Classical Meta-heuristic Algorithms Research Articles

Multi-Strategy Improved Dung Beetle Optimization Algorithm and Its Applications.

The dung beetle optimization (DBO) algorithm, a swarm intelligence-based metaheuristic, is renowned for its robust optimization capability and fast convergence speed. However, it also suffers from low population diversity, susceptibility to local optima solutions, and unsatisfactory convergence speed when facing complex optimization problems. In response, this paper proposes the multi-strategy improved dung beetle optimization algorithm (MDBO). The core improvements include using Latin hypercube sampling for better population initialization and the introduction of a novel differential variation strategy, termed "Mean Differential Variation", to enhance the algorithm's ability to evade local optima. Moreover, a strategy combining lens imaging reverse learning and dimension-by-dimension optimization was proposed and applied to the current optimal solution. Through comprehensive performance testing on standard benchmark functions from CEC2017 and CEC2020, MDBO demonstrates superior performance in terms of optimization accuracy, stability, and convergence speed compared with other classical metaheuristic optimization algorithms. Additionally, the efficacy of MDBO in addressing complex real-world engineering problems is validated through three representative engineering application scenarios namely extension/compression spring design problems, reducer design problems, and welded beam design problems.

Reentrant hybrid flow shop scheduling with stockers in automated material handling systems using deep reinforcement learning

In smart factories, automated material handling system (AMHSs) replace manual material handling to increase production efficiency, in which stockers serve as the temporary storage for work-in-process inventories. Furthermore, costly machines and complex processes in advanced manufacturing lead to the need for multiple reentrant processes of the same workstation. However, previous works on scheduling problems rarely considered the function of stockers and reentrant processes, and their approaches were mostly based on metaheuristic algorithms. Recent advances in artificial intelligence enable the possibility of solving scheduling problems using deep reinforcement learning (DRL). Therefore, this work investigates the reentrant hybrid flow shop scheduling problem with stockers (RHFS2) inspired by AMHSs, and solves it by DRL. Firstly, the states, actions, and rewards of the Markov decision process for this problem are designed; and then, two deep Q network (DQN) approaches based on the actions for determining machines and jobs, respectively, are proposed. Simulation results demonstrate that our proposed DQN approaches outperform for finding better solutions of different-scale problems than classical metaheuristic algorithms.

Hybrid Whale Optimization with a Firefly Algorithm for Function Optimization and Mobile Robot Path Planning.

With the wide application of mobile robots, mobile robot path planning (MRPP) has attracted the attention of scholars, and many metaheuristic algorithms have been used to solve MRPP. Swarm-based algorithms are suitable for solving MRPP due to their population-based computational approach. Hence, this paper utilizes the Whale Optimization Algorithm (WOA) to address the problem, aiming to improve the solution accuracy. Whale optimization algorithm (WOA) is an algorithm that imitates whale foraging behavior, and the firefly algorithm (FA) is an algorithm that imitates firefly behavior. This paper proposes a hybrid firefly-whale optimization algorithm (FWOA) based on multi-population and opposite-based learning using the above algorithms. This algorithm can quickly find the optimal path in the complex mobile robot working environment and can balance exploitation and exploration. In order to verify the FWOA's performance, 23 benchmark functions have been used to test the FWOA, and they are used to optimize the MRPP. The FWOA is compared with ten other classical metaheuristic algorithms. The results clearly highlight the remarkable performance of the Whale Optimization Algorithm (WOA) in terms of convergence speed and exploration capability, surpassing other algorithms. Consequently, when compared to the most advanced metaheuristic algorithm, FWOA proves to be a strong competitor.

Snow Geese Algorithm: A novel migration-inspired meta-heuristic algorithm for constrained engineering optimization problems

This paper proposes a novel nature-inspired meta-heuristic algorithm, named Snow Geese Algorithm. It is inspired by the migratory behavior of snow geese and emulates the distinctive “Herringbone” and “Straight Line” shaped flight patterns observed during their migration. The algorithm is structured into three main phases for benchmark testing. In the first phase, the Snow Geese Algorithm's numerical results are compared with those of several classical meta-heuristic algorithms using the same test functions and original data from these algorithms. In the second phase, in order to minimize potential variations during the comparison, all algorithms undergo evaluation on a standardized testing platform. In the third phase, this paper applies the Snow Geese Algorithm to solve four widely recognized engineering optimization problems: the tubular column design, piston lever optimization design, reinforced concrete beam design and car side impact design. These real-world engineering problems serve as test cases to assess Snow Geese Algorithm problem-solving capabilities. The primary objective of the Snow Geese Algorithm is to provide an alternative perspective for tackling complex optimization problems. Please note that the complete source code for the Snow Geese Algorithm is publicly available at https://github.com/stones3421/SGA-project.

