Pareto Ranking Research Articles

An integrated framework for measuring sustainable rural development towards the SDGs

Rural areas are essential for achieving the Sustainable Development Goals (SDGs), necessitating the evaluation of rural sustainability transitions. A precondition for evaluating sustainable rural development is to adapt the SDGs to rural contexts while maintaining global comparability. Besides, tradeoff relationships among evaluation indicators should be captured to overcome the limitations of traditional weighting aggregation methods. In this study, we propose an integrated framework that incorporates the SDGs into a composite indicator system for assessing sustainable rural development. Our framework quantifies these indicators using the index thresholds outlined in the SDGs dashboard, along with customized rules that reflectrural contexts. Furthermore, we employ the Pareto ranking approach to evaluate the sustainable development potentials of rural areas regarding tradeoff relationships among different indicators. In a case study of Zhaoyuan City, China, our findings indicate that only 20 % of the sustainable development goals are achieved on average across all indicator dimensions. The economic and social dimensions are closer to their ideal targets and more compatible with each other, while the ecological dimension remains relatively unsustainable. Sustainable rural development depends primarily on natural resource endowment, ecological sustainability, and residential sustainability of villages. Understanding tradeoff relationships among different indicators enables us to better adapt rural policies to the specific contexts of villages.

Investigation of sustainable operation oriented- economic, process and environment based multi-criteria optimization of large scale methanol production plant

Methanol is one of the main raw materials and a building block for many essential chemicals in the industry, and it is widely produced from the synthesis gas. It is used as a fuel, solvent, inhibitor, antifreeze agent, and in producing a variety of chemicals including olefins and paraffin. The global demand for methanol is 70 million tons per year and it is estimated to grow by a factor of 6–8% per year so it is necessary to develop a methanol production process, which is economical, and environment friendly. The multi-objective optimization (MOO) approach is used to optimize the large scale Lurgi two-step methanol synthesis process using the non-dominated sorting genetic algorithm-II (NSGA-II) to find the Pareto front solutions involving the process, economic, and environmental-based objectives. Methanol production rate, total energy consumption, total annual CO2 emissions, and profit are considered as objectives. The five decision variables include syngas molar flow, reactor feed pressure, reactor 1 temperature, reactor 2 temperature, and by-product mass flow in the methanol distillation column. Two constraints are set on ethanol mole fraction in the methanol stream, and methanol mole flow in water to the treatment stream. Three optimization cases (two bi-objective and one tri-objective) are developed. In the Case 1 study, CO2 emissions decreased by 65.83%, the methanol production rate increased by 10.11%, and total energy was reduced by 57.55%. The Pareto ranking is carried out using the Preference Ranking On the Basis of Ideal-average Distance (PROBID) method and the best optimal solution is reported for all cases.

Multi-epitope vaccine design of African swine fever virus considering T cell and B cell immunogenicity

T and B cell activation are equally important in triggering and orchestrating adaptive host responses to design multi-epitope African swine fever virus (ASFV) vaccines. However, few design methods have considered the trade-off between T and B cell immunogenicity when identifying promising ASFV epitopes. This work proposed a novel Pareto front-based ASFV screening method PFAS to identify promising epitopes for designing multi-epitope vaccines utilizing five ASFV Georgia 2007/1 sequences. To accurately predict T cell immunogenicity, four scoring methods were used to estimate the T cell activation in the four stages, including proteasomal cleavage probability, transporter associated with antigen processing transport efficiency, class I binding affinity of the major histocompatibility complex, and CD8 + cytotoxic T cell immunogenicity. PFAS ranked promising epitopes using a Pareto front method considering T and B cell immunogenicity. The coefficient of determination between the Pareto ranks of multi-epitope vaccines and survival days of swine vaccinations was R2 = 0.95. Consequently, PFAS scored complete epitope profiles and identified 72 promising top-ranked epitopes, including 46 CD2v epitopes, two p30 epitopes, 10 p72 epitopes, and 14 pp220 epitopes. PFAS is the first method of using the Pareto front approach to identify promising epitopes that considers the objectives of maximizing both T and B cell immunogenicity. The top-ranked promising epitopes can be cost-effectively validated in vitro. The Pareto front approach can be adaptively applied to various epitope predictors for bacterial, viral and cancer vaccine developments. The MATLAB code of the Pareto front method was available at https://github.com/NYCU-ICLAB/PFAS.Graphical

