Article
Computer Science, Information Systems
Kai Zhang, Chaonan Shen, Juanjuan He, Gary G. Yen
Summary: The proposed MMO-EvoKnee algorithm incorporates MCDM strategy to efficiently search for a complete set of global knee solutions for MMOPs. It outperforms existing state-of-the-art MMOEAs and provides decision makers with well-converged alternative solutions.
INFORMATION SCIENCES
(2021)
Article
Computer Science, Artificial Intelligence
Joao Luiz Junho Pereira, Guilherme Antonio Oliver, Matheus Brendon Francisco, Sebastiao Simoes Cunha, Guilherme Ferreira Gomes
Summary: The Multi-objective Lichtenberg Algorithm is a hybrid meta-heuristic algorithm capable of dealing with multiple objectives, distributing points for evaluation through Lichtenberg patterns in each iteration. It has shown promising results in terms of convergence and maximum spread, outperforming traditional and recent algorithms.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Multidisciplinary Sciences
Tong Wu, Jing Li, Xuan Qin
Summary: In order to enhance braking performance, engineers improve braking systems by upgrading structures and optimizing parameters, with multi-objective optimal design of electro-mechanical brake (EMB) parameters being an effective method. Research results show that this optimal design can reduce braking pressure response time, shorten stopping distance, increase mean fully developed deceleration, and reduce lateral displacement of the body.
Article
Engineering, Environmental
Jiahao Wang, Xiaomin Liu, Yue Wang
Summary: This study proposes an improved multi-objective density-based topology optimization model for Newtonian and non-Newtonian fluid micro-channel reactors (NMCR, nNMCR) to enhance their overall performance. A multi-physics model of continuous-flow exothermic reaction is built to describe flow, catalytic reaction, and heat transfer in the reactors. The improved model uses material density as the design variable to control changes in fluid and solid domains. Various techniques, such as polynomial function, Helmholtz partial differential filter, and hyperbolic tangent projection, are employed to eliminate checkerboard phenomenon and structural ill-conditioning. An improved multi-objective algorithm based on weighted-sum method and pseudo-design domain concept is proposed to avoid local topological structure simplification and irregularities. The effects of reactor geometries and outlet angles on optimal configurations and performance parameters are investigated. The study also provides an efficient optimization method for designing micro-channel reactors with higher comprehensive performance.
CHEMICAL ENGINEERING JOURNAL
(2023)
Review
Construction & Building Technology
Wang Chen, Mulian Zheng
Summary: This review provides an overview of the state-of-the-art development and application of multi objective optimization methods in pavement maintenance and rehabilitation decision-making, including strategic goals, decision processes, uncertainty handling methods, and intelligent algorithm applications globally. The importance of establishing a unified multi-objective optimization model for intelligent decision-making is emphasized, and existing deficiencies and future research directions are discussed.
AUTOMATION IN CONSTRUCTION
(2021)
Article
Construction & Building Technology
A. U. Weerasuriya, Xuelin Zhang, Jiayao Wang, Bin Lu, K. T. Tse, Chun-Ho Liu
Summary: This study evaluated the performance of four optimization algorithms and three decision-making techniques in optimizing an unconventional building design. The algorithms showed improvement with increasing population and iterations, with PSO performing well in many aspects. The decision-making techniques tended to choose similar optimal designs, and the multi-objective optimization resulted in a better trade-off solution compared to single-objective optimization.
BUILDING AND ENVIRONMENT
(2021)
Review
Computer Science, Interdisciplinary Applications
Natee Panagant, Nantiwat Pholdee, Sujin Bureerat, Ali Riza Yildiz, Seyedali Mirjalili
Summary: This study compares the performance of 14 new and established multi-objective metaheuristics in solving truss optimization problems, providing insights into the pros and cons of these algorithms and aiding in designing customized algorithms for such problems.
ARCHIVES OF COMPUTATIONAL METHODS IN ENGINEERING
(2021)
Article
Engineering, Multidisciplinary
Mostafa Borhani
Summary: The study analyzes the air transport network structure using a multi-objective genetic algorithm to reduce air routes, improve passenger travel length, and minimize the number of stops per passenger. By combining point-to-point and hub-and-spoke topologies, the model successfully decreased air routes while increasing average travel length and route changes. The optimization model proved effective in improving airline topologies and reducing operational costs in the Iran airline industry.
