Hybrid Multi-objective Optimization Research Articles

The Case for Hybrid Multi-Objective Optimisation in High-Stakes Machine Learning Applications

The Case for Hybrid Multi-Objective Optimisation in High-Stakes Machine Learning Applications

Maximum pitting corrosion depth prediction of buried pipeline based on theory-guided machine learning

Buried pipelines are crucial for the transportation of oil and natural gas resources. However, pipeline failure accidents have frequently occurred due to corrosion. Therefore, an accurate corrosion depth prediction model is necessary for the reliable supply of energy. In this paper, a theory-guided machine learning (ML) model is developed for maximum pitting corrosion depth prediction, the engineering theory and domain knowledge are integrated into feature space to improve the model interpretability. Firstly, several new feature variables are constructed based on the interactions between independent variables. Then, feature importance of all feature variables is obtained using random forest (RF). A hybrid multi-objective grey wolf optimization (HMOGWO) is proposed to optimize the hyper-parameters of RF model, considering feature number, prediction accuracy, and stability simultaneously. Finally, a comprehensive pitting corrosion dataset is utilized for performance evaluation. The results indicate that the proposed theory-guided model can achieve high prediction accuracy and stability, the optimal feature subset can be determined using multi-objective optimization method simultaneously, which solves the problems of model interpretability and feature selection for traditional ML models with the single-objective optimizer. This study is of great significance to the transportation safety of buried pipelines.

Demand side management strategy for smart building using multi-objective hybrid optimization technique

This study proposes a home energy management system that uses the load-shifting technique for demand-side management as a way to improve the energy consumption patterns of a smart house. This system's goal is to optimize the energy of household appliances in order to effectively regulate load demand, with the end result being a reduction in the peak-to-average ratio (PAR) and a consequent minimization of electricity costs. This is accomplished while also keeping user comfort as a priority. Load scheduling based on both a next-day and real-time basis is what is used to meet the load demand requested by energy customers. In addition to providing a fitness criterion, utilizing a multi-objective hybrid optimization technique makes it easier to achieve an equitable distribution of workload between on-peak and off-peak hours. Moreover, the idea of developing coordination among home appliances in order to achieve real-time rescheduling is now being studied as a concept. Because of the inherent parallels between the two problems, the real-time rescheduling issue is framed as a knapsack problem and is solved using a dynamic programming strategy. The performance of the suggested methodology is evaluated in this study in relation to real-time pricing (RTP), time-of-use pricing (ToU), and crucial peak pricing (CPP). The simulation findings, which were assessed using a confidence interval that was set at 95 %, provide proof of the relevance that has been shown to be associated with the proposed optimization method. During scheduling RTP signal showcases a minimum PAR of 2.22 and a cost reduction of 24.06 % for HAG compared to the unscheduled case. Under the TOU tariff, HAG manages to reduce PAR by 46.14 % and cost by 20.44 %. Similarly, in the case of CPP, HAG outperforms by reducing PAR by up to 29.5 % and cost by up to 31.47 %.

Re-Entrant Green Scheduling Problem of Bearing Production Shops Considering Job Reworking

To solve various reworking and repair problems caused by unqualified bearing product quality inspections, this paper introduces a green re-entrant scheduling optimization method for bearing production shops considering job reworking. By taking into account quality inspection constraints, this paper establishes an integrated scheduling mathematical model based on the entire processing–transportation–assembly process of bearing production shops with the goals for minimizing the makespan, total carbon emissions, and waste emissions. To solve these problems, the concepts of the set of the longest common machine routes (SLCMR) and the set of the shortest recombination machine combinations (SSRMC) were used to propose the re-entrant scheduling optimization method, based on system reconfiguration, to enhance the system stability and production scheduling efficiency. Then, a multi-objective hybrid optimization algorithm, based on a neighborhood local search (MOOA-LS), is proposed to improve the search scope and optimization ability by constructing a multi-level neighborhood search structure. Finally, this paper takes a bearing production shop as an example to carry out the case study and designs a series of experimental analyses and comparative tests. The final results show that in the bearing production process, the proposed model and algorithm can effectively realize green and energy-saving re-entrant manufacturing scheduling.