Effective drug-target affinity prediction via generative active learning

Drug-target affinity (DTA) prediction is a critical early-stage task in drug discovery. Recently, deep learning has demonstrated remarkable efficacy in DTA prediction. However, acquiring experimentally verified data for target proteins proves to be a time-consuming, labor-intensive, and costly endeavor. In this study, we introduce an innovative generative active learning method for DTA prediction, referred to as GAL-DTA. GAL-DTA comprises two modules, data augmentation and generator fine-tuning. In the data augmentation module, the algorithm uses an optimized generator to produce informative and diverse molecules, thereby enhancing training of the predictor. The generator fine-tuning module introduces Fisher's informativeness and molecule diversity as objectives and employs the Pareto ranking algorithm to compute rewards. The final generator is fine-tuned using the policy-gradient method. GAL-DTA performs data augmentation by directly generating diverse and informative data, effectively reducing annotation costs while preserving model performance. Extensive experiments on independent test sets involving four target proteins consistently demonstrated that GAL-DTA achieves superior performance, resulting in an average reduction of 8.402% in mean squared error.

QEMCGAN: Quantized Evolutionary Gradient Aware Multiobjective Cyclic GAN for Medical Image Translation.

Generative adversarial networks (GANs) have become popular in medical imaging because of their remarkable performance and ability to translate images across different domains. However, GANs face several issues in image-to-image translation, including training instability, lack of diversity, and mode collapse. These issues become even more complex when using cyclic GANs. Additionally, collecting paired images required for GANs may be costly, especially in the medical domain. Cyclic GANs are a favorable choice for addressing this issue, as they can convert cross-domain images. However, no pre-existing technique or algorithm is comprehensive enough to handle diverse datasets and applications. To address these issues, we propose a novel Quantized Evolutionary Gradient Aware Multiobjective Cyclic GAN (QEMCGAN) that employs evolutionary computation, multiobjective optimization, and an intelligent selection scheme. We use simulated annealing and Pareto ranking selection using three fitness criteria to address local optima stagnation. Additionally, we use model quantization because of its suitability for low-cost IoT-based applications. Extensive trials reveal that EMCGAN and QEMCGAN produces more visually realistic images than other approaches while preserving both background information and salient features. In addition, QEMCGAN performs on par with the baseline approach even when the model size is halved, making it more efficient.

Theoretical morphospace analysis of neuropteran wings reveals little evidence of optimization for flight performance

Abstract The wing is the key evolutionary innovation of pterygote insects and wing morphology is commonly envisaged as finely attuned to functional performance. Here, we use a theoretical morphospace approach to analyse the evolution of disparity and functional optimality in neuropteran wings, thus, investigating how wings are adapted for flight and how varied factors constrain wing shape. Analysing 738 neuropteran wings from the Permian to the present, we construct a theoretical morphospace with 1092 hypothetical shapes. These theoretical wings were subjected to functional analysis, creating performance landscapes for aerodynamic traits. We used a Pareto ranking scheme to create an optimality landscape for the trade-off between functional traits, which identifies wings with a convex leading edge as optimal. After projecting neuropteran wings onto this optimal landscape, we find that neuropterans do not occupy this optimal region. Aerodynamic constraint is evident solely in increasing aspect ratio, increasing the lift-to-drag ratio and therefore flight efficiency. Phylomorphospaces reveal a large overlap between different neuropteran clades, reflecting convergence on these same functionally sub-optimal wing shapes. Forewings and hindwings are subject to different evolutionary constraints with forewings more aerodynamically related and hindwings more phylogenetically related. The theoretical morphology approach facilitates morphological analyses without prior assumptions of adaptive optimality.

Sustainable Material Selection in New Constructions: A Brute‐Force Optimization Framework Using Parallel Computing, Cost Benefits, and Thermal Performance Analysis

Sustainable construction practices rely on carefully selecting building materials and balancing environmental and economic considerations. This study examines the complex link between local climate, market dynamics, and building material selection. Market data analysis, parametric modeling, and brute‐force optimization are used to provide insights into construction decision‐making. Across 5540 simulations, a thorough assessment of the financial and energy performance of various materials for walls, roofs, windows, and floors is conducted. Incorporating Pareto ranking, parallel simulation, and sensitivity analysis, the comprehensive evaluation reveals the intricate tradeoffs between cost, thermal properties, and energy savings. The findings highlight the potential for optimal external wall solutions to reduce U‐values by up to 30% and achieve source energy savings of up to 25% source energy savings across diverse climates. By emphasizing the importance of local context in material selection, this study highlights how energy consumption patterns and transmission losses influence financial and energy performance, thus advancing sustainable construction practices.