Opposition-Based Learning Equilibrium Optimizer with Application in Mobile Robot Path Planning

Abstract: The objective of mobile robot path planning (MRPP) is to devise the shortest obstacle-free path for autonomous mobile robots based on a given terrain. Numerous MRPP methods have been extensively researched. This paper presents a novel approach called Opposition-based Learning Equilibrium Optimizer (OEO) for generating smooth paths for mobile robots. The fundamental idea behind OEO is to introduce an opposition-based learning mechanism while maintaining the overall framework of the basic EO algorithm. This modification alleviates the susceptibility of the basic EO algorithm to local optima. The OEO algorithm is employed to provide smooth paths for autonomous mobile robots, and the results are compared with several classical metaheuristic algorithms. Comparative analysis across different environments demonstrates that the proposed OEO-based path planning method consistently yields the shortest and most collision-free paths with superior stability.

Enhanced Harris hawks optimization-based fuzzy k-nearest neighbor algorithm for diagnosis of Alzheimer's disease

In order to stop deterioration and give patients with Alzheimer's disease (AD) early therapy, it is crucial to correctly diagnose AD and its early stage, mild cognitive impairment (MCI). A framework for diagnosing AD is presented in this paper, which includes magnetic resonance imaging (MRI) image preprocessing, feature extraction, and the Fuzzy k-nearest neighbor algorithm (FKNN) model. In particular, the framework's novelty lies in the use of an improved Harris Hawks Optimization (HHO) algorithm named SSFSHHO, which integrates the Sobol sequence and Stochastic Fractal Search (SFS) mechanisms for optimizing the parameters of FKNN. The HHO method improves the quality of the initial population overall by incorporating the Sobol sequence, and the SFS mechanism increases the algorithm's capacity to get out of the local optimum solution. Comparisons with other classical meta-heuristic algorithms, state-of-the-art HHO variants in low and high dimensions, and enhanced meta-heuristic algorithms on 30 typical IEEE CEC2014 benchmark test problems show that the overall performance of SSFSHHO is significantly better than other comparative algorithms. Moreover, the created framework based on the SSFSHHO-FKNN model is employed to classify AD and MCI using MRI scans from the ADNI dataset, achieving high classification performance for 6 representative cases. Experimental findings indicate that the proposed algorithm performs better than a number of high-performance optimization algorithms and classical machine learning algorithms, thus offering a promising approach for AD classification. Additionally, the proposed strategy can successfully identify relevant features and enhance classification performance for AD diagnosis.

DETDO: An adaptive hybrid dandelion optimizer for engineering optimization

Dandelion Optimizer (DO) is a recently proposed swarm intelligence algorithm that coincides with the process of finding the best reproduction site for dandelion seeds. Compared with the classical Meta-heuristic algorithms, DO exhibits strong competitiveness, but it also has some drawbacks. In this paper, we proposed an adaptive hybrid dandelion optimizer called DETDO by combining three strategies of adaptive tent chaotic mapping, differential evolution (DE) strategy, and adaptive t-distribution perturbation to address the shortcomings of weak DO development, easy to fall into local optimum and slow convergence speed. Firstly, the adaptive tent chaos mapping is used in the initialization phase to obtain a uniformly distributed high-quality initial population, which helps the algorithm to enter the correct search region quickly. Secondly, the DE strategy is introduced to increase the diversity of dandelion populations to avoid algorithm stagnation, which improves the exploitation capability and the accuracy of the optimal solution. Finally, adaptive t-distribution perturbation around the elite solution successfully balances the exploration and exploitation phases while improving the convergence speed through a reasonable conversion from Cauchy to Gaussian distribution. The proposed DETDO is compared with classical or advanced optimization algorithms on CEC2017 and CEC2019 test sets, and the experimental results and statistical analysis demonstrate that the algorithm has better optimization accuracy and speed. In addition, DETDO has obtained the best results in solving six real-world engineering design problems. Finally, DETDO is applied to two bar topology optimization cases. Under a series of complex constraints, DETDO produces a lighter bar structure than the current scheme. It further illustrates the effectiveness and applicability of DETDO in practical problems. The above results mean that DETDO with strong competitiveness will become a preferred swarm intelligence algorithm to cope with optimization problems.