Enhancing the resilience of error-prone computing environments using a hybrid multi-objective optimization algorithm for edge-centric cloud computing systems

Enhancing the resilience of error-prone computing environments using a hybrid multi-objective optimization algorithm for edge-centric cloud computing systems

Dynamic Load Balancing in Cloud Computing: Optimized RL-Based Clustering with Multi-Objective Optimized Task Scheduling

Dynamic load balancing in cloud computing is crucial for efficiently distributing workloads across available resources, ensuring optimal performance. This research introduces a novel dynamic load-balancing approach that leverages a deep learning model combining Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs) to calculate load values for each virtual machine (VM). The methodology aims to enhance cloud performance by optimizing task scheduling and stress distribution. The proposed model employs a dynamic clustering mechanism based on computed loads to categorize VMs into overloaded and underloaded clusters. To improve clustering efficiency, the approach integrates Reinforcement Learning (RL) with a sophisticated Hybrid Lyrebird Falcon Optimization (HLFO) algorithm. HLFO merges the Lyrebird Optimization Algorithm (LOA) and Falcon Optimization Algorithm (FOA), enhancing the effectiveness of load balancing. A Multi-Objective Hybrid Optimization model is introduced to optimize task scheduling while considering Quality of Service (QoS) parameters, including makespan minimization, energy consumption reduction, balanced CPU utilization, efficient memory usage, and task prioritization. The implementation, conducted in Python and CloudSim, demonstrates the model’s ability to effectively allocate work between virtual machines (VMs) and physical machines (PMs), resulting in improved resource utilization, shortened makespan, enhanced CPU usage, and rigorous assessments affirming its efficacy. This research addresses the complexity of dynamic load balancing in cloud environments by combining deep learning, reinforcement learning, and hybrid optimization techniques, offering a comprehensive solution to optimize cloud performance under varying workloads and resource conditions.

Multi-Objective Multi-Skill Resource-Constrained Project Scheduling Considering Flexible Resource Profiles

This paper addresses a novel multi-skill resource-constrained project scheduling problem with flexible resource profiles (F-MSRCPSP), in which the resource allocation of each activity consists of a certain number of discrete resources and is allowed to be adjusted over its duration. The F-MSRCPSP aims, therefore, to determine the flexible resource profile of each activity to minimize the make-span and total cost simultaneously. Then, a hybrid multi-objective fruit fly optimization algorithm is proposed to handle the concerned problem. In the proposed algorithm, two flexible parallel and serial schedule generation schemes are introduced, aiming to schedule activities and adjust allocated resource combinations. Additionally, two heuristic strategies are proposed to effectively select suitable resource combinations for activities. Moreover, a series of operators has been developed, including the rejoining operator, empirical re-arrangement operator, and empirical re-selection operator. These operators aim to accelerate the convergence speed and enhance the exploration of the proposed algorithm. Finally, the orthogonal test is used to select the optimal parameter combination, and comparative experiments based on tests with different scales are conducted, along with a t-test. The experimental results demonstrate that MOFOA-HS is effective in solving the F-MSRCPSP.

Crashworthiness Analysis and Optimization of Thin-Walled Structures Based on the TSSA-LSSVM Surrogate Model

Thin-Walled Structures (TWS) play a crucial role in enhancing both the collision safety and lightweight characteristics of vehicles. Structural design and optimization have demonstrated significant potential in improving their crashworthiness and overall weight reduction. Several interconnected factors, including the cross-sectional shape and parameters of distinctive structural characteristics (known as induction grooves), impact the TWS’ resistance to collisions. Another critical aspect in the optimization of TWS design is striking the delicate balance between the surrogate model’s effectiveness and precision. Taking all these factors into account, this paper conducts an in-depth investigation into the cross-sectional shape, grooves, and other geometric attributes of various TWSs through a comparative analysis. It reveals that the thin-walled structure with a Double-Ribbed Rectangle cross-section (DR-TWS) proves to be the optimal choice, particularly in low-speed axial impact scenarios. Furthermore, we present an enhanced Least Squares Support Vector Machine (LSSVM) method, augmented with an adaptive [Formula: see text]-distribution Sparrow Search Algorithm (TSSA), referred to as the TSSA-LSSVM surrogate model. The optimized design of the DR-TWS is realized through the integration of the TSSA-LSSVM surrogate model with the adaptive hybrid multi-objective particle swarm optimization (AHMOPSO) algorithm. This paper not only provides substantial theoretical support but also offers valuable guidelines for the application of TWSs in the automotive industry.