Identifying critical landscape patterns for simultaneous provision of multiple ecosystem services – A case study in the central district of Wuhu City, China

Landscape patterns have a global impact on ecosystem services (ESs). Managing land to enhance the provision of multiple ESs is vital for ecological resilience and sustainable development. Identifying critical landscape patterns that simultaneously influence multiple ESs provision based on understanding ES trade-offs is needed. Our study aims to develop an approach to find the relationship between multiple ESs provision and landscape patterns, identify the critical landscape patterns that simultaneously promote multiple ESs provision, and give suggestions on land management based on an understanding of the relationships of ESs. Taking the central district of Wuhu City as an example, production possibility frontier (PPF) and Pareto ranks are used to identify relationships between ESs and extract the landscape pattern characteristics of samples that effectively enhance multiple ESs. Our analysis of trade-offs between ESs highlights the complexity of ES relationships and emphasizes the necessity for multi-dimensional investigative approaches. Landscape metrics of Mean Perimeter Area Ratio (PARA_MN), Coefficient of Variation of Perimeter (PERI_CV), Area-weighted Mean Fractal Dimension (FRAC_AM), and Relative Mutual Information (RELMUTINF) emerged as critical determinants of multiple ESs supply. Our findings suggest enhancing patch shape complexity when conducting land use planning to avoid homogenization and keeping differences between the diversity of land use categories and the diversity of adjacencies small. Also, we recommend caution in using high contagion as the only criterion for land use planning, as its effects may vary on different ES targets.

A multi-objective optimization framework for functional arrangement in smart floating cities

Before the terms “smart city” and “floating city” were introduced, the world's population had increased and land shortage across the world was already widely recognized. As a first challenge, the previous studies have developed the concept of a smart city as a creative answer, following that, several scientists proposed the floating city concept in the literature as a solution to the increased sea levels. Moreover, engineers, architects, and designers deal with city planning, for smart and floating settlements as a difficult design challenge, and evolutionary algorithms could be employed to address this complex problem by optimizing residents' needs. As a continuation of our previous studies on this topic, this time, we develop a multi-objective continuous genetic algorithm with differential evolution (DE) mutation strategy (MO_CGADE) and a multi-objective ensemble differential evolution algorithm (MO_EDE) to solve the problem on hand. Then, we compare the performance of the MO_CGADE and MO_EDE algorithms with the non-dominated sorting genetic algorithm (NSGAII) to maximize two conflicted objective functions, namely, scenery, and walkability in the proposed smart floating city model created in the Grasshopper Algorithmic Modeling Environment. The parametric model that we create in the Grasshopper software includes 64 decision variables, area constraints and objective functions to be optimized by MO_CGADE, MO_EDE, and NSGAII algorithms. Computational results show that MO_CGADE and MO_EDE algorithms generate better Pareto ranking results than the traditional NSGAII algorithm in terms of cardinality, distribution spacing, and coverage metrics.

Comparative evaluation of SPE methods for biotoxicity assessment of water and wastewater: Linkage between chemical extracting efficiency and biotoxicity outcome

Biotoxicity assessment results of environmental waters largely depend on the sample extraction protocols that enrich pollutants to meet the effect-trigger thresholds of bioassays. However, more chemical mixture does not necessarily translate to higher combined biotoxicity. Thus, there is a need to establish the link between chemical extracting efficiency and biotoxicity outcome to standardize extraction methods for biotoxicity assessment of environmental waters. This study compares the performance of five different extraction phases in solid phase extraction (SPE), namely HLB, HLB+Coconut, C18 cartridge, C18 disk and Strata-X, and evaluated their chemical extracting efficiencies and biotoxicity outcomes. We quantitatively assessed cytotoxicity, acute toxicity, genotoxicity, estrogenic activity, and neurotoxicity of the extracts using in vitro bioassays and characterized the chemical extracting efficiencies of the SPE methods through chemical recoveries of 23 model compounds with different polarities and total organic carbon. Using Pareto ranking, we identified HLB+Coconut as the optimal SPE method, which exhibited the highest level of water sample biotoxicity and recovered the most chemicals in water samples. We found that the biotoxicity outcomes of the extracted water samples significantly and positively correlated with the chemical extracting efficiencies of the SPE methods. Moreover, we observed synchronous changing patterns in biotoxicity outcome and chemical extracting efficiencies in response to increasing sample volumes per cartridge (SVPC) during SPE. Our findings underscore that higher chemical extracting efficiency of SPE corresponds to higher biotoxicity outcome of environmental water samples, providing a scientific basis for standardization of SPE methods for adequate assessment of biotoxicities of environmental waters.

