Multi-objective Genetic Algorithm Approach Research Articles

Optimization of landscape garden greening design based on multi objective genetic algorithm

This paper presents a novel approach to optimize landscape garden greening design using a multi-objective genetic algorithm (MOGA)[1]. Incorporating genetic algorithms into landscape architecture offers a promising avenue for efficiently navigating the complex and multidimensional design space inherent in green infrastructure projects. Through a comprehensive bibliometric analysis of existing literature, this study synthesizes key insights into the application of genetic algorithms in landscape design and identifies gaps for further exploration[2]. Leveraging the evolutionary process of genetic algorithms, our methodology focuses on simultaneously optimizing multiple objectives such as biodiversity conservation, aesthetic appeal, water efficiency, and ecosystem services provisioning[3]. By iteratively evolving and selecting landscape configurations based on fitness criteria derived from these objectives, the MOGA enables designers to explore a diverse range of design alternatives and identify Pareto-optimal solutions that balance competing priorities. The integration of genetic algorithms into landscape design facilitates an iterative and adaptive design process, allowing for the exploration of complex trade-offs and the generation of innovative design solutions. Through a case study application, we demonstrate the effectiveness of the MOGA approach in optimizing landscape garden greening designs, showcasing its potential to enhance sustainability, resilience, and functionality in urban green spaces. This research contributes to the growing body of knowledge on computational design methods in landscape architecture and provides valuable insights for practitioners and researchers seeking to leverage genetic algorithms for optimizing green infrastructure projects.

A multi-objective genetic algorithm approach applied to compact meander branch line couplers design for 5G-enabled IoT applications

A multi-objective genetic algorithm approach applied to compact meander branch line couplers design for 5G-enabled IoT applications

MULTI-OBJECTIVE GENETIC ALGORITHM FOR THE ASSEMBLY LINE WORKER ASSIGNMENT AND BALANCING PROBLEM: A CASE STUDY IN THE AUTOMOTIVE SUPPLY INDUSTRY

The primary challenge in assembly line design is the need for more appropriately allocating tasks and workers to workstations. This study addresses the problem of line balancing and worker assignments, considering the performance disparities among workers during the line balancing process. In the relevant literature, this problem is known as the Assembly Line Worker Assignment and Balancing (ALWAB) problem. This research examines a multi-objective ALWAB Type-2 problem, simultaneously evaluating cycle time and squared load assignment objectives. The study is conducted based on a real-life scenario in a sub-industry automotive industry that manufactures cable equipment. To solve this problem, a multi-objective genetic algorithm approach is proposed. Recognising that the selection of parameter values will influence the algorithm’s performance, parameter calibration has been performed. A full factorial experimental design and the irace method have been utilised for this purpose. The results are compared with those using parameter values utilised for similar problems in the literature. Furthermore, a sensitivity analysis has been carried out to examine the impact of various relative weight values of the objectives on the result. The results indicate that the experimental design generally yields superior results compared to other methods.

Multi-objective genetic algorithm approach for enhanced cumulative hydrogen and methane-rich syngas emission through co-pyrolysis of de-oiled microalgae and coal blending

Towards establishing low-carbon bio-economy, the energy-rich syngas is considered a global energy carrier. Targeting hydrogen as a promising advanced fuel, the significance of methane also increases due to its direct conversion capability into hydrogen. The current study aims to use a co-pyrolysis-based valorization of de-oiled microalgae and low-rank coal blend to generate H2 and CH4 rich syngas. Pyrolysis kinetic models, Kissinger-Akahira-Sunose (KAS) and Starink (STK) are used to evaluate apparent activation energy (Ea). The gradual addition of microalgae (0⎼100%) in coal reduces Ea from 189.11⎼55.87 kJ/mol and 180.16⎼54.61 kJ/mol by KAS and STK method, respectively. The maximum hymethane carrying ratio is observed 2.51 and 3.51 at optimized conditions of response surface methodology (RSM) and artificial neural network-based multi-objective genetic algorithm (ANN-MOGA), respectively. Maximum H2 (54.5 %) in the syngas is observed at mid pyrolysis stage (451 °C) using ANN-MOGA optimized conditions (blending ratio - 42.25 % and heating rate-13.8 °C/min). This study highlights the advantage of ANN-MOGA optimization over statistical based optimization. Hence, incorporating of the evolutionary algorithm as integrated ANN-MOGA optimization could be an efficient way for hymethane rich syngas emission in co-pyrolysis approach to gain carbon-neutral energy with zero waste discharge.

