool to interact with a part of this framework. Beyond the creation of these maps, this case study demonstrates the positive impact of adapting optimization-based methodologies for political redistricting in practice.

A parallel variable neighborhood search for [formula omitted]-neighbor facility location problems

In this paper, we employ the less is more approach to develop a Parallel Variable Neighborhood Search (VNS) algorithm for the α-neighbor p-center problem (αNpCP) and the α-neighbor p-median problem (αNpMP). The αNpCP and the αNpMP are generalizations of the p-center (pCP) and p-median (pMP) problems, respectively. In the α-neighbor problems, one seeks to open p facilities and assign each of the n customers to their closest α ones. The objective is to minimize the maximum distance of a customer to its αth facility, in the case of the αNpCP, and the sum of the distances from each customer to their α nearest facilities, in the case of the αNpMP. Our VNS adapts simple but efficient algorithms and data structures from the pCP and pMP literature to the αNpCP and αNpMP context. We also introduce an updated objective function for the αNpCP, which adds more information to the solution cost and helps the VNS to escape from local optima. Several experimental tests show that our VNS outperforms more complex state-of-the-art algorithms. Regarding the αNpCP, on 120 instances derived from the OR-library set, our algorithm improved best-known solutions for 22, with an average improvement of 34.26%; the overall gap on the 120 instances is 6.18% in favor of our algorithm. Moreover, on 231 instances derived from the TSPLIB set, we improved the solutions for 115, with an average improvement of 5.30%, and an overall improvement gap of 2.47% for all 231 instances. Considering the αNpMP results, our heuristic obtained better results than a heuristic from literature in all 80 instances tested, finding optimal solutions in all these instances.

A Method for Enhancing the Simulation Continuity of the Snesim Algorithm in 2D Using Multiple Search Trees

Multiple-point geostatistics (MPS) has more advantages than two-point geostatistics in reproducing the continuity of geobodies in subsurface reservoir modeling. For fluvial reservoir modeling, the more continuous a channel, the more consistent it is with geological knowledge in general, and fluvial continuity is also of paramount importance when simulating fluid flow. Based on the pixel-based MPS algorithm Snesim, this study proposes a method that utilizes multiple search trees (MSTs) to enhance simulation continuity in 2D fluvial reservoir modeling. The objective of the MST method is to capture complete data events from a training image (TI), which aims to achieve enhanced continuity in fluvial reservoir sublayer modeling. By resorting to search neighborhoods based on their proximity to the central node of the data template, multiple data templates that correspond to the MSTs will be generated. Here, four data templates were generated by arranging the relative search neighborhood coordinates in ascending and descending order with respect to the central node. Parallel computing was tried for the construction of the search trees. This work calculated the conditional probability distribution function (CPDF) of the simulating nodes by averaging the CPDFs derived from the MSTs, and double retrieval was employed to filter out the search trees that possessed an inaccurate local CPDF for the simulating nodes. In addition, the connected component labeling (CCL) method was introduced to evaluate the simulation continuity in MPS. The results indicated that the MST method can enhance the simulation continuity of the Snesim algorithm by reproducing the fine connectivity of channel facies in 2D fluvial reservoir modeling.

Dynamic repositioning in bike-sharing systems with uncertain demand: An improved rolling horizon framework

The bike-sharing system (BSS) serves as a vital shared mobility mechanism, providing first- and last-mile services in urban transportation. However, one ongoing challenge in the BSS is the spatiotemporal imbalance caused by fluctuating user demand. To address this issue, this paper studies a dynamic bike repositioning problem under scenario-based demand to achieve a balance in inventory levels for electronic fences. Specifically, a multi-period two-stage stochastic model is proposed to make the trade-off between operational costs and service quality. The model incorporates a vehicle no-return (VNR) strategy, which allows vehicles to remain at the last station at the end of each period, awaiting the next repositioning task. To solve the model, an improved rolling horizon framework is introduced to effectively handle the temporal dimension complexity of the multi-period problem. Then a customized hybrid heuristic algorithm based on parallel genetic algorithm and variable neighborhood search is developed to obtain high-quality solutions. To validate the effectiveness and applicability of the proposed method, small-scale and practical-scale numerical experiments are specifically conducted based on real-world BSS data from Beijing, China. The results demonstrate the proposed method (i) creates a substantial reduction in the number of unmet demands when multiple scenarios potentially occur compared to the determined model, (ii) yields an average reduction of 29.04% in operational costs compared to the scheme without the VNR strategy, (iii) overcomes the short-sightedness inherent in the conventional single-period rolling horizon approach, exhibiting closer to optimal performance, and (iv) is superior to baseline approaches in terms of solution quality and efficiency.