Sustainable design and multi-objective optimization of heat pump assisted extractive distillation process for separating a ternary mixture of methyl acetate, tetrahydrofuran and methanol

The separation of a ternary mixture of methyl acetate (MeAc), tetrahydrofuran (THF), and methanol (MeOH) produced from chemical or pharmaceutical plants can not only recover the valuable chemicals but also reduce the environmental pollution caused by the emission of waste organics. In this study, a two-step extractive distillation process with four columns (FCED) with dimethyl sulfoxide and p-xylene serving as the entrainers was firstly proposed to separate the ternary mixture by using Aspen Plus simulation. The vapor recompression heat pump (VRHP) technology was introduced to improve the thermodynamic efficiency of the proposed FCED process. Subsequently, a multi-objective optimization using the non-dominated sorting genetic algorithm-II was employed to optimize the designed processes to accomplish a Pareto front trade-off in three crucial indicators, i.e., total annual cost (TAC), CO2 emission, and exergy efficiency. A Pareto ranking method, i.e., the technique for order of preference by similarity to ideal solutions combined with entropy weighting approach, was used to choose the best solution from the Pareto front. The results indicate that, compared to the FCED process, TAC (8-year payback period) and CO2 emissions of the VRHP-assisted FCED are reduced by 8.75% and 33.26%, respectively, while the exergy efficiency is increased by 8.06% from 3.67% to 3.96%, demonstrating that the introduction of VRHP can significantly improve the performance of the FCED process.

Multi-objective optimization of fed-batch bioreactor for lysine production

ABSTRACT Lysine production via the fermentation process is one of the most economical production routes that involves simultaneous conflicting objectives. Two multi-objective differential evolution algorithms MODE-III and MODE-III-IMS are used to obtain the optimal control parameters of the lysine bioreactor. Two conflicting objectives, namely yield and productivity, are studied. Singular and constant feeding policies using both algorithms are studied, and respective Pareto fronts are reported. Tournament selection and penalty constraint handling methods are used, and their performances are compared for both algorithms. The higher bound of feeding rate (2.0 L/s) is found to be the best feeding rate compared to other feeding rates. The MODE-III-IMS algorithm converged to the Pareto front faster than the MODE-III algorithm. The COPRAS method is used to carry out the Pareto ranking, and the best optimal solution is reported. The decision tree method is used to predict and report the best optimal solution with acceptable accuracy.

Multi-objective Hardware-aware Neural Architecture Search with Pareto Rank-preserving Surrogate Models

Deep learning (DL) models such as convolutional neural networks (ConvNets) are being deployed to solve various computer vision and natural language processing tasks at the edge. It is a challenge to find the right DL architecture that simultaneously meets the accuracy, power, and performance budgets of such resource-constrained devices. Hardware-aware Neural Architecture Search (HW-NAS) has recently gained steam by automating the design of efficient DL models for a variety of target hardware platforms. However, such algorithms require excessive computational resources. Thousands of GPU days are required to evaluate and explore an architecture search space such as FBNet [ 45 ]. State-of-the-art approaches propose using surrogate models to predict architecture accuracy and hardware performance to speed up HW-NAS. Existing approaches use independent surrogate models to estimate each objective, resulting in non-optimal Pareto fronts. In this article, HW-PR-NAS, 1 a novel Pareto rank-preserving surrogate model for edge computing platforms, is presented. Our model integrates a new loss function that ranks the architectures according to their Pareto rank, regardless of the actual values of the various objectives. We employ a simple yet effective surrogate model architecture that can be generalized to any standard DL model. We then present an optimized evolutionary algorithm that uses and validates our surrogate model. Our approach has been evaluated on seven edge hardware platforms from various classes, including ASIC, FPGA, GPU, and multi-core CPU. The evaluation results show that HW-PR-NAS achieves up to 2.5× speedup compared to state-of-the-art methods while achieving 98% near the actual Pareto front.