Discrete komodo algorithm for traveling salesman problem

This work presents a Discrete Komodo Algorithm (DKA) to solve the traveling salesman problem (TSP). Inspired by the Komodo Mlipir Algorithm, a swarm intelligence algorithm based on the behavior of komodo dragons, DKA introduces an enhanced edge-construction operator and a novel edge-destruction operator to explore the discrete search space of TSP. To prioritize the exploration of potential edges, DKA uses a Priority Queue that stores all possible edges with their weights, and its counts appear on tour as its priority. The edges with smaller weights and counts have higher probabilities of being selected for the tour. DKA is tested with 45 different TSP instances from TSPLIB. Experimental results show that DKA guarantees produced the best-known solution for small test problems for up to 24 nodes, and the best solutions are within 6.92% of the known optimal solutions. Compared with some of the previous algorithms, DKA outperforms the recent state-of-the-art algorithms and is significantly better than classical metaheuristic algorithms by achieving a lower relative error.

White Blood Cells Classification Using Entropy-Controlled Deep Features Optimization.

White blood cells (WBCs) constitute an essential part of the human immune system. The correct identification of WBC subtypes is critical in the diagnosis of leukemia, a kind of blood cancer defined by the aberrant proliferation of malignant leukocytes in the bone marrow. The traditional approach of classifying WBCs, which involves the visual analysis of blood smear images, is labor-intensive and error-prone. Modern approaches based on deep convolutional neural networks provide significant results for this type of image categorization, but have high processing and implementation costs owing to very large feature sets. This paper presents an improved hybrid approach for efficient WBC subtype classification. First, optimum deep features are extracted from enhanced and segmented WBC images using transfer learning on pre-trained deep neural networks, i.e., DenseNet201 and Darknet53. The serially fused feature vector is then filtered using an entropy-controlled marine predator algorithm (ECMPA). This nature-inspired meta-heuristic optimization algorithm selects the most dominant features while discarding the weak ones. The reduced feature vector is classified with multiple baseline classifiers with various kernel settings. The proposed methodology is validated on a public dataset of 5000 synthetic images that correspond to five different subtypes of WBCs. The system achieves an overall average accuracy of 99.9% with more than 95% reduction in the size of the feature vector. The feature selection algorithm also demonstrates better convergence performance as compared to classical meta-heuristic algorithms. The proposed method also demonstrates a comparable performance with several existing works on WBC classification.

Improved bald eagle search algorithm for global optimization and feature selection

The use of metaheuristics is one of the most encouraging methodologies for taking care of real-life problems. Bald eagle search (BES) algorithm is the latest swarm-intelligence metaheuristic algorithm inspired by the intelligent hunting behavior of bald eagles. In recent research works, BES algorithm has performed reasonably well over a wide range of application areas such as chemical engineering, environmental science, physics and astronomy, structural modeling, global optimization, engineering design, energy efficiency, etc. However, it still lacks adequate searching efficiency and has a tendency to stuck in local optima which affects the final outcome. This paper introduces a modified BES (mBES) algorithm that removes the shortcomings of the original BES algorithm by incorporating three improvements; Opposition-based learning (OBL), Chaotic Local Search (CLS), and Transition & Pharsor operators. OBL is embedded in different phases of the standard BES viz. initial population, selecting, searching in space, and swooping phases to update the positions of individual solutions to strengthen exploration, CLS is used to enhance the position of the best agent which will lead to enhancing the positions of all individuals, and Transition & Pharsor operators help to provide sufficient exploration–exploitation trade-off. The efficiency of the mBES algorithm is initially evaluated with 29 CEC2017 and 10 CEC2020 global optimization benchmark functions. In addition, the practicality of the mBES is tested with a real-world feature selection problem and five engineering design problems. Results of the mBES algorithm are compared against a number of classical metaheuristic algorithms using statistical metrics, convergence analysis, box plots, and the Wilcoxon rank sum test. In the case of composite CEC2017 test functions F21-F30, mBES wins against compared algorithms in 70% test cases, whereas for the rest of the test functions, it generates good results in 65% cases. The proposed mBES produces best performance in 55% of the CEC2020 test functions, whereas for the rest of the 45% test cases, it generated competitive results. On the other hand, for five engineering design problems, the mBES is the best among all compared algorithms. In the case of the feature selection problem, the mBES also showed competitiveness with the compared algorithms. Results and observations for all tested optimization problems show the superiority and robustness of the proposed mBES over the baseline metaheuristics. It can be safely concluded that the improvements suggested in the mBES are proved to be effective making it competitive enough to solve a variety of optimization problems.