Integrated location and inventory planning in service parts logistics with last-mile delivery outsourcing

Outsourcing last-mile delivery to express companies provides an innovative strategy for capital goods manufacturers to improve service parts logistics. Given delivery dilemma due to geographically dispersed customers and low-frequency uncertain demands, we develop an integrated location and inventory problem with last-mile delivery outsourcing in a two-echelon service parts logistics. The proposed non-linear mixed integer programming model captures important decisions involving (1) location of local warehouses, (2) customer allocation to opened local warehouses, (3) inventory levels at these local warehouses, (4) outsourcing or self-operating last-mile delivery. We then replace service level constraints that cause nonlinearity with two new sets of constraints, and construct a linear mixed integer formulation that could be solved by GUROBI solver for some small-scale problems. It is still a challenging problem for large scale case, for which we propose an adaptive large neighbourhood search (ALNS) algorithm. Our extensive computational experiments verify the effectiveness of our model and algorithm. Computational experiments’ results indicate that outsourcing last-mile delivery amounting to around 12 % total cost savings on average, compared to self-operating in a real-world large instance from our research partner. Furthermore, we test some key parameters with computational experiments, and provide some managerial implications.

A multi-population multi-tasking variable neighborhood search algorithm with diversity enhancements for inter-domain path computation problem

In recent years, as networks have become increasingly complex and expansive, network navigation faces new challenges in routing and resource utilization, particularly in the case of multi-domain networks. The Inter-Domain Path Computation under Node-defined Domain Uniqueness Constraint (IDPC-NDU) problem is one of many such scenarios. The problem requires finding the shortest path in a multi-domain network, but it cannot visit any domain twice. Since this is a proven NP-Hard problem, it is reasonable to approach IDPC-NDU from a meta-heuristics perspective. By integrating the search prowess of Variable Neighborhood Search (VNS) with the meta-learning capabilities of Multi-population-based Multitasking (MM), this paper proposes a Multi-population multi-tasking VNS (MM-VNS) algorithm to solve the Inter-Domain Path Computation under Domain Uniqueness constraint (IDPC-DU). In this combination, VNS is used to exploit solution space while meta-learning maintains diversity by transferring knowledge. Numerous experiments performed indicate the proposed approach finds better solutions in comparison with the state-of-the-art metaheuristics in many cases.

Bridging Software-Hardware for CXL Memory Disaggregation in Billion-Scale Nearest Neighbor Search

We propose CXL-ANNS , a software-hardware collaborative approach to enable scalable approximate nearest neighbor search (ANNS) services. To this end, we first disaggregate DRAM from the host via compute express link (CXL) and place all essential datasets into its memory pool. While this CXL memory pool allows ANNS to handle billion-point graphs without an accuracy loss, we observe that the search performance significantly degrades because of CXL’s far-memory-like characteristics. To address this, CXL-ANNS considers the node-level relationship and caches the neighbors in local memory, which are expected to visit most frequently. For the uncached nodes, CXL-ANNS prefetches a set of nodes most likely to visit soon by understanding the graph traversing behaviors of ANNS. CXL-ANNS is also aware of the architectural structures of the CXL interconnect network and lets different hardware components collaborate with each other for the search. Furthermore, it relaxes the execution dependency of neighbor search tasks and allows ANNS to utilize all hardware in the CXL network in parallel. Our evaluation shows that CXL-ANNS exhibits 93.3% lower query latency than state-of-the-art ANNS platforms that we tested. CXL-ANNS also outperforms an oracle ANNS system that has unlimited local DRAM capacity by 68.0%, in terms of latency.

Research on a Joint Distribution Vehicle Routing Problem Considering Simultaneous Pick-Up and Delivery under the Background of Carbon Trading

In order to explore the positive impact of the joint distribution model on the reduction in logistics costs in small-scale logistics enterprises, considering the demand on enterprises for simultaneous pick-up and delivery, as well as the cost of carbon emissions, this study considers the vehicle routing problem of simultaneous pick-up and delivery under a joint distribution model. First of all, an independent distribution model and a joint distribution model including fixed transportation, variable transportation, time penalty, and carbon emissions costs are established; second, by adding the self-adaption cross-mutation probability and the destruction and repair mechanism in the large-scale neighborhood search algorithm, the genetic algorithm is improved to adapt to the solution of the model in this paper, and the effectiveness of the improved algorithm is verified and analyzed. It is found that the improved genetic algorithm is more advantageous than the original algorithm for solving the problems of both models designed in this paper. Finally, the improved genetic algorithm is used to solve the two models, and the results are compared and analyzed. It is found that the joint distribution model can reduce the total cost by 6.61% and the carbon emissions cost by 5.73%. Additionally, the impact of the carbon trading mechanism on the simultaneous pick-up and delivery vehicle routing problem under the joint distribution model is further explored. The results of this study prove that enterprises can effectively reduce costs, improve profits, reduce carbon emissions, and promote the sustainable development of logistics enterprises under the condition of joint distribution.