A Multi-Objective Genetic Algorithm Approach for Silicon Photonics Design

A multi-objective genetic algorithm approach is formulated to optimize the design of silicon-photonics complex circuits with contradicting performance metrics and no closed-form expression for the circuit performance. A case study is the interleaver/deinterleaver circuit which mixes/separates optical signals into/from different physical channels while preserving the wavelength-division-multiplexing specifications. These specifications are given as channel spacing of 50 GHz, channel 3-dB bandwidth of at least 20 GHz, channel free spectral range of 100 GHz, crosstalk of −23 dB or less, and signal dispersion less than 30 ps/nm. The essence of the proposed approach lies in the formulation of the fitness functions and the selection criteria to optimize the values of the three coupling coefficients, which govern the circuit performance, in order to accommodate the contradicting performance metrics of the circuit. The proposed approach achieves the optimal design in an incomparably short period of time when contrasted with the previous tedious design method based on employing Z-transform and visual inspection of the transmission poles and zeros.

Optimization of Process Parameters for Better Surface Morphology of Electrical Discharge Machining-Processed Inconel 825 Using Hybrid Response Surface Methodology-Desirability Function and Multi-objective Genetic Algorithm Approaches

Optimization of Process Parameters for Better Surface Morphology of Electrical Discharge Machining-Processed Inconel 825 Using Hybrid Response Surface Methodology-Desirability Function and Multi-objective Genetic Algorithm Approaches

Design of Hybrid-Laminar-Flow-Control Wing and Suction System for Transonic Midrange Aircraft

Hybrid laminar flow control (HLFC) has shown significant promise in the viscous drag reduction of aircraft. However, the use of HLFC for commercial applications requires further simplification. The current study proposes tools for the conceptual design of transonic HLFC wing and suction system. In the first part of the study, airfoil sections for the wing are optimized for minimum total drag using a multi-objective genetic algorithm approach at six spanwise locations. The induced drag of the wing is estimated using a vortex lattice method solver. In the second part of the study, suction system design is performed using ASPeCT, an in-house solver for HLFC system design. A simplified inner structure for the suction system is proposed, which can be integrated easily within the wing structure. A total drag penalty approach is proposed to establish a tradeoff between matching the target suction distribution and the complexity of the suction system. Finally, the additional weight and off-design performance of the suction system are analyzed for a change in the design lift coefficient. A maximum fuel reduction of 7% can be expected with the HLFC system taking into account the additional weight added and power off-take from the engine.

Multi-Objective Parametric Shape Optimisation of Body-Centred Cubic Lattice Structures for Additive Manufacturing

There has been significant interest in additively manufactured lattice structures in recent years due to their enhanced mechanical and multi-physics properties, making them suitable candidates for various applications. This study presents a multi-parameter implicit equation model for designing body-centred cubic (BCC) lattice structures. The model is used in conjunction with a multi-objective genetic algorithm (MOGA) approach to maximise the stiffness of the BCC lattice structure while minimising von-Mises stress within the structure under a specific loading condition. The selected design from the MOGA at a specific lattice density is compared with the classical BCC lattice structure and the designs generated by a single-objective genetic algorithm, which focuses on maximising stiffness or minimising von-Mises stress alone. By conducting a finite element analysis on the optimised samples and performing mechanical testing on the corresponding 3D-printed specimens, it was observed that the optimised lattice structures exhibited a substantial improvement in mechanical performance compared to the classical BCC model. The suitability of multi-objective and single-objective optimisation approaches for designing lattice structures was further investigated by comparing the corresponding designs in terms of their stiffness and maximum von-Mises stress values. The results from the numerical analysis and experimental testing demonstrate the significance of the application of an appropriate optimisation strategy for designing lattice structures for additive manufacturing.