A point-feature label placement algorithm based on spatial data mining.

The point-feature label placement (PFLP) refers to the process of positioning labels near point features on a map while adhering to specific rules and guidelines, finally obtaining clear, aesthetically pleasing, and conflict-free maps. While various approaches have been suggested for automated point feature placement on maps, few studies have fully considered the spatial distribution characteristics and label correlations of point datasets, resulting in poor label quality in the process of solving the label placement of dense and complex point datasets. In this paper, we propose a point-feature label placement algorithm based on spatial data mining that analyzes the local spatial distribution characteristics and label correlations of point features. The algorithm quantifies the interference among point features by designing a label frequent pattern framework (LFPF) and constructs an ascending label ordering method based on the pattern to reduce interference. Besides, three classical metaheuristic algorithms (simulated annealing algorithm, genetic algorithm, and ant colony algorithm) are applied to the PFLP in combination with the framework to verify the validity of this framework. Additionally, a bit-based grid spatial index is proposed to reduce cache memory and consumption time in conflict detection. The performance of the experiments is tested with 4000, 10000, and 20000 points of POI data obtained randomly under various label densities. The results of these experiments showed that: (1) the proposed method outperformed both the original algorithm and recent literature, with label quality improvements ranging from 3 to 6.7 and from 0.1 to 2.6, respectively. (2) The label efficiency was improved by 58.2% compared with the traditional grid index.

Construction cost prediction system based on Random Forest optimized by the Bird Swarm Algorithm.

Predicting construction costs often involves disadvantages, such as low prediction accuracy, poor promotion value and unfavorable efficiency, owing to the complex composition of construction projects, a large number of personnel, long working periods and high levels of uncertainty. To address these concerns, a prediction index system and a prediction model were developed. First, the factors influencing construction cost were first identified, a prediction index system including 14 secondary indexes was constructed and the methods of obtaining data were presented elaborately. A prediction model based on the Random Forest (RF) algorithm was then constructed. Bird Swarm Algorithm (BSA) was used to optimize RF parameters and thereby avoid the effect of the random selection of RF parameters on prediction accuracy. Finally, the engineering data of a construction company in Xinyu, China were selected as a case study. The case study showed that the maximum relative error of the proposed model was only 1.24%, which met the requirements of engineering practice. For the selected cases, the minimum prediction index system that met the requirement of prediction accuracy included 11 secondary indexes. Compared with classical metaheuristic optimization algorithms (Particle Swarm Optimization, Genetic Algorithms, Tabu Search, Simulated Annealing, Ant Colony Optimization, Differential Evolution and Artificial Fish School), BSA could more quickly determine the optimal combination of calculation parameters, on average. Compared with the classical and latest forecasting methods (Back Propagation Neural Network, Support Vector Machines, Stacked Auto-Encoders and Extreme Learning Machine), the proposed model exhibited higher forecasting accuracy and efficiency. The prediction model proposed in this study could better support the prediction of construction cost, and the prediction results provided a basis for optimizing the cost management of construction projects.

Algorithms comparison for hydrogen storage predictive control on an islanded microgrid

This paper presents a comparison of some classical metaheuristic (Particle Swarm Optimsation, Simulated Anneling) and deterministic (Branch & Bound, Pattern Search) algorithms for the optimization of the storage control of a microgrid over a whole year with real data. These algorithms are compared on the relevance of the controls in terms of energy autonomy power losses and wear of the hydrogen chain as well as on the computational burden for a small embedded controller. This allows to highlight a type of algorithm adapted to a limited time and means of calculation that we find in real conditions. This study also points out the importance of prediction errors and their impacts when controlling real systems.

Comparative study of metaheuristic optimization algorithms for image steganography based on discrete Fourier transform domain

