Article
Computer Science, Artificial Intelligence
Jose M. Granado-Criado, Alvaro Rubio-Largo, Sergio Santander-Jimenez, Miguel A. Vega-Rodriguez
Summary: Research supports the relationship between Single Nucleotide Polymorphisms (SNPs) and neurodegenerative diseases. This paper proposes the application of two successful multiobjective evolutionary algorithms to identify genetic interactions. The results demonstrate the advantages of the NSGA-III algorithm in both multiobjective and biological terms, and it reveals new genetic interactions.
APPLIED SOFT COMPUTING
(2022)
Article
Operations Research & Management Science
N. Eslami, B. Najafi, S. M. Vaezpour
Summary: We extend and analyze the trust region method for solving smooth and unconstrained multicriteria optimization problems on Riemannian manifolds. A quadratic model is assigned to each component of the vectorial objective function at each iteration by considering the notion of retractions. A subproblem is constructed and solved to find a new descent direction. The convergence of the algorithm is investigated by considering radially Lipschitz continuously differentiable functions.
JOURNAL OF OPTIMIZATION THEORY AND APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Jie Cao, Jianlin Zhang, Fuqing Zhao, Zuohan Chen
Summary: A novel algorithm named MOEA/D-TS is proposed in this paper, which effectively solves multi-objective optimization problems through two-stage evolution strategies. The performance of the algorithm is validated in real world problems and shows advantages in terms of convergence and diversity over other algorithms.
EXPERT SYSTEMS WITH APPLICATIONS
(2021)
Article
Computer Science, Artificial Intelligence
Ying Xu, Huan Zhang, Lei Huang, Rong Qu, Yusuke Nojima
Summary: This research investigates the grid-based decomposition methods in multi-objective optimization to address the issues of diversity and convergence. A new concept of Pareto Front grid and a statistical analysis-based nadir point estimation strategy are proposed to improve computational efficiency. Furthermore, a novel grid-based knee point selection method is proposed. Experimental analysis demonstrates the effectiveness of the proposed PFG-MOEA algorithm.
APPLIED SOFT COMPUTING
(2023)
Article
Operations Research & Management Science
Shahabeddin Najafi, Masoud Hajarian
Summary: This paper introduces a Riemannian BFGS method for addressing multiobjective optimization problems with strongly retraction-convex objective functions. The method is an extension of the Euclidean version and converges to a Pareto optimal point regardless of the initial point. The main component of the globalization strategy is a generalized Wolfe line search. Numerical experiments demonstrate the superiority and effectiveness of the proposed algorithm.
COMPUTATIONAL OPTIMIZATION AND APPLICATIONS
(2023)
Article
Engineering, Chemical
Hongbo Jiao, Huaibin Wei, Qi Yang, Min Li
Summary: The CFD-MOEA/D algorithm is proposed for flood control operation of large-scale reservoirs, showing better performance than traditional algorithms. By using a decomposition method, it obtains non-dominated solutions with higher water levels and outperforms the NSGA-II algorithm. The optimal dispatching scheme of the algorithm matches the actual reservoir dispatching, improving scheduling efficiency.
Article
Operations Research & Management Science
Shahabeddin Najafi, Masoud Hajarian
Summary: In this paper, we introduce the conjugate gradient multiobjective optimization methods on Riemannian manifolds. We redefine the concepts of optimality and Wolfe conditions, as well as Zoutendijk's theorem, in this context. We prove that under some standard assumptions, a sequence generated by these algorithms converges to a critical Pareto point, defined when the step sizes satisfy the multiobjective Wolfe conditions. We propose the Fletcher-Reeves, Dai-Yuan, Polak-Ribiere-Polyak, and Hestenes-Stiefel parameters and further analyze the convergence behavior of the first two methods and compare their performance to the steepest descent method.
JOURNAL OF OPTIMIZATION THEORY AND APPLICATIONS
(2023)
Article
Operations Research & Management Science
Ellen H. Fukuda, L. M. Grana Drummond, Ariane M. Masuda
Summary: The proposed extension of the real-valued conjugate directions method is used for unconstrained quadratic multiobjective problems, aiming to find weak Pareto and Pareto optima through specific steps and calculations in each iteration.