Multi-criteria analysis of cell-recycle based continuous lactic acid production process

ABSTRACT This study focuses on the optimization of the cell-recycle based continuous lactic acid (LA) fermentation production process with the estimation of improved kinetic parameters. The non-dominated sorting genetic algorithm II (NSGA-II) is used considering seven variables and five imposed constraints for multi-objective optimization (MOO) study. Three objectives namely LA productivity, substrate conversion, and yield are formed for the maximization purpose. Three different cases are constructed by using a combination of two bi-objectives and one tri-objective optimization problem, and the modified differential evolution (MDE) algorithm is used to individually optimize each case. Pareto ranking is carried out for MOO, and the best optimal solution is determined using the technique for order preference by similarity to the ideal solution (TOPSIS) method. The highest value of volumetric productivity of LA (184.363 kg/m3.h) is reported in case-4a. The highest yield (1.176) is reported in case-2, while the highest substrate conversion (0.963) is reported in case-1.

Insight into single- and bi-objective optimization of industrial problems

ABSTRACT Evolutionary single- and bi-objective optimization of industrial problems, namely, naphtha cracking and styrene reactor are considered. Bi-objective optimization is solved using the HMODE-DLS algorithm, whereas an improved black hole optimizer (BHO) is employed for the single-objective optimization (SOO) problem. For the naphtha cracking process, ethylene selectivity (SE) and severity index (SI) are selected as objectives. Similarly, styrene selectivity (SST) and styrene flow rates (FST) are considered objectives for styrene reactor. Pareto ranking is carried out by using the net flow method (NFM) and the best solution is compared with the single- and multi-objective optimization results. In a single objective optimization study, the optimum solution corresponds to the lowest value of the SI (1.571) at the cost of the worst value of SE (0.269). Similarly, the highest values of FST (16.642 kmol/h) and SST (96.158%) are resulted in the SOO study of styrene reactor.

Social vulnerability to long-duration power outages

Although long-duration power outages can lead to significant damages to the local economy, the human catastrophe that can potentially unfold due to the failure of essential services can far outweigh the financial damages incurred. Furthermore, power outages do not impact individuals equally, and access to proper resources (or lack thereof) can significantly affect how individuals deal with long-duration outages. Various socioeconomic and demographic characteristics have been shown in the literature to correlate with increased health risks, levels of power outage preparedness, and willingness and means to evacuate if necessary. This highlights the need to identify socially vulnerable groups and communities so that during such events information, assistance, and resources can be provided in a more targeted manner. This study presents a three-dimensional metric of social vulnerability to quantify the degree to which a person's life or livelihood is put at risk by a long-duration power outage. Dimensions of vulnerability include health, preparedness, and evacuation. Principal component analysis and an L2 norm model are applied to produce a single metric for each of the three dimensions of vulnerability. These three scores are then aggregated using Pareto ranking to determine an overall vulnerability score. A case study is presented for the state of Colorado using data from the 2020 US Census as well as other relevant federal and state datasets. The index put forth in this paper can be used by power utilities and other federal, state, or local authorities to plan and manage their operations during long-duration power outages in a more equitable way.

SCREENING OF MOST EFFECTIVE VARIABLES FOR DEVELOPMENT OF TELMISARTAN LIPOSOMES BY TAGUCHI DESIGN

The aim of this study was the selection of the most effective variable for the formulation of telmisartan liposomes. Liposomes were prepared by dry film deposition method using a Rota evaporator. The effects of formulation and processing variables on different response variables were studied using standard Taguchi orthogonal array L8 design. The independent variables selected were the amount of lipid, amount of cholesterol, sonication time, speed of Rota evaporator, hydration time, hydration volume and temperature, and the dependent variable selected were particle size (nm) and entrapment efficiency (%). All the batches of liposomes were prepared as per the Taguchi design. The influence of variables analyzed by Pareto ranking showed that the most effective factors for the selected responses were the amount of cholesterol, amount of lipid and sonication time (P<0.05).

