Minimizing durations in repetitive projects through adaptive large neighborhood search

A new constraint programming model and a linear programming-based adaptive large neighborhood search for the vehicle routing problem with synchronization constraints

In this paper, a two-stage stochastic mathematical model is developed for an asset protection routing problem under a wildfire. The main aim of this study is to reduce the negative impact of a wildfire. Some parameters, such as travel and service times, obtaining profit by protecting an asset, and upper bounds of time windows, are considered as stochastic parameters. Generating proper scenarios for uncertain parameters has a large impact on the accuracy of the obtained solutions. Therefore, artificial neural networks are employed to extract possible scenarios according to previous actual wildfire events. The problem cannot be solved by exact solvers for large instances, so two matheuristic algorithms are proposed in this study to solve the problem in a reasonable time. In the first algorithm, a set of feasible routes is generated based on a heuristic approach, then a route-based mathematical model is used to obtain the final solution. Also, another matheuristic algorithm based on adaptive large neighbourhood search (ALNS) is proposed. In this algorithm, routing decisions are determined using the ALNS algorithm while other decisions are achieved by solving an intermediate mathematical model. The numerical analysis confirms the efficiency of both proposed algorithms; however, the first algorithm performs more efficiently.

Adaptive large neighborhood search heuristics for multi-tier service deployment problems in clouds

Cross-docking is considered as a method to manage and control the inventory flow, which is essential in the context of supply chain management. This paper studies the integration of the vehicle routing problem with cross-docking, namely VRPCD which has been extensively studied due to its ability to reduce the overall costs occurring in a supply chain network. Given a fleet of homogeneous vehicles for delivering a single type of product from suppliers to customers through a cross-dock facility, the objective of VRPCD is to determine the number of vehicles used and the corresponding vehicle routes, such that the vehicle operational and transportation costs are minimized. An adaptive large neighborhood search (ALNS) algorithm is proposed to solve the available benchmark VRPCD instances. The experimental results show that ALNS is able to improve 80 (out of 90) best known solutions and obtain the same solution for the remaining 10 instances within short computational time. We also explicitly analyze the added value of using an adaptive scheme and the implementation of the acceptance criteria of Simulated Annealing (SA) into the ALNS, and also present the contribution of each ALNS operator towards the solution quality.

An efficient adaptive large neighborhood search algorithm based on heuristics and reformulations for the generalized quadratic assignment problem

This article introduces an advanced solution to optimize street sweeping operations by extending a multi-depot arc routing problem. The key enhancement involves flexible end depot assignments, where vehicles start and conclude shifts at designated depots. A notable constraint requires subsequent shifts to begin from the destination depot of the preceding shift. The problem involves servicing highway exclusively during night shifts, while other arc types can be addressed during both day and night. The objective is to identify optimal shifts meeting practical criteria while adhering to constraints like maximum shift duration. To address this, a mixed-integer linear programming (MILP) model is presented. It aims to minimize the number of shifts and total travel time. Given the computational complexity of large instances, an adaptive large neighbourhood search (ALNS) metaheuristic was developed. This approach incorporates specialized operators that address unique attributes such as arc type and depot assignments, ensuring arcs are repositioned based on their type and proximity to depots. This tailored approach provides a distinct advantage over classical ALNS operators, as numerical tests indicate that the specialized operators are more efficient in comparison. The approach is evaluated on larger and a real-world instances, demonstrating notable performance in solution quality and computational efficiency.

This article systematically reviews the literature on adaptive large neighborhood search (ALNS) to gain insights into the operators used for vehicle routing problems (VRPs) and their effectiveness. The ALNS has been successfully applied to a variety of optimization problems, particularly variants of the VRP. The ALNS gradually improves an initial solution by modifying it using removal and insertion operators. However, relying solely on adaptive operator selection is not advisable. Instead, authors often conduct experiments to identify operators that improve the solution quality or remove detrimental ones. This process is mostly cumbersome due to the wide variety of operators, further complicated by inconsistent nomenclature. The objectives of this review are threefold: First, to classify ALNS operators using a unified terminology; second, to analyze their performance; and third, to present guidelines for the development and analysis of ALNS algorithms in the future based on the outcomes of the performance evaluation. In this review, we conduct a network meta-analysis of 211 articles published between 2006 and 2023 that have applied ALNS algorithms in the context of VRPs. We employ incomplete pairwise comparison matrices, similar to rankings used in sports, to rank the operators. We identify 57 distinct removal and 42 insertion operators, and the analysis ranks them based on their effectiveness. Sequence-based removal operators, which remove sequences of customers in the current solution, are found to be the most effective. The best-performing insertion operators are those that exhibit foresight, such as regret insertion operators. Finally, guidelines and possible future research directions are discussed.