AIN SHAMS ENGINEERING JOURNAL
(2021)
Article
Computer Science, Information Systems
Tomoaki Takagi, Keiki Takadama, Hiroyuki Sato
Summary: This work introduces the concepts of directional and estimated directional Pareto front to encourage multi-objective decision-making when the Pareto front exists in limited regions. The directional Pareto front is a superset of the Pareto front and supplements the objective value trade-off. The estimated directional Pareto front represents the directional Pareto front using a limited number of points, enhancing the understanding of the objective space.
Article
Thermodynamics
T. A. Z. de Souza, J. L. J. Pereira, M. B. Francisco, C. A. R. Sotomonte, B. Jun Ma, G. F. Gomes, C. J. R. Coronado
Summary: This study provides a detailed analysis of hydrogen production using Response Surface Methodology and Lichtenberg Algorithm, aiming to quickly optimize steam reforming cycles considering different feedstock compositions and other characteristics. Comparison with other optimization studies demonstrates that this new methodology offers a quick and consistent method for optimizing steam reforming and potentially other thermodynamic cycles.
INTERNATIONAL JOURNAL OF GREEN ENERGY
(2023)
Article
Business
Qun Wu, Xinwang Liu, Jindong Qin, Ligang Zhou, Abbas Mardani, Muhammet Deveci
Summary: This paper aims to develop a hybrid SRI portfolio selection model using multi-criteria decision making and multi-objective optimization techniques. A case study on medical stock investment is conducted to demonstrate the robustness, effectiveness, and superiority of the proposed methodology.
TECHNOLOGICAL FORECASTING AND SOCIAL CHANGE
(2022)
Article
Computer Science, Artificial Intelligence
Kai Zhang, Minshi Chen, Xin Xu, Gary G. Yen
Summary: The paper proposes an evolution strategy MMO-MOES for solving multimodal multi-objective optimization problems, focusing on searching for multiple groups of optimal solutions in decision space. By using a novel niching strategy and requiring a small population size, MMO-MOES is effective in finding well-distributed and well-converged Pareto optimal solutions. Experimental results show exceptional performance compared to leading-edge MMOEAs in various test problems.
APPLIED SOFT COMPUTING
(2021)
Article
Thermodynamics
Hamed Kamali, Mehdi Mehrpooya, Seyed Hamed Mousavi, Mohammad Reza Ganjali
Summary: This work presents a thermally regenerative electrochemical refrigerator model based on finite-time analysis. The system is analyzed in different temperature ranges, and sensitivity analysis is performed. A multi-objective genetic algorithm is used to optimize the system parameters, and the optimum values for cooling capacity and coefficient of performance are obtained. This study is significant for the lab-scale design of thermally regenerative electrochemical refrigerators.
ENERGY CONVERSION AND MANAGEMENT
(2022)
Article
Computer Science, Artificial Intelligence
Ke-Jing Du, Jian-Yu Li, Hua Wang, Jun Zhang
Summary: Evolutionary multi-objective multi-task optimization is an emerging paradigm for solving multi-objective multi-task optimization problems using evolutionary computation. This paper proposes treating these problems as multi-objective multi-criteria optimization problems and develops an algorithm framework that utilizes the knowledge of all tasks in the same population. The algorithm selects fitness evaluation functions as criteria, guided by a probability-based selection strategy and an adaptive parameter learning method. Extensive experiments show the effectiveness and efficiency of the proposed algorithm. Treating MO-MTOP as MO-MCOP is a potential and promising direction for solving these problems.
COMPLEX & INTELLIGENT SYSTEMS
(2023)
Article
Transportation Science & Technology
Xiangkun He, Chen Lv
Summary: This article presents a novel constrained multi-objective reinforcement learning technique for personalized decision making in autonomous driving. By introducing a non-linear constraint and a vectorized action-value function, this method is able to learn decision behaviors that align efficiently between user preferences and optimal policies.