Image steganography and watermarking have demonstrated their efficiency in creating a covert communication channel and in digital image authentication respectively. The importance of cybersecurity tasks sets high efficiency requirements for information embedding methods. Many studies in this area focus on finding new ways to improve the embedding quality. One of them is to use metaheuristic optimization algorithms. The use of classical metaheuristics is common enough for a data hiding area. However, a lot of new metaheuristics have been proposed over the past few years. Their applicability has not yet been evaluated in the field of steganography and digital watermarking. In this paper, we present a comparative study on the efficiency of seven metaheuristic optimization algorithms for finding the best embedding options in the phase spectrum of the Discrete Fourier Transform (DFT). One of the contributions of our research is an improved DFT-based data hiding algorithm, which is a development of the error-free embedding algorithm we obtained earlier. The new algorithm also provides error-free extraction of embedded data, and it is distinguished by the high invisibility of embedding through the use of metaheuristic optimization. The use of metaheuristic optimization led to an increase in the PSNR value by an average of 2%–6% and an increase in the capacity value by an average of 16%–25%. Another important contribution of our research is the original formulation of the optimization problem for different classes of metaheuristics. The experimental results demonstrate the efficiency of metaheuristic optimization for solving data hiding problem. The Differential Evolution and the Particle Swarm Optimization showed the best values of embedding indicators among the classical metaheuristic optimization algorithms. The Gradient-Based Optimizer showed the highest efficiency among modern algorithms. Thus, our research indicates the relevance of further studies of other modern optimization algorithms for a data hiding area.

Hierarchical Harris hawks optimization for epileptic seizure classification

The intelligent recognition of electroencephalogram (EEG) signals is a valuable tool for epileptic seizure classification. Given that visual inspection of EEG signals is time-consuming, and that mutant signals dramatically increase the workload of neurologists, automatic epilepsy diagnosis systems are extremely helpful. However, the existing epilepsy diagnosis methods suffer from some shortcomings. For example, they tend to fall into local optima quickly because of their failure to fully consider the discriminative features of EEG signals. To tackle this problem, in this article, an enhanced automatic epilepsy diagnosis method is proposed using time-frequency analysis and improved Harris hawks optimization (IHHO) with a hierarchical mechanism. Specifically, the signal is decomposed into five rhythms using continuous wavelet transform, with the local and global features extracted using the local binary pattern and the gray level co-occurrence matrix. Discriminative features are then selected and further mapped to the final recognition results using both IHHO and the k-nearest neighbor classifier. To evaluate its performance, the proposed method was compared with a variety of classical meta-heuristic algorithms on 23 benchmark functions. Moreover, the proposed approach achieved more than 99.67% accuracy on the Bonn dataset and 99.06% accuracy on the CHB-MIT dataset, out-performing a multitude of state-of-the-art methods. Taken together, these results demonstrate the utility of our approach in the automatic diagnosis of epilepsy. Supportive datasets and source codes for this research are publicly available at https://github.com/sstudying/lzzhen, and latest updates for the HHO algorithm are provided at https://aliasgharheidari.com/HHO.html.

Solution of Multi-Crew Depots Railway Crew Scheduling Problems: The Chinese High-Speed Railway Case

This paper presents a novel mathematical formulation in crew scheduling, considering real challenges most railway companies face such as roundtrip policy for crew members joining from different crew depots and stricter working time standards under a sustainable development strategy. In China, the crew scheduling is manually compiled by railway companies respectively, and the plan quality varies from person to person. An improved genetic algorithm is proposed to solve this large-scale combinatorial optimization problem. It repairs the infeasible gene fragments to optimize the search scope of the solution space and enhance the efficiency of GA. To investigate the algorithm’s efficiency, a real case study was employed. Results show that the proposed model and algorithm lead to considerable improvement compared to the original planning: (i) Compared with the classical metaheuristic algorithms (GA, PSO, TS), the improved genetic algorithm can reduce the objective value by 4.47%; and (ii) the optimized crew scheduling plan reduces three crew units and increases the average utilization of crew unit working time by 6.20% compared with the original plan.

Improved bat algorithm for roundness error evaluation problem.

In the production and processing of precision shaft-hole class parts, the wear of cutting tools, machine chatter, and insufficient lubrication can lead to changes in their roundness, which in turn affects the overall performance of the relevant products. To improve the accuracy of roundness error assessments, Bat algorithm (BA) is applied to roundness error assessments. An improved bat algorithm (IBA) is proposed to counteract the original lack of variational mechanisms, which can easily lead BA to fall into local extremes and induce premature convergence. First, logistic chaos initialisation is applied to the initial solution generation to enhance the variation mechanism of the population and improve the solution quality; second, a sinusoidal control factor is added to BA to control the nonlinear inertia weights during the iterative process, and the balance between the global search and local search of the algorithm is dynamically adjusted to improve the optimization-seeking accuracy and stability of the algorithm. Finally, the sparrow search algorithm (SSA) is integrated into BA, exploiting the ability of explorer bats to perform a large range search, so that the algorithm can jump out of local extremes and the convergence speed of the algorithm can be improved. The performance of IBA was tested against the classical metaheuristic algorithm on eight benchmark functions, and the results showed that IBA significantly outperformed the other algorithms in terms of solution accuracy, convergence speed, and stability. Simulation and example verification show that IBA can quickly find the centre of a minimum inclusion region when there are many or few sampling points, and the obtained roundness error value is more accurate than that of other algorithms, which verifies the feasibility and effectiveness of IBA in evaluating roundness errors.