Heuristic Algorithms for Heterogeneous and Multi-Trip Electric Vehicle Routing Problem with Pickup and Delivery

In light of the widespread use of electric vehicles for urban distribution, this paper delves into the electric vehicle routing problem (EVRP): specifically addressing multiple trips per vehicle, diverse vehicle types, and simultaneous pickup and delivery. The primary objective is to minimize the overall cost, which encompasses travel expenses, waiting times, recharging costs, and fixed vehicle costs. The focal problem is formulated as a heterogeneous and multi-trip electric vehicle routing problem with pickup and delivery (H-MT-EVRP-PD). Additionally, we introduce two heuristic algorithms to efficiently approximate solutions within a reasonable computational time. The variable neighborhood search (VNS) algorithm and the adaptive large neighborhood search (ALNS) algorithm are presented and compared based on our computational experiences with both. Through solving a series of large-scale real-world instances for the H-MT-EVRP-PD and smaller instances using an exact method, we demonstrate the efficacy of the proposed approaches.

A Quantum-Inspired Ant Colony Optimization Algorithm for Parking Lot Rental to Shared E-Scooter Services

Electric scooter sharing mobility services have recently spread in major cities all around the world. However, the bad parking behavior of users has become a major source of issues, provoking accidents and compromising urban decorum of public areas. Reducing wild parking habits can be pursued by setting reserved parking spaces. In this work, we consider the problem faced by a municipality that hosts e-scooter sharing services and must choose which locations in its territory may be rented as reserved parking lots to sharing companies, with the aim of maximizing a return on renting and while taking into account spatial consideration and parking needs of local residents. Since this problem may result difficult to solve even for a state-of-the-art optimization software, we propose a hybrid metaheuristic solution algorithm combining a quantum-inspired ant colony optimization algorithm with an exact large neighborhood search. Results of computational tests considering realistic instances referring to the Italian capital city of Rome show the superior performance of the proposed hybrid metaheuristic.

Solving the probabilistic drone routing problem: Searching for victims in the aftermath of disasters

AbstractSeveral major industrial disasters happen each year around the world. They usually involve limited accessibility, poor ground conditions, and toxic wastes. As a consequence, this reduces the efficiency of humanitarian operations. In such a context, flying drones may be a viable alternative: faster, no dependency on ground conditions, and larger areas scanned. They are also better suited for following the population and the crisis dynamic. For such a purpose, various issues have to be addressed such as defining and optimizing the drone's routes, their energy consumption, choosing the relay points for recharging equipment, among others. In this study, several additional features from existing works are considered: first, a probability of identifying individuals is defined. Thus, each node can be scanned several times in order to improve the observation. In addition, the nodes are prioritized according to a given heatmap. The probabilistic drone routing problem (PDRP) consists of finding a route, that is, a sequence of trips, for each drone such that the sum of the expected number of identified individuals on all routes is maximized. Constraints on energy consumption, collision avoidance and drone‐base assignment are considered. We propose a heuristic and metaheuristics based on the adaptive large neighborhood search for the PDRP. The methods are tested on theoretical instances, as well as on a case study of the Beirut Port explosion on August 4, 2020, in order to analyze the performance of the proposed methods.

A New Simheuristic Approach for Stochastic Runway Scheduling

We consider a stochastic, dynamic runway scheduling problem involving aircraft landings on a single runway. Sequencing decisions are made with knowledge of the estimated arrival times (ETAs) of all aircraft due to arrive at the airport, and these ETAs vary according to continuous-time stochastic processes. Time separations between consecutive runway landings are modeled via sequence-dependent Erlang distributions and are affected by weather conditions, which also evolve continuously over time. The resulting multistage optimization problem is intractable using exact methods, and we propose a novel simheuristic approach based on the application of methods analogous to variable neighborhood search in a high-dimensional stochastic environment. Our model is calibrated using flight tracking data for over 98,000 arrivals at Heathrow Airport. Results from numerical experiments indicate that our proposed simheuristic algorithm outperforms an alternative based on deterministic forecasts under a wide range of parameter values, with the largest benefits seen when the underlying stochastic processes become more volatile and also when the on-time requirements of individual flights are given greater weight in the objective function. Funding: This work was supported by the Engineering and Physical Sciences Research Council [Grant EP/M020258/1]. Supplemental Material: The online appendix and data files are available at https://doi.org/10.1287/trsc.2022.0400 .

CRISPR-COPIES: an in silico platform for discovery of neutral integration sites for CRISPR/Cas-facilitated gene integration.