In this paper an adaptation of the Adaptive Large Neighborhood Search (ALNS) to a patient’s care planning problem is proposed. We formalize it as an RCPSP problem that consists of assigning a start date and medical resources to a set of medical appointments. Different intensification and diversification movements for the ALNS are presented. We test this approach on real-life problems and compare the results of ALNS to a version without the adaptive layer, called (\(\lnot \)A)LNS. We also compare our results with the ones obtained with a 0–1 linear programming model. On small instances, ALNS obtains results close to optimality, with an average difference of 1.39 of solution quality. ALNS outperforms (\(\lnot \)A)LNS with a gain of up to 18.34% for some scenarios.

Large Neighborhood Search (LNS) heuristics are among the most powerful but also most expensive heuristics for mixed integer programs (MIP). Ideally, a solver adaptively concentrates its limited computational budget by learning which LNS heuristics work best for the MIP problem at hand. To this end, this work introduces Adaptive Large Neighborhood Search (ALNS) for MIP, a primal heuristic that acts as a framework for eight popular LNS heuristics such as Local Branching and Relaxation Induced Neighborhood Search (RINS). We distinguish the available LNS heuristics by their individual search spaces, which we call auxiliary problems. The decision which auxiliary problem should be executed is guided by selection strategies for the multi armed bandit problem, a related optimization problem during which suitable actions have to be chosen to maximize a reward function. In this paper, we propose an LNS-specific reward function to learn to distinguish between the available auxiliary problems based on successful calls and failures. A second, algorithmic enhancement is a generic variable fixing prioritization, which ALNS employs to adjust the subproblem complexity as needed. This is particularly useful for some LNS problems which do not fix variables by themselves. The proposed primal heuristic has been implemented within the MIP solver SCIP. An extensive computational study is conducted to compare different LNS strategies within our ALNS framework on a large set of publicly available MIP instances from the MIPLIB and Coral benchmark sets. The results of this simulation are used to calibrate the parameters of the bandit selection strategies. A second computational experiment shows the computational benefits of the proposed ALNS framework within the MIP solver SCIP.

To overcome the limitations of traditional methods in emergency response scenarios—such as limited adaptability during the search process and a tendency to fall into local optima, which reduce the overall efficiency of emergency supply distribution—this study develops a Vehicle Routing Problem (VRP) model that incorporates multiple constraints, including service time windows, demand satisfaction, and fleet size. A multi-objective optimization function is formulated to minimize the total travel time, reduce distribution imbalances, and maximize demand satisfaction. To solve this problem, a hybrid deep reinforcement learning framework is proposed that integrates an Adaptive Large Neighborhood Search (ALNS) with Proximal Policy Optimization (PPO). In this framework, ALNS provides the baseline search, whereas the PPO policy network dynamically adjusts the operator weights, acceptance criteria, and perturbation intensities to achieve adaptive search optimization, thereby improving global solution quality. Experimental validation of benchmark instances of different scales shows that, compared with two baseline methods—the traditional Adaptive Large Neighborhood Search (ALNS) and the Improved Ant Colony Algorithm (IACA)—the proposed algorithm reduces the average objective function value by approximately 23.6% and 25.9%, shortens the average route length by 7.8% and 11.2%, and achieves notable improvements across multiple performance indicators.

Meta-analysis, a systematic statistical examination that combines the results of several independent studies, has the potential of obtaining problem- and implementation-independent knowledge and understanding of metaheuristic algorithms, but has not yet been applied in the domain of operations research. To illustrate the procedure, we carried out a meta-analysis of the adaptive layer in adaptive large neighborhood search (ALNS). Although ALNS has been widely used to solve a broad range of problems, it has not yet been established whether or not adaptiveness actually contributes to the performance of an ALNS algorithm. A total of 134 studies were identified through Google Scholar or personal e-mail correspondence with researchers in the domain, 63 of which fit a set of predefined eligibility criteria. The results for 25 different implementations of ALNS solving a variety of problems were collected and analyzed using a random effects model. This dataset contains a detailed comparison of ALNS with the non-adaptive variant per study and per instance, together with the meta-analysis summary results. The data enable to replicate the analysis, to evaluate the algorithms using other metrics, to revisit the importance of ALNS adaptive layer if results from more studies become available, or to simply consult the ready-to-use formulas in the summary file to carry out a meta-analysis of any research question. The individual studies, the meta-analysis and its results are described and interpreted in detail in Renata Turkeš, Kenneth Sörensen, Lars Magnus Hvattum, Meta-analysis of Metaheuristics: Quantifying the Effect of Adaptiveness in Adaptive Large Neighborhood Search, in the European Journal of Operational Research.