Multi-objective distributionally robust approach for optimal location of renewable energy sources

Wind turbines and solar photovoltaic (PV) systems are intermittent and uncertain energy sources that disturb grid planning operations. In this paper, we establish a multi-objective distributionally robust optimization model for optimal locations of wind turbines and solar photovoltaics (PV) that minimize the variance of renewable energy sources and maximize power production. Moreover, this paper evaluates the accuracy of the Autoregressive Moving Average (ARMA), Deep Learning Gated Recurrent unit (GRU), and Deep Learning Long Short-Term Memory (LSTM) as forecasting models for wind speed and solar irradiation and compares their root mean square errors (RMSE). Using the forecasting error information, we characterize the uncertain variables in the ambiguity set, incorporating the bounds, means, and covariance values. Furthermore, we propose a modified multi-objective non-dominated sorting genetic algorithm (NSGA-II) approach to achieve a tractable Pareto front solution. To verify the effectiveness of the model, we use the actual candidate sites for wind turbines and solar photovoltaic (PV) systems in Saudi Arabia. The results demonstrate that our proposed model is an attractive and less conservative solution than a multi-objective robust optimization model when considering forecasting uncertainties.

Multi-objective optimization design of the Hinge Sleeve of Cubic based on Kriging.

In this study, the multi-objective optimal design of the Hinge Sleeve of Cubic (HSC) was achieved by combining the central composite design (CCD), Kriging and multi-objective genetic algorithm (MOGA) approaches. Firstly, the model of the HSC was established and the appropriate design variables were selected. The mass, the maximum deformation and the maximum equivalent stress of the HSC were taken as the optimization objectives. After comparative analysis of the parameters, the parameter with the greatest influence on the optimization objectives was selected as the geometric constraint. Subsequently, according to the results of the experimental design, the Kriging model was used to establish the response surface optimization model of the objective function. And finally the best optimization results were obtained by using MOGA. The experimental results show that the optimization strategy is reliable and the mass of the optimized model is reduced by 24.84%, which achieves the lightweight design of the HSC while meeting the actual production requirements, saves the design cost and improves the material utilization.

Parametric study and optimization of pillow-plate heat exchanger using multi-objective genetic algorithm and entropy generation minimization approaches

Parametric study and optimization of pillow-plate heat exchanger using multi-objective genetic algorithm and entropy generation minimization approaches

Formulation and characterization of a low-cost high-energy grain-based beverage premix: A multi-objective genetic algorithm approach

There is a huge demand for low cost healthy energy drinks amongst the consumers. A novel multigrain beverage premix (MGBP) was formulated using barley, pearl millet, quinoa, and yellow split pea by multi-objective genetic algorithm (MOGA) technique. The effects of processing like roasting and malting on nutritional compositions and correlations were studied using principal component analysis (PCA). Malting had significant effect on almost all nutritional components (p < 0.05). Roasting increased the total carbohydrate content by 67.7% in yellow split pea and 77.2% in barley flours. The Pearson correlation coefficient demonstrated a strong correlation of energy content with carbohydrate and fat content. The optimum combination obtained from MOGA and validated by fuzzy analysis was 30, 28, 30 and 12 g of roasted yellow pea flour (RYPF), roasted pearled barley flour (RPBF), malted pearl millet flour (MPMF) and malted quinoa flour (MQF), respectively per 100 g of MGBP with an energy value of 345.8 kcal and cost Rs. 4.88. The processing triggered the protein-starch interaction and agglomeration as confirmed by scanning electron microscopy (SEM); however, the chemical structure remained intact as revealed by Fourier transform infrared (FTIR) spectroscopy.