DPb-MOPSO: A Dynamic Pareto bi-level Multi-objective Particle Swarm Optimization Algorithm

Particle Swarm Optimization (PSO) system based on the distributed architecture over multiple sub-swarms is very efficient for static multi-objective optimization but has not been considered for solving dynamic multi-objective problems (DMOPs). Tracking the most effective solutions over time and ensuring good exploitation and exploration are the main challenges of solving DMOP. This study proposes a Dynamic Pareto bi-level Multi-Objective Particle Swarm Optimization (DPb-MOPSO) algorithm including two parallel optimization levels. At the first level, all solutions are managed in a single search space. When a dynamic change is successfully detected in the objective values, the Pareto ranking operator is used to enable multiple sub-swarm’ subdivisions and processing which drives the second level of enhanced exploitation. A dynamic handling strategy based on random detectors is used to track the changes in the objective function due to time-varying parameters. A response strategy consisting in reevaluating all unimproved solutions and replacing them with newly generated ones is also implemented. The DPb-MOPSO system is tested on DMOPs with different types of time-varying Pareto Optimal Set (POS) and Pareto Optimal Front (POF). Inverted generational distance (IGD), mean inverted generational distance (MIGD), hypervolume difference (HVD), Robust IGD (RIGD), and Robust General Distance (RGD) metrics are used to assess the DPb-MOPSO performance. Quantitative results are analyzed using Friedman’s analysis of variance, and the Wilcoxon sum ranks test, while the stability is analyzed using Lyapunov’s theorem. The DPb-MOPSO is more robust than several dynamic multi-objective evolutionary algorithms in solving 21 complex problems over a range of changes in both the POS and POF. On IGD and HVD, DPb-MOPSO can solve 8/13 and 8/13 of the 13 UDF and ZJZ functions with moderate changes. DPb-MOPSO can resolve 7/8 FDA and DMOP benchmarks with severe changes to the MIGD, and 6/8 with moderate changes. DPb-MOPSO assumes 7/8, 6/8, and 5/8 for solving FDA, and dMOP functions on IGD and 6/8, 5/8, and 5/8 on HVD metrics considering severe, moderate, and slight environmental changes respectively. Also, it is the winner for solving 8 DMOPs based on RIGD, and RGD metrics.

RESISTOR: A New OSPREY Module to Predict Resistance Mutations.

Computational, in silico prediction of resistance-conferring escape mutations could accelerate the design of therapeutics less prone to resistance. This article describes how to use the Resistor algorithm to predict escape mutations. Resistor employs Pareto optimization on four resistance-conferring criteria-positive and negative design, mutational probability, and hotspot cardinality-to assign a Pareto rank to each prospective mutant. It also predicts the mechanism of resistance, that is, whether a mutant ablates binding to a drug, strengthens binding to the endogenous ligand, or a combination of these two factors, and provides structural models of the mutants. Resistor is part of the free and open-source computational protein design software OSPREY.

A novel multi-level population hybrid search evolution algorithm for constrained multi-objective optimization problems

Constrained multi-objective optimization problem (CMOP) considers the convergence, diversity and feasibility of the population in the optimization process, so it is challenging to find desirable solutions of CMOP. Existing evolutionary multi-objective optimization algorithms have good performance on unconstrained multi-objective optimization problems, but have difficulties in solving CMOPs in discrete feasible regions. Aiming at this issue, this paper proposes a novel multi-level population hybrid search evolution algorithm (MLHSEA). First, a new multi-level hybrid search strategy (i.e. MHSS) is designed in the algorithm, which divides the population into three-level subpopulations based on Pareto ranks, constraint violation degree values, and feasible thresholds. Each subpopulation has its own unique evolution strategy to maximize the evolutionary potential of each subpopulation, which is beneficial to make some feasible solutions break through the discrete feasible region and reach the Pareto frontier. Then, a new population fusion degree strategy (i.e. PFDS) is proposed to timely perform population fusion and information exchange according to the population fusion degree (PFD) of each sub-population, thus improving the searchability of the target space. Finally, a novel detection reset strategy (i.e. DRS) is proposed for the lowest inferior subpopulation. This strategy can make inferior subpopulations avoid unnecessary evolutionary iterations and improve population diversity. Based on constrained test suites with four different characteristics, the experimental results show that the proposed MLHSEA outperforms other state-of-the-art constrained multi-objective optimization algorithms in performance.