TRANSPORTATION RESEARCH PART C-EMERGING TECHNOLOGIES
(2023)
Article
Thermodynamics
T. A. Z. de Souza, J. L. J. Pereira, M. B. Francisco, C. A. R. Sotomonte, B. Jun Ma, G. F. Gomes, C. J. R. Coronado
Summary: This study provides a detailed analysis of hydrogen production using Response Surface Methodology and Lichtenberg Algorithm, aiming to quickly optimize steam reforming cycles considering different feedstock compositions and other characteristics. Comparison with other optimization studies demonstrates that this new methodology offers a quick and consistent method for optimizing steam reforming and potentially other thermodynamic cycles.
INTERNATIONAL JOURNAL OF GREEN ENERGY
(2023)
Article
Mechanics
Daniel Brighenti Bortoluzzi, Camila Aparecida Diniz, Joao Luiz Junho Pereira, Ronny Francis Ribeiro Junior, Guilherme Ferreira Gomes, Antonio Carlos Ancelotti Junior
Summary: This study is the first to use Artificial Neural Networks (ANN) to predict and assess the mode II interlaminar fracture toughness in z-pinned composites. Z-pins with different rectangular shapes and pin area densities are manufactured using design variables generated from Design of Experiments (DOE). The experimental results show an improvement in the mode II delamination resistance of z-pinned specimens compared to unpinned ones. Response Surface Model (RSM) and Analysis of Variance (ANOVA) are used for statistical analysis, and the generated data are used to train an Artificial Neural Network to predict and evaluate the influence of pin size and density configuration on interlaminar fracture properties of z-pinned composite structures.
COMPOSITE STRUCTURES
(2023)
Article
Automation & Control Systems
Danilo Pazeto, Joao Luiz Junho Pereira, Guilherme Ferreira Gomes
Summary: This study conducted numerical simulation and analysis of aluminum extrusion using the finite element method and metamodel optimization.
INTERNATIONAL JOURNAL OF ADVANCED MANUFACTURING TECHNOLOGY
(2023)
Article
Computer Science, Artificial Intelligence
Benedict Jun Ma, Joao Luiz Junho Pereira, Diego Oliva, Shuai Liu, Yong-Hong Kuo
Summary: This article introduces an enhanced manta ray foraging optimizer based on oppositional learning and vertical crossover search for color image segmentation. The proposed algorithm, OL-MRFO-VC, is integrated with Kapur entropy to identify the best threshold configuration in each image component. The technique is tested over three datasets and compared with fourteen competitive metaheuristics.
KNOWLEDGE-BASED SYSTEMS
(2023)
Article
Engineering, Civil
Matheus Brendon Francisco, Joao Luiz Junho Pereira, Sebastiao Simoes da Cunha, Guilherme Ferreira Gomes
Summary: This paper focuses on the multi-objective optimization of a tubular sandwich structure with an auxetic reentrant core. The Response Surface Methodology and the Multi-objective Lichtenberg Algorithm were used to find the optimized configuration of the structure. A parametric analysis was also conducted to study the influence of design factors on different responses. The results demonstrated the significant effect of unit cell height on failure load, natural frequency, and mass, while revealing the lack of correlation between unit cell height and Poisson's ratio.
ENGINEERING STRUCTURES
(2023)
Article
Computer Science, Artificial Intelligence
Joao Luiz Junho Pereira, Benedict Jun Ma, Matheus Brendon Francisco, Ronny Francis Ribeiro Jr, Guilherme Ferreira Gomes
Summary: Feature selection is a valuable tool in understanding problems in data mining, improving patterns and reducing computational costs. This study introduces a new metaheuristic called binary sunflower optimization (BSFO), which shows promising results in terms of fitness value and computational costs. The improved version, IBSFO, is compared with eight other metaheuristics and performs better in terms of fitness value and execution time.
Article
Materials Science, Characterization & Testing
Lucas Antonio de Oliveira, Guilherme Ferreira Gomes, Joao Luiz Junho Pereira, Matheus Brendon Francisco, Anthonin Demarbaix, Sebastiao Simoes Cunha Jr
Summary: Infrared thermography technique is used to detect damage in metallic and non-metallic materials, and it helps in determining the safety of mechanical structures. Vibrothermography combines vibration and infrared thermography to identify internal damages through temperature mapping. The current study focuses on identifying different research approaches for damage identification and characterization, evaluating heat generation mechanisms in damaged regions, and utilizing mathematical methods to enhance the efficiency of damage detection and characterization.