Optimizing pipe network design and central plant positioning of district heating and cooling System: A Graph-Based Multi-Objective genetic algorithm approach

• A graph-based multi-objective genetic algorithm for district energy system design. • Simultaneously optimize DHC pipe network layout and central plant positioning. • Improve valid offspring design generation and enable plant positioning optimization. • Six real-world case study to demonstrate outperformance of proposed work. District heating and cooling systems are one of the most promising technologies to support rapid decarbonization of our energy system as they enable significant energy and emissions savings. However, realizing such savings is highly dependent on the pipe network design and the central plant positioning. Today, the pipe network design and the central plant positioning still rely on expert experience and most of them are sub-optimal in terms of cost and efficiency performance. To tackle this issue, we propose a graph-based multi-objective genetic algorithm for simultaneous optimization of the pipe network design and the position of a central plant in a district heating and cooling system. Differing from a traditional genetic algorithm, our proposed algorithm uses a custom mutation and crossover technique to ensure the generation of valid offspring. The extra relocation step in the algorithm enables the optimization of the central plant location. Finally, we used the proposed algorithm to design the pipe network layout and decide the central plant location for six city districts in a metropolitan area in China. To assess and benchmark the performance of our proposed method, we utilized an expert-based approach in which several experienced engineers produced reference designs. Results demonstrate that our proposed algorithm outperforms the reference designs for all six cases. The proposed method was able to achieve a maximal pressure drop reduction of 37% and a maximal cost saving of 10% compared to the reference design.

Multi-objective optimization of µ-EDM parameters for µ-hole drilling on Hastelloy C 276 super alloy using Response Surface Methodology and multi-objective genetic algorithm

Hastelloy C 276 is one of the nickel-based super alloys which has an excellent resistance against corrosion at high temperature. It has a major application in nuclear and aerospace industry. Micro-electrical discharge machining (µ-EDM) is a highly accurate and precise non-conventional micro-machining process to fabricate this alloy. Thus, in this research paper an effort has been made to examine the impact of input process parameters like capacitance, gap voltage and pulse on time (t on ) on the response variables i.e., material removal rate (MRR), tool wear rate (TWR) and diametral overcut (DOC) during µ-EDM of Hastelloy C 276. Response Surface Methodology (RSM) is a collection of statistical and mathematical tools which determine the relationship between input and output characteristics. Box-Behnken design (BBD) is a special type of RSM based design that reduces the no of experiments without affecting the variation between the input and response parameters. RSM based BBD is used in the current study to develop the mathematical models for each response variable. RSM based desirability function (RSM-DF) which convert multiple responses into a scale free desirability, is used as an optimization tool in the present study. The current work used the multi-objective genetic algorithm (MOGA), a non-conventional searching based multi-response optimization algorithm, to optimize the machining parameters for maximum MRR along with minimum TWR and DOC. The lowest machining time is achieved by experiment no 10 and the least entry µ-hole diameter is drilled by experiment no 15. ANOVA analysis accomplished that capacitance has the highest significant contribution towards MRR, TWR and DOC. 3D response surface plot discloses that MRR, TWR and DOC is increased with capacitance until 5600 pF, 5600 pF and 8900 pF and then there is little reduction took place for all the three responses. The optimization result from RSM-DF is capacitance of 1001.560 pF, gap voltage of 137.6670 volt and t on of 41.532 µs for maximum MRR along with minimum TWR and DOC. MOGA approach reveals that maximum MRR of 0.005153 mm 3 /min along with minimum TWR of 0.002423 mm 3 /min and minimum DOC of 0.028691 mm is found at a capacitance of 1001.610 pF, gap voltage of 137.6671 volt and t on of 48.025 µs. Conformation experiments are conducted and the absolute percentage errors between the experimental and optimized values are found to be less than 5% for both approaches. Field emission scanning electron microscopy (FESEM) analysis reveals that MOGA settings drilled µ-hole with less DOC, more circularity and less surface cracks compared to the initial and RSM-DF parametric settings. A deposition of C and Cu took place in the side wall of the micro-hole from the dielectric fluid and tool material respectively.