Article
Computer Science, Software Engineering
Matteo Lapucci, Pierluigi Mansueto, Fabio Schoen
Summary: This paper considers the solution methods for multi-objective optimization problems over a box. It compares the advantages and disadvantages of evolutionary methods and descent methods through numerical experiments, and proposes a new method that combines the strengths of both. The resulting algorithm, called Non-dominated Sorting Memetic Algorithm, performs excellently in numerical tests on widely used test functions.
MATHEMATICAL PROGRAMMING COMPUTATION
(2023)
Article
Engineering, Chemical
Shashwat Srivastava, Nitin Padhiyar
Summary: In this study, a comparative analysis between a fixed bed reactor (FBR) and a reverse flow chromatographic reactor (RFCR) for a series reaction is carried out using mathematical model based single and multi-objective optimization. The uncertainty in the model parameters is considered and the optimization problem considers the inlet concentration, Damkohler number, and dimensionless switching time as decision variables. The results show the superiority of RFCR over FBR, with a representative Pareto point solution of 3-objective problem in RFCR corresponding to 51.36% higher yield, 27.06% higher selectivity, and 19.15% higher conversion compared to FBR.
CHEMICAL ENGINEERING RESEARCH & DESIGN
(2023)
Article
Computer Science, Artificial Intelligence
Qian Wang, Qinghua Gu, Lu Chen, Yueping Guo, Naixue Xiong
Summary: This paper proposes an improved multi-objective evolutionary algorithm with global and local cooperative mechanisms for complicated bi-objective optimization problems. The overall optimization is carried out through the coordination of global and local search phases. The proposed algorithm shows great potential for addressing complex bi-objective optimization problems.
APPLIED SOFT COMPUTING
(2023)
Article
Computer Science, Information Systems
Mingjing Wang, Xiaoping Li, Yong Dai, Long Chen, Huiling Chen, Ruben Ruiz
Summary: Researchers have developed a method called Copula Incremental Learning (CIL) to improve the performance of the MOEA/D algorithm in problems with irregular Pareto Fronts (PFs) by generating non-uniform direction vectors. They also employ the Niche Hierarchical Selection (NHS) method to construct the neighborhood structure and prevent duplicate solutions. The use of convergence-guided direction (CGD) ensures efficiency by approximating irregular PFs. Statistical analysis shows that this method outperforms other competitive algorithms, particularly in handling multi-objective optimization problems with irregular PFs.
INFORMATION SCIENCES
(2023)
Article
Energy & Fuels
Seong-Tae Jo, Woo-Hyeon Kim, Young-Keun Lee, Yong-Joo Kim, Jang-Young Choi
Summary: In this study, a multi-objective optimal design method for the SPMSM of an EV air conditioner system compressor was proposed and applied using NSGA-II and an analytical method. The validity of the proposed method was confirmed by comparing the characteristics of the optimal design model with those of the initially designed model.
Article
Computer Science, Information Systems
Jinlong Zhou, Juan Zou, Shengxiang Yang, Jinhua Zheng, Dunwei Gong, Tingrui Pei
Summary: This paper proposes niche-based and angle-based selection strategies for many objective evolutionary optimization, which have been shown to be competitive and scalable to handle constrained many-objective optimization problems in experimental studies.
INFORMATION SCIENCES
(2021)
Article
Computer Science, Artificial Intelligence
Fei Liu, Qingfu Zhang, Zhonghua Han
Summary: This paper studies the method of expensive multiobjective optimization when gradients are available. We propose a method, called MOEA/D-GEK, which combines MOEA/D and gradient-enhanced kriging. Experimental results demonstrate the high efficiency and effectiveness of our proposed method in a set of test instances and an engineering problem of aerodynamic design optimization for a transonic airfoil.
Article
Computer Science, Artificial Intelligence
Bekir Afsar, Ana B. Ruiz, Kaisa Miettinen
Summary: This paper highlights the importance of solving multiobjective optimization problems with interactive methods and comparing different methods to find the most suitable one. It introduces a new artificial decision maker (ADM-II) that can handle different types of preference information and assess the performance of interactive evolutionary methods. By considering the learning and decision phases separately, ADM-II can generate preference information in different ways to reflect the nature of each phase.