A Two-Stage Hybrid Multi-Objective Optimization Evolutionary Algorithm for Computing Offloading in Sustainable Edge Computing

A Two-Stage Hybrid Multi-Objective Optimization Evolutionary Algorithm for Computing Offloading in Sustainable Edge Computing

An interactive possibilistic programming approach for sustainable supplier selection and inventory management problem in the physical internet

PurposeThis paper aims to introduce a novel structure for the physical internet (PI)–enabled sustainable supplier selection and inventory management problem under uncertain environments.Design/methodology/approachTo address hybrid uncertainty both in the objective function and constraints, a novel interactive hybrid multi-objective optimization solution approach combining Me-based fuzzy possibilistic programming and interval programming approaches is tailored.FindingsVarious numerical experiments are introduced to validate the feasibility of the established model and the proposed solution method.Originality/valueDue to its interconnectedness, the PI has the opportunity to support firms in addressing sustainability challenges and reducing initial impact. The sustainable supplier selection and inventory management have become critical operational challenges in PI-enabled supply chain problems. This is the first attempt on this issue, which uses the presented novel interactive possibilistic programming method.

Optimization of the octagonal origami tube energy absorption using the artificial neural network and heuristic method

This article presents a hybrid optimization approach for improving the crashworthiness characteristics of thin-walled metallic tubes using a unique multilayered origami pattern. In the quasi-static axial loading condition, a total of 120 thin-walled tube samples were simulated using the full factorial design method to assess the impact of geometric parameters on their crashworthiness behavior, including the dihedral angle, module number, and outer layer height. Additionally, based on the input parameters, an artificial neural network was implemented to estimate the mechanical response of the tubes. Then, to determine the ideal design parameters, a hybrid multi-objective optimization approach was used. The Technique for Order of Preference by Similarity to Ideal Solution technique was utilized to extract the optimal design to balance the objectives of energy absorption and initial peak force. The specific energy absorption of the optimal tube was discovered to be 68% higher than that of the base tube. The simulation results demonstrate that the proposed multilayered origami pattern provides a unique structure that improves the crashworthiness properties of the thin-walled metallic tubes.

Multi-objective hybrid optimization for micro strip patch antenna design

Due to their low price, light weights, as well as simple installation, Micro strip Patch Antennas (MPAs) have been made to perform in a double and multi-band applications. The MP receiver is created with an Electromagnetic Band Gap (EBG) structure in order to decrease the micro strip patch cross-polarized radiation but also achieve the crucial radiation criteria. The polymeric liquid crystals substratum is employed to decrease raw material costs, and also the applicable shape framework are employed to enhance receiver execution. We have established a new optimization based method which has two operating stages. In the begining stage, we have designed a Micro strip patch antenna with certain parameters. Afterwards, these design parameters length, width, height, substrate thickness under area such as get optimized by the newly introduced Battle Royale Customized Spider Monkey Optimization (BRCSMO) algorithm in order to get an antenna with higher performance. We have evaluated the proposed method with regard to measures like receiver profit, productivity, bandwidth, decline loss as well as Total Active Reflection coefficient (TARC) and the outcomes showed that this proposed technique can offer superior outcomes than other approaches.

Multi-objective impact mechanism on the performance characteristic for a diesel particulate filter by RF-NSGA III-TOPSIS during soot loading

Diesel particulate filter (DPF) provides an effective control method for particulate matter (PM) emissions from diesel vehicles. In the PM trap process of DPF, pressure drop and filtration efficiency present conflicting relationships as the main indicators of the comprehensive trap performance. To fully optimize the comprehensive trap performance of DPF, this research develops a hybrid multi-objective optimization approach of FGRA-RF-NSGA III-TOPSIS. Specifically, critical structures are selected from DPF structural features by fuzzy grey correlation analysis (FGRA). Three machine learning (ML) models are trained on the computational fluid dynamics (CFD) model dataset to relate the critical structures to the initial filtration efficiency and pressure drop. Then, statistical indicators are introduced to evaluate three ML models, and the best-performing model – The random forest (RF) model is selected as the input of the optimization algorithm. Finally, a hybrid Critic-TOPSIS approach is used to select the most ideal solution from the Pareto frontier obtained by the NSGA III as the output of the optimization results. The optimized DPF reduces the pressure drop by 49.84 % and improves the initial filtration efficiency by 49.31 % compared to the original DPF under standard operating conditions. The optimization effect is more significant at high-speed and high-load conditions.