JOURNAL OF NONDESTRUCTIVE EVALUATION
(2023)
Article
Engineering, Mechanical
Ronny Francis Ribeiro Junior, Fabricio Alves de Almeida, Ariosto Bretanha Jorge, Joao Luiz Junho Pereira, Matheus Brendon Francisco, Guilherme Ferreira Gomes
Summary: Fault diagnosis is crucial for maintenance industries to prevent catastrophic failures and save time and money. This paper proposes a model using uniaxial acceleration signals to cluster, identify, and diagnose six different failures in electric motors. Experiment results demonstrate the efficiency of the proposed method, with an average accuracy of 97.9%, especially in identifying bearing, unbalanced, and mechanical loss failures. The method can be used for early detection of fault conditions based on real electric motor experiments.
JOURNAL OF THE BRAZILIAN SOCIETY OF MECHANICAL SCIENCES AND ENGINEERING
(2023)
Article
Materials Science, Multidisciplinary
Guilherme Antonio Oliver, Joao Luiz Junho Pereira, Matheus Brendon Francisco, Guilherme Ferreira Gomes
Summary: This study proposes a damage index for identifying delaminations in laminated composite structures using a discrete wavelet transform. The index achieved high-quality results in identifying damage in both numerical cases and real carbon fiber-reinforced polymer beams. The method proved effective at locating damage in almost all positions along the beam, with some issues at the free end due to signal discontinuity.
MECHANICS OF ADVANCED MATERIALS AND STRUCTURES
(2023)
Article
Mechanics
Guilherme Ferreira Gomes, Ronny Francis Ribeiro Ribeiro Junior, Joao Luiz Junho Pereira, Matheus Brendon Francisco
Summary: In this paper, a deep learning model is trained to predict 16 different structural responses of a complex composite isogrid structure tube. The results demonstrate the substantial capacity of the deep learning model to fit the isogrid physical behavior and predict the structural responses. Additionally, the model is able to predict the design parameters of the isogrid structure from desired results.
COMPOSITE STRUCTURES
(2023)
Article
Computer Science, Artificial Intelligence
Joao Luiz Junho Pereira, Kate Smith-Miles, Mario Andres Munoz, Ana Carolina Lorena
Summary: This paper investigates the importance of gathering a diverse benchmark of datasets and introduces an optimization method to choose challenging datasets. Experimental results demonstrate that this method is more effective in selecting diverse benchmarks and challenging ML algorithms.
DATA MINING AND KNOWLEDGE DISCOVERY
(2023)
Article
Computer Science, Interdisciplinary Applications
Matheus Francisco, Joao Pereira, Lucas Oliveira, Sebastiao Simoes Cunha Jr, G. F. Gomes
Summary: The purpose of this study is to perform multi-objective optimization of a reentrant hexagonal cell auxetic structure. It also aims to analyze the impact of design factors on each response through parametric analysis.
ENGINEERING COMPUTATIONS
(2023)
Article
Engineering, Civil
Camila Aparecida Diniz, Joao Luiz Junho Pereira, Daniel Brighenti Bortoluzzi, Sebastia Simoes Cunha Jr, Guilherme Ferreira Gomes
Summary: A comparative study was conducted on the static and dynamic behavior of a new type of tubular structure with drop-offs used in lower limb prostheses. Numerical and experimental methods were used to obtain results on natural frequencies, damping loss factors, and maximum compression load. The hybrid structure with drop-offs was found to be the best cost-effective option for pylon tubes.
ENGINEERING STRUCTURES
(2023)
Article
Computer Science, Artificial Intelligence
Joao Luiz Junho Pereira, Matheus Brendon Francisco, Fabricio Alves de Almeida, Benedict Jun Ma, Sebastiao Simoes Cunha, Guilherme Ferreira Gomes
Summary: This paper presents different strategies for tuning and accelerating meta-heuristics using the first hybrid algorithm in the literature, which is inspired by lightning and Lichtenberg figures. After discussing the best tuning tools, the algorithm's parameters are tuned using response surface methodology. The chaotic Lichtenberg algorithm equipped with the piecewise function and tuned parameters proves to be the best version with only 16% similarity to the original algorithm, outperforming other popular algorithms in terms of average accuracy and computational cost.