COMPLEX & INTELLIGENT SYSTEMS
(2023)
Article
Engineering, Multidisciplinary
Bhuvan Khoshoo, Julian Blank, Thang Q. Pham, Kalyanmoy Deb, Shanelle N. Foster
Summary: This article investigates a complex electric machine design problem and proposes a computationally efficient optimization method based on evolutionary algorithms. The method generates feasible solutions using a repair operator and addresses time-consuming objective functions by incorporating surrogate models. The study successfully establishes the superiority of the proposed method in optimization tasks.
ENGINEERING OPTIMIZATION
(2023)
Article
Management
Bekir Afsar, Johanna Silvennoinen, Giovanni Misitano, Francisco Ruiz, Ana B. Ruiz, Kaisa Miettinen
Summary: Interactive multiobjective optimization methods operate iteratively, allowing decision makers to provide preference information and generate desired solutions. Different methods vary in technical aspects and preference information used, making it challenging to select the most suitable method. Published research lacks specific information on conducted experiments, impeding replication. We propose a novel questionnaire and experimental design for comparing methods and develop user interfaces for a sustainability problem with multiple objectives.
JOURNAL OF THE OPERATIONAL RESEARCH SOCIETY
(2023)
Article
Automation & Control Systems
Bhupinder Singh Saini, Debalay Chakrabarti, Nirupam Chakraborti, Babooshka Shavazipour, Kaisa Miettinen
Summary: This paper tackles the challenges of solving real-life data-driven multiobjective optimization problems, involving preprocessing, modelling, formulation, and decision support. It focuses on optimizing the composition of microalloyed steels to achieve desired mechanical properties. The proposed MultiDM/IOPIS algorithm combines multiobjective evolutionary algorithms and scalarization functions to enable meaningful decision-making for multiple objective functions and decision makers. Through the use of the DESDEO framework, the methodology successfully provides microalloyed steel compositions that satisfy both decision makers.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2023)
Article
Computer Science, Artificial Intelligence
Kalyanmoy Deb, Zhichao Lu, Ian Kropp, J. Sebastian Hernandez-Suarez, Rayan Hussein, Steven Miller, A. Pouyan Nejadhashemi
Summary: Many societal and industrial problems can be decomposed into hierarchical subproblems. This article introduces a new evolutionary approach that allows upper level decision makers to analyze the impact of lower level decision making when choosing a solution. This method can be applied to similar hierarchical management problems to achieve minimum deviation and more reliable outcomes.
IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION
(2023)
Article
Computer Science, Artificial Intelligence
Adhe Kania, Bekir Afsar, Kaisa Miettinen, Juha Sipila
Summary: We propose DESMILS, a decision support approach that tackles multi-item lot sizing problems with a large number of items using single-item multiobjective lot sizing models. DESMILS considers multiple conflicting objective functions and incorporates decision maker preferences to find the most preferred Pareto optimal solutions. Through clustering, DESMILS treats items in a cluster utilizing preferences provided for a representative item. This approach reduces the decision maker's workload and time while still achieving acceptable solutions.
JOURNAL OF INTELLIGENT MANUFACTURING
(2023)
Article
Cardiac & Cardiovascular Systems
Arto J. Hautala, Babooshka Shavazipour, Bekir Afsar, Mikko P. Tulppo, Kaisa Miettinen
Summary: This study evaluated the applicability of machine learning tools for predicting healthcare costs in patients with acute coronary syndrome based on known risk markers, and found that depression score is the most significant predictor of healthcare costs.
CARDIOVASCULAR DIGITAL HEALTH JOURNAL
(2023)
Article
Engineering, Multidisciplinary
Jana Burkotova, Pouya Aghaei Pour, Tomas Kratky, Kaisa Miettinen
Summary: This article introduces a surrogate-assisted evolutionary interactive multiobjective optimization method applied to pump stator design. The preferences of a decision maker are iteratively incorporated into the solution process, demonstrating the advantages of the interactive method in reducing computation time and finding preferred solutions. The decision maker expressed satisfaction with the interactive solution process, and the final solution accurately reflected his preferences. Importantly, this method could save days of computation time.