Multi-objective sparse synthesis optimization of concentric circular antenna array via hybrid evolutionary computation approach

Sparse synthesis can reduce the maximum sidelobe level (SLL) and overhead of the concentric circular antenna array (CCAA) by switching off partial array elements. However, previous works have not considered finding a balance between the number of switched-on elements of CCAA and reducing the maximum SLL, or jointly optimizing the excitation current weights of array elements to further suppress the maximum SLL. In this work, we address these challenges by formulating a hybrid multi-objective optimization problem (MOP) with discrete and continuous solutions. To solve this hybrid NP-hard problem, we consider two approaches: the two-step and one-step methods. In the two-step method, we convert the original MOP into two subproblems and propose an enhanced non-dominated sorting genetic algorithm-II (ENSGA-II) and an improved genetic algorithm with new mutation and attraction (IGAnMA) to solve these two subproblems. ENSGA-II and IGAnMA involve the hierarchy mechanism and attraction operator to enhance the solving abilities, respectively. In the one-step method, we propose an improved hybrid non-dominated sorting genetic algorithm-II (IHNSGA-II) with global search and hybrid solution update methods to solve the MOP in one step. Numerical simulation results show that the proposed methods are effective and stable for the sparse synthesis of CCAA. Additionally, electromagnetic simulation results demonstrate the validity of the proposed approaches in a more practical environment.

Optimal design of triangular side orifice using multi-objective optimization NSGA-II.

Triangular orifices are widely used in industrial and engineering applications, including fluid metering, flow control, and measurement. Predicting discharge through triangle orifices is critical for correct operation and design optimization in various industrial and engineering applications. Traditional approaches like empirical equations have accuracy and application restrictions, whereas computational fluid dynamics (CFD) simulations can be computationally costly. Alternatively, artificial neural networks (ANNs) have emerged as a successful solution for predicting discharge through orifices. They offer a dependable and efficient alternative to conventional techniques for estimating discharge coefficients, especially in intricate relationships between input parameters and discharge. In this study, ANN models were created to predict discharge through the triangle orifice and velocity at the downstream of the main channel, and their effectiveness was assessed by comparing the performance with the earlier models proposed by researchers. This paper also proposes a novel hybrid multi-objective optimization model (NSGA-II) that uses genetic algorithms to discover the best values for design parameters that maximize discharge and downstream velocity simultaneously.

Multiobjective optimization of injection molding parameters based on the GEK-MPDE method

Abstract In plastic injection molding (PIM), the process parameters determine the quality and productivity of molded parts. The traditional injection molding process analysis method mainly relies on production experience. It is lack of advanced and rationality and seriously increases production costs. In this paper, a hybrid multiobjective optimization method is proposed to minimize the warpage, volumetric shrinkage and cycle time. The method integrates orthogonal experimental design, numerical simulation, and the metamodel method with multiobjective optimization. The orthogonal experiment chooses seven parameters as the design variables to generate sampling data and determines key factors that affect product quality by the numerical simulation. A gradient-enhanced Kriging (GEK) surrogate model strategy is introduced to construct the response predictors to calculate objective responses in the global design space. Multipopulation differential evolution (MPDE) is conducted to locate the Pareto-optimal solutions, where the response predictors are taken as the fitness functions. This study shows that the proposed GEK-MPDE method can reduce warpage, volumetric shrinkage and cycle time by 5.7 %, 4.7 %, and 18.1 %, respectively. It helps plastic industry to realize collaborative scheduling of multiple tasks between different production lines by providing a low-cost and effective dynamic control method.

An Efficient Hybrid Multi-Objective Optimization Method Coupling Global Evolutionary and Local Gradient Searches for Solving Aerodynamic Optimization Problems