Background: The Multi Depot Pickup and Delivery Problem with Time Windows and Heterogeneous Vehicle Fleets (MDPDPTWHV) is a strongly practically oriented routing problem with many real-world constraints. Due to its complexity, solution approaches with sufficiently good quality ideally contain several operators with certain probabilities.Thus, automatically selecting the best parameter configurations enhances the overall solution quality. Methods: To solve the MDPDPTWHV, we present a Grouping Genetic Algorithm (GGA) framework with several operators and population management variants. A Bayesian Optimization (BO) approach is introduced to optimize the GGA’s parameter configuration. The parameter tuning is evaluated on five data sets which differ in several structural characteristics and contain 1200 problem instances. The outcomes of the parameter-tuned GGA are compared to both the initial GGA parameter configuration and a state-of-the-art Adaptive Large Neighborhood Search (ALNS). Results: The presented GGA framework achieves a better solution quality than the ALNS, even for the initial parameter configuration used. The mean value of the relative error is less than 0.9% and its standard deviation is less than 1.31% for every problem class. For the ALNS, these values are up to three times higher and the GGA is up to 38% faster than the ALNS. Conclusions: It is shown that the BO, as a parameter tuning approach, is a good choice in improving the performance of the considered meta-heuristic over all instances in each data set. In addition, the best parameter configuration per problem class with the same characteristics is able to improve both the frequency of finding the best solution, as well as the relative error to this solution, significantly.