ENGINEERING OPTIMIZATION
(2023)
Article
Computer Science, Artificial Intelligence
Dhish Kumar Saxena, Sukrit Mittal, Sarang Kapoor, Kalyanmoy Deb
Summary: This article proposes a high-fidelity-dominance principle that factors in all three critical human decision-making elements and implements it in a computationally efficient many-objective evolutionary algorithm (MaOEA). The experimental results show statistically better performance in about 60% of instances, making it practical and worthy of further investigation and application.
IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION
(2023)
Proceedings Paper
Computer Science, Artificial Intelligence
Deepanshu Yadav, Palaniappan Ramu, Kalyanmoy Deb
Summary: Evolutionary multi-objective optimization (EMO) algorithms are commonly used to solve multi- and many-objective optimization problems and find the Pareto front. It is important for decision makers to consider objective vectors that are less sensitive to perturbations in design variables and problem parameters. This paper proposes and evaluates different algorithmic implementations that integrate multi-objective optimization, robustness consideration, and multi-criterion decision-making. The results provide insights for developing more efficient multi-objective robust optimization and decision-making procedures for practical problems with uncertainties.
PROCEEDINGS OF THE 2023 GENETIC AND EVOLUTIONARY COMPUTATION CONFERENCE, GECCO 2023
(2023)
Proceedings Paper
Computer Science, Artificial Intelligence
Ritam Guha, Wei Ao, Stephen Kelly, Vishnu Boddeti, Erik Goodman, Wolfgang Banzhaf, Kalyanmoy Deb
Summary: Automated machine learning (AutoML) greatly simplifies architecture engineering by building machine-learning algorithms using basic primitives. AutoML-Zero expands on this concept by exploring novel architectures beyond human knowledge without utilizing feature or architectural engineering. However, it currently lacks a mechanism to satisfy real-world application constraints. We propose MOAZ, a multi-objective variant of AutoML-Zero, which trades off accuracy with computational complexity, distributes solutions on a Pareto front, and efficiently explores the search space.
PROCEEDINGS OF THE 2023 GENETIC AND EVOLUTIONARY COMPUTATION CONFERENCE, GECCO 2023
(2023)
Article
Computer Science, Artificial Intelligence
Deepanshu Yadav, Palaniappan Ramu, Kalyanomy Deb
Summary: This paper proposes an approach that combines the Pareto-Race MCDM method with the interpretable self-organizing map (iSOM) based visualization method. The approach assists decision makers in multi-criteria decision-making by generating iSOM plots of objectives and considering metrics such as closeness to constraint boundaries, trade-off value, and robustness. The proposed iSOM-enabled Pareto-Race approach improves the quality of preferred solutions.
APPLIED SOFT COMPUTING
(2023)
Article
Mathematics, Interdisciplinary Applications
Kalyanmoy Deb, Matthias Ehrgott
Summary: This paper analyzes the properties of generalized dominance structures and introduces the concept of anti-dominance structure to explain the identification of resulting optimal solutions. The anti-dominance structure is applied to analyze the optimal solutions of commonly used dominance structures.
MATHEMATICAL AND COMPUTATIONAL APPLICATIONS
(2023)
Proceedings Paper
Computer Science, Artificial Intelligence
Pouya Aghaei Pour, Sunith Bandaru, Bekir Afsar, Kaisa Miettinen
Summary: This article introduces the importance of interactive methods in multiobjective optimization problems and points out the challenges in choosing the appropriate interactive method and comparing them using indicators. It proposes a set of desirable properties of indicators for assessing interactive methods in order to fill a gap in the literature.
PROCEEDINGS OF THE 2022 GENETIC AND EVOLUTIONARY COMPUTATION CONFERENCE COMPANION, GECCO 2022
(2022)
Article
Forestry
Babooshka Shavazipour, Dmitry Podkopaev, Kaisa Miettinen
Summary: Sustainable environmental management involves uncertainties that cannot be addressed using probabilistic models. This paper proposes a multi-scenario multi-objective approach to support decision-making in forest landscape planning, helping experts find robust strategies.
CANADIAN JOURNAL OF FOREST RESEARCH
(2022)