The CRISPR/Cas system has emerged as a powerful tool for genome editing in metabolic engineering and human gene therapy. However, locating the optimal site on the chromosome to integrate heterologous genes using the CRISPR/Cas system remains an open question. Selecting a suitable site for gene integration involves considering multiple complex criteria, including factors related to CRISPR/Cas-mediated integration, genetic stability, and gene expression. Consequently, identifying such sites on specific or different chromosomal locations typically requires extensive characterization efforts. To address these challenges, we have developed CRISPR-COPIES, a COmputational Pipeline for the Identification of CRISPR/Cas-facilitated intEgration Sites. This tool leverages ScaNN, a state-of-the-art model on the embedding-based nearest neighbor search for fast and accurate off-target search, and can identify genome-wide intergenic sites for most bacterial and fungal genomes within minutes. As a proof of concept, we utilized CRISPR-COPIES to characterize neutral integration sites in three diverse species: Saccharomyces cerevisiae, Cupriavidus necator, and HEK293T cells. In addition, we developed a user-friendly web interface for CRISPR-COPIES (https://biofoundry.web.illinois.edu/copies/). We anticipate that CRISPR-COPIES will serve as a valuable tool for targeted DNA integration and aid in the characterization of synthetic biology toolkits, enable rapid strain construction to produce valuable biochemicals, and support human gene and cell therapy applications.

Quantum‐Behaved Particle Swarm Optimization Based on Concentration Selection Probability Assignment Weights for Power System Economic Dispatch

The economic dispatch (ED) problem is very important in the economics of power systems, and a suitable dispatch algorithm can help power plants save huge amounts of money and use energy efficiently. To ensure the economics of the power system dispatch strategy, this paper proposes a quantum‐behaved particle swarm improvement algorithm for solving the ED problem. First, the algorithm addresses the problem that the classical quantum‐behaved particle swarm optimization (QPSO) tends to be premature and falls into local optimality. We propose a method to calculate the mean best position by assigning weights to different particles, by adopting the idea of concentration selection in the immune concentration regulation mechanism, and designing a Weight‐Concentration selection probability function, so that those particles located outside the local optimal solution region but with smaller fitness value and better evolutionary trend can be retained. Second, to make the algorithm focus on global search at the beginning of the iteration and more focused search at the end to enhance the neighborhood search capability at the later stage, a decreasing function of the slope K in the Weight‐Concentration selection probability function is designed so that the influence of the weighting term on the calculation of the mean best position decreases gradually. The algorithm is evaluated with four power systems, and the results show that the overall performance of the improved algorithm is better than other algorithms in solving the ED problem in terms of minimum solution, convergence, and stability. © 2024 Institute of Electrical Engineer of Japan and Wiley Periodicals LLC.

Hybrid truck-drone delivery system with multi-visits and multi-launch and retrieval locations: Mathematical model and adaptive variable neighborhood search with neighborhood categorization

Drones have recently been suggested as a means of performing last-mile deliveries, as they have several advantages compared to traditional delivery vehicles. A recent research avenue is to adopt a hybrid truck-drone delivery system that integrates drones with traditional delivery methods such as trucks. This paper deals with the problem of optimizing the delivery operation of a truck working in tandem with a drone capable of visiting multiple customers per dispatch. We also introduce practical attributes such as allowing the truck to launch and retrieve the drone from both customer and non-customer nodes as well as permitting cyclic and acyclic drone operations. An integer linear programming model is first formulated followed by the development of an effective variable neighborhood search (VNS)-based approach. The novelty of the latter is that it incorporates an effective categorisation of neighborhoods due to their large number that is needed while retaining their individual impact through an adaptive selection scheme. The performance of this powerful VNS-based heuristic is empirically assessed against three variants of the VNS. The VNS heuristic was also shown to be flexible and effective at handling the issue of synchronization. A sensitivity analysis, based on some of the critical parameters of the drone, is conducted alongside highlights of some interesting insights.

Toward Polydisperse Flows With mfix-exa

Abstract In the presence of large size disparities, single-grid neighbor search algorithms lead to inflated neighbor lists that significantly degrade the performance of Lagrangian particle solvers. If Eulerian–Lagrangian (EL) frameworks are to remain performant when simulating realistic systems, improved neighbor detection approaches must be adopted. To this end, we consider the application of a multigrid neighbor search (MGNS) algorithm in the mfix-exa software package, an exascale EL solver built upon the AMReX library. Details regarding the implementation and verification of MGNS are provided along with speedup curves for a bidisperse mixing layer. MGNS is shown to yield up to 15× speedup on CPU and 6× speedup on GPU for the problems considered here. The mfix-exa software is then validated for a variety of polydisperse flows. Finally, a brief discussion is given for how dynamic MGNS may be completed, with application to spatially varying particle size distributions.