Metaheuristics: a comprehensive overview and classification along with bibliometric analysis

Research in metaheuristics for global optimization problems are currently experiencing an overload of wide range of available metaheuristic-based solution approaches. Since the commencement of the first set of classical metaheuristic algorithms namely genetic, particle swarm optimization, ant colony optimization, simulated annealing and tabu search in the early 70s to late 90s, several new advancements have been recorded with an exponential growth in the novel proposals of new generation metaheuristic algorithms. Because these algorithms are neither entirely judged based on their performance values nor according to the useful insight they may provide, but rather the attention is given to the novelty of the processes they purportedly models, these area of study will continue to periodically see the arrival of several new similar techniques in the future. However, there is an obvious reason to keep track of the progressions of these algorithms by collating their general algorithmic profiles in terms of design inspirational source, classification based on swarm or evolutionary search concept, existing variation from the original design, and application areas. In this paper, we present a relatively new taxonomic classification list of both classical and new generation sets of metaheuristic algorithms available in the literature, with the aim of providing an easily accessible collection of popular optimization tools for the global optimization research community who are at the forefront in utilizing these tools for solving complex and difficult real-world problems. Furthermore, we also examined the bibliometric analysis of this field of metaheuristic for the last 30 years.

Orthogonally-designed adapted grasshopper optimization: A comprehensive analysis

Grasshopper optimization algorithm (GOA) is a newly proposed meta-heuristic algorithm that simulates the biological habits of grasshopper seeking for food sources. Nonetheless, some shortcomings exist in the basic version of GOA. It may quickly drop into local optima and show slow convergence rates when facing some complex basins. In this work, an improved GOA is proposed to alleviate the core shortcomings of GOA and handle continuous optimization problems more efficiently. For this purpose, two strategies, including orthogonal learning and chaotic exploitation, are introduced into the conventional GOA to find a more stable trade-off between the exploration and exploitation cores. Adding orthogonal learning to GOA can enhance the diversity of agents, whereas a chaotic exploitation strategy can update the position of grasshoppers within a limited local region. To confirm the efficacy of GOA, we compared it with a variety of famous classical meta-heuristic algorithms performed on 30 IEEE CEC2017 benchmark functions. Also, it is applied to feature selection cases, and three structural design problems are employed to validate its efficacy in terms of different metrics. The experimental results illustrate that the above tactics can mitigate the deficiencies of GOA, and the improved variant can reach high-quality solutions for different problems.

A Comparative Study of Four Metaheuristic Algorithms, AMOSA, MOABC, MSPSO, and NSGA-II for Evacuation Planning

Evacuation planning is an important activity in disaster management to reduce the effects of disasters on urban communities. It is regarded as a multi-objective optimization problem that involves conflicting spatial objectives and constraints in a decision-making process. Such problems are difficult to solve by traditional methods. However, metaheuristics methods have been shown to be proper solutions. Well-known classical metaheuristic algorithms—such as simulated annealing (SA), artificial bee colony (ABC), standard particle swarm optimization (SPSO), genetic algorithm (GA), and multi-objective versions of them—have been used in the spatial optimization domain. However, few types of research have applied these classical methods, and their performance has not always been well evaluated, specifically not on evacuation planning problems. This research applies the multi-objective versions of four classical metaheuristic algorithms (AMOSA, MOABC, NSGA-II, and MSPSO) on an urban evacuation problem in Rwanda in order to compare the performances of the four algorithms. The performances of the algorithms have been evaluated based on the effectiveness, efficiency, repeatability, and computational time of each algorithm. The results showed that in terms of effectiveness, AMOSA and MOABC achieve good quality solutions that satisfy the objective functions. NSGA-II and MSPSO showed third and fourth-best effectiveness. For efficiency, NSGA-II is the fastest algorithm in terms of execution time and convergence speed followed by AMOSA, MOABC, and MSPSO. AMOSA, MOABC, and MSPSO showed a high level of repeatability compared to NSGA-II. It seems that by modifying MOABC and increasing its effectiveness, it could be a proper algorithm for evacuation planning.