Aligning protein-protein interaction networks from different species is a useful mechanism for figuring out orthologous proteins, predicting/verifying protein unknown functions or constructing evolutionary relationships. The network alignment problem is proved to be NP-hard, requiring exponential-time algorithms, which is not feasible for the fast growth of biological data. In this paper, we present a novel protein-protein interaction global network alignment algorithm, which is enhanced with an extended large neighborhood search heuristics. Evaluated on benchmark datasets of yeast, fly, human and worm, the proposed algorithm outperforms state-of-the-art. Furthermore, the complexity of ours is polynomial, thus being scalable to large biological networks in practice.
 Keywords
 Heuristic, Protein-protein interaction networks, network alignment, neighborhood search
 References
 [1] R.L. Finley, R. Brent, Interaction mating reveals binary and ternary connections between drosophila cell cycle regulators. Proc. Natl Acad. Sci. USA. 91 (1994) 12980-12984.[2] R. Aebersold, M. Mann, Mass spectrometry-based proteomics, Nature. 422 (2003) 198-207.[3] C.S. Goh, F.R. Cohen, Co-evolutionary analysis reveals insights into protein-protein interactions, J. Mol. Biol. 324 (2002) 177-192.[4] J.D. Han et al, Evidence for dynamically organized modularity in the yeast proteinprotein interaction network, Nature. 430 (2004) 88-93.[5] G.D. Bader, C.W. Hogue, Analyzing yeast protein-protein interaction data obtained from different sources, Nat. Biotechnol. 20 (2002) 991-997.[6] H.B. Hunter et al, Evolutionary rate in the protein interaction network, Science. 296 (2002) 750-752.[7] J. Dutkowski, J. Tiuryn,J, Identification of functional modules from conserved ancestral protein-protein interactions, Bioinformatics. 23 (2007) i149-i158.[8] B.P. Kelley et al, Conserved pathways within bacteria and yeast as revealed by global protein network alignment, Proc. Natl Acad. Sci. USA. 100 (2003) 11394-11399.[9] O. Kuchaiev, N. Przˇ ulj, Integrative network alignment reveals large regions of global network similarity in yeast and human, Bioinformatics. 27 (2011) 1390-1396.[10] M. Remm et al, Automatic clustering of orthologs and in-paralogs from pairwise species comparisons, J. Mol. Biol. 314 (2001) 1041-1052. [11] L. Chindelevitch et al, Local optimization for global alignment of protein interaction networks, In: Pacific Symposium on Biocomputing, Hawaii, USA, 2010, pp. 123-132.[12] E. hmet, Aladağ, Cesim Erten, SPINAL: scalable protein interaction network alignment, Bioinformatics. Volume 29(7) (2013) 917-924. https://doi.org/10.1093/bioinformatics/btt071.[13] B.P. Kelley et al, Pathblast: a tool for alignment of protein interaction networks, Nucleic Acids Res. 32 (2004) 83-88.[14] R. Sharan et al, Conserved patterns of protein interaction in multiple species, Proc. Natl Acad. Sci. USA. 102 (2005) 1974-1979.[15] M. Koyuturk et al, Pairwise alignment of protein interaction networks, J. Comput. Biol. 13 (2006) 182-199.[16] M. Narayanan, R.M. Karp, Comparing protein interaction networks via a graph match-and-split algorithm, J. Comput. Biol. 14 (2007) 892-907.[17] J. Flannick et al, Graemlin: general and robust alignment of multiple large interaction networks, Genome Res. 16 (2006) 1169-1181.[18] R. Singh et al, Global alignment of multiple protein interaction networks. In: Pacific Symposium on Biocomputing, 2008, pp. 303-314.[19] M. Zaslavskiy et al, Global alignment of protein-protein interaction networks by graph matching methods, Bioinformatics. 25 (2009) 259-267.[20] L. Chindelevitch, Extracting information from biological networks. PhD Thesis, Department of Mathematics, Massachusetts Institute of Technology, Cambridge, 2010.[21] Do Duc Dong et al, An efficient algorithm for global alignment of protein-protein interaction networks, Proceeding of ATC15, 2015, pp. 332-336.[22] S. Ropke, D. Pisinger, An Adaptive Large Neighborhood Search Heuristic for the Pickup and Delivery Problem with Time Windows. Transportation Science. 40 (2006) 455-472. https:// doi.org/10.1287/trsc.1050.0135.[23] P. Shaw, A new local search algorithm providing high quality solutions to vehicle routing problems, Technical report, Department of Computer Science, University of Strathclyde, Scotland, 1997.[24] Roman Lutz, Adaptive Large Neighborhood Search, Bachelor thesis, Ulm University, 2014.[25] M.A. Trick, A linear relaxation heuristic for the generalized assignment prob-lem, Naval Research Logistics. 39 (1992) 137-151.[26] J.Y. Potvin, M. Rousseau, Parallel Route Building Algorithm for the Vehicle Routing and Scheduling Problem with Time Windows, European Journal of Operational Research. 66(3) (1993) pp. 331-340.[27] https://www.researchgate.net/figure/Network-alignment-a-A-dashed-arrow-from-a-node-i-V1-from-the-first-network-G1-V1-E_fig1_24017968[28] J.M. Peter, Van Laarhoven, H.L. Emile, Aarts. Simulated annealing. Springer, 1987.

The generalized lock scheduling problem (GLSP) is a mixed integer optimization problem which consists of a ship placement (SP) and a lockage operation scheduling (LOS) sub-problem. In previous research, the GLSP is solved by different exact and heuristic methods, which are confirmed inferior with respect to computation time and solution quality. Consequently, none of those methods is efficient for handling practical large-scale GLSP. For the first time, we show that high-quality solutions of GLSP can be efficiently obtained by using an innovative approach proposed in this paper. Specifically, an ingenious solution structure of GLSP is designed, by which the GLSP is converted to a combinatorial optimization problem. Furthermore, an adaptive large neighborhood search (ALNS) heuristic based on the principle of destruction and reconstruction of solutions is proposed for solving the GLSP. Test results using a large number of instances reported in the literature are compared with those obtained by two exact methods, the mixed integer linear programming (MILP) and combinatorial Benders' decomposition (CBD) method. The results show that our ALNS achieves optimal solutions within less time in terms of most of the small-scale instances. Much better solutions are obtained by the ALNS within a few minutes for those large-scale instances that cannot be solved to optimality by exact methods within 2 h. Especially, the advantage of the proposed method is more remarkable when there is no specific chronological rules forced, which indicates that the proposed method is capable of handling the GLSP in a broader scope of situations.

An adaptive large neighborhood search procedure applied to the dynamic patient admission scheduling problem.