Aerodynamic shape optimization is frequently complicated and challenging due to the involvement of multiple objectives, large-scale decision variables, and expensive cost function evaluation. This paper presents a bilayer parallel hybrid algorithm framework coupling multi-objective local search and global evolution mechanism to improve the optimization efficiency and convergence accuracy in high-dimensional design space. Specifically, an efficient multi-objective hybrid algorithm (MOHA) and a gradient-based surrogate-assisted multi-objective hybrid algorithm (GS-MOHA) are developed under this framework. In MOHA, a novel multi-objective gradient operator is proposed to accelerate the exploration of the Pareto front, and it introduces new individuals to enhance the diversity of the population. Afterward, MOHA achieves a trade-off between exploitation and exploration by selecting elite individuals in the local search space during the evolutionary process. Furthermore, a surrogate-assisted hybrid algorithm based on the gradient-enhanced Kriging with the partial least squares(GEKPLS) approach is established to improve the engineering applicability of MOHA. The optimization results of benchmark functions demonstrate that MOHA is less constrained by dimensionality and can solve multi-objective optimization problems (MOPs) with up to 1000 decision variables. Compared to existing MOEAs, MOHA demonstrates notable enhancements in optimization efficiency and convergence accuracy, specifically achieving a remarkable 5–10 times increase in efficiency. In addition, the optimization efficiency of GS-MOHA is approximately five times that of MOEA/D-EGO and twice that of K-RVEA in the 30-dimensional test functions. Finally, the multi-objective optimization results of the airfoil shape design validate the effectiveness of the proposed algorithms and their potential for engineering applications.

Multi-objective optimization of energy-efficient remanufacturing system scheduling problem with lot-streaming production mode

Most previous studies on the scheduling problem in remanufacturing systems have focused on single or two production stages and economic criteria such as makespan or tardiness. However, environmental criteria like energy consumption or carbon emissions, as well as the lot-streaming production mode, have not been adequately addressed. This paper presents a novel approach to the energy-efficient remanufacturing system scheduling problem with lot-streaming production mode (ERSSP-LS) for the first time, which integrates disassembly, reprocessing, and reassembly three pivotal production stages. Firstly, a multi-objective mathematical model aimed at simultaneously minimizing the total energy consumption (TEC) and makespan (Cmax) is established and presented. To further reduce the TEC, the well-accepted energy-saving measure, known as the turn off and on strategy, is also integrated. Subsequently, a hybrid multi-objective optimization algorithm called HMOFFO, which combines the fruit fly optimization algorithm and simulated annealing mechanism is developed to seek the promising Pareto solution set. Finally, several experiments on a real-life case and a group of ten random test instances are carried out and the computational results affirm the feasibility and effectiveness of the proposed HMOFFO in solving the ERSSP-LS.

Knowledge-Based Reinforcement Learning and Estimation of Distribution Algorithm for Flexible Job Shop Scheduling Problem

In this study, a flexible job shop scheduling problem with time-of-use electricity price constraint is considered. The problem includes machine processing speed, setup time, idle time, and the transportation time between machines. Both maximum completion time and total electricity price are optimized simultaneously. A hybrid multi-objective optimization algorithm of estimation of distribution algorithm and deep Q-network is proposed to solve this. The processing sequence, machine assignment, and processing speed assignment are all described using a three-dimensional solution representation. Two knowledge-based initialization strategies are designed for better performance. In the estimation of distribution algorithm component, three probability matrices corresponding to solution representation are provided. In the deep Q-network component, 34 state features are selected to describe the scheduling situation, while nine knowledge-based actions are defined to refine the scheduling solution, and the reward based on the two objectives is designed. As the knowledge for initialization and optimization strategies, five properties of the considered problem are proposed. The proposed mixed integer linear programming model of the problem is validated by exact solver CPLEX. The results of the numerical testing on wide-range scale instances show that the proposed hybrid algorithm is efficient and effective at solving the integrated flexible job shop scheduling problem.

A hybrid multi-objective optimization method for nuclear essential service water system design

Evolutionary algorithms have proven to be very successful in solving multi-objective optimization problems (MOPs). However, their performance often deteriorates when constraints are introduced. This paper proposes a hybrid heuristic intelligence (HHI) approach to remedy this issue. First, a classification operator is designed to sort individuals by their relative constraint values, which enlarges the size of feasible solutions in the early process. Then, by combining constraints with the non-dominated sorting method, we define a new selection operator to increase the probability of selecting individuals who have low constraint values but high objective values. Finally, we use hybrid optimized operations to simultaneously enhance the convergence and diversity performance by performing different operators in a certain domain. The proposed method achieves state-of-the-art performance on test functions ZDT1, Binh2, and OSY, with a remarkable decrease of 24% in the inverted generational distance (IGD) on OSY. For further applications, our method is also tested with an engineering model established for the design of the nuclear essential service water system (SEC). The final result shows that without taking hundreds of hours, the final solution of HHI has met all constraints and reduced the total cost by 8.9%© YEAR The Authors. Published by Elsevier Ltd.