A Decision Support System for Multi-Objective Porcelain Container Loading Problem Based on Genetic Algorithm

The multi-objective genetic algorithm approach for solving the Porcelain Container Loading Problem (PCLP) has a vital role in the global logistics industry. In this paper, a logistical problem with one constrained container loading problem that has to be filled with a set of boxes has been the focus. This study addresses a real-life problem that exports departments in the international porcelain industry face. Our objective is to maximize product profitability and delivery priority. Since the CLP is known as an NP-hard problem, the Genetic Algorithm (GA) approach is proposed to solve the problem based on these objective functions. The parameters of the GA affect the obtained results. We made tuning by using an experimental design in order to determine the appropriate parameters. The main contribution of the study is to present a new decision support system taking into account the objectives of the delivery time and profit rate priority of the manufacturer in the porcelain sector. Thus, loading according to the company’s priority and distribution in the shortest distance has been successfully achieved. The results show the efficiency of the proposed decision support system, which solves the CLP with up to 12 different products in boxes of different volumes.

A Multi-Objective Genetic Algorithms Approach for Modelling of Order Picking

A Multi-Objective Genetic Algorithms Approach for Modelling of Order Picking

Multi Objective Optimization of Machining Parameters in End Milling of AISI1020

In current research, artificial neural network (ANN) and Multi objective genetic algorithm (MOGA) have been used for the prediction and multi objective optimization of the end milling operation. Cutting speed, feed rate, depth of cut, material density and hardness have been considered as input variables. The predicted values and optimized results obtained through ANN and MOGA are compared with experimental results. A good correlation has been established between the ANN predicted values and experimental results with an average accuracy of 91.983% for material removal rate, 99.894% for tool life, 92.683% for machining time, 92.671% for tangential cutting force, 92.109% for power and 90.311% for torque. The MOGA approach has been proposed to obtain the cutting condition for optimization of each responses. The MOGA gives average accuracy of 96.801% for MRR, 99.653% for tool life, 86.833% for machining time, 93.74% for cutting force, 93.74% for power and 99.473% for torque. It concludes that ANN and MOGA are efficiently and effectively used for prediction and multi objective optimization of end milling operation for any selected materials before the experimental. Implementation of these techniques in industries before the experimentation is useful to reduce the lead time, experimental cost and power consumption also increase the productivity of the product.

Performance investigation and multi-objective optimization of helical baffle heat exchangers based on thermodynamic and economic analyses

The application of helical baffle heat exchangers (HBHXs) is attracting much more attention in both industrial and academic fields. In this paper, the heat transfer and fluid flow characteristics of HBHXs are explored by the aid of numerical and experimental analyses, and a good agreement is reached between the simulated and measured data. A series of helical baffle configurations involving parameter combinations of helix angles (10°, 15°, 20°, 25°, 30°, 35°, 40°, 45°) and axial overlapped ratios (0, 10%, 20%, 30%, 40%, 50%) are taken into account. Then a multi-objective genetic algorithm approach is proposed for HBHXs optimization design. The tube number, tube outer diameter, tube length, helix angle and axial overlapped ratio are perceived as design variables, and the revised entropy generation number and total cost are considered as objective functions. The results illustrate that the thermal enhancement and axial velocity reduction are attributed to diminished helix angle and enlarged axial overlapped ratio. For the sake of a narrow helix pitch, the baffle schemes of 45° helix angle together with no more than 20% axial overlapped ratio are given priority over the one of 40° helix angle, while the schemes of 40° helix angle as well as at most 30% axial overlapped ratio are preferable to the one of 30° helix angle in view of better shell-side comprehensive performance. A set of Pareto optimal solutions are obtained in order to trade off the thermodynamic and economic behaviors, and the optimized results make known that the multi-objective optimization method is more suitable compared with the single-objective optimization strategy.

A multi-objective genetic algorithm approach to design optimal zoning systems for freight transportation planning

This paper contributes to the existing research on freight transportation, spatial and land use planning by investigating an improved spatial aggregation technique to delineate desirable freight traffic analysis zones. Zoning is a process of spatially aggregating several predefined basic spatial units (BSUs) into multiple zones. It plays a vital role in the transportation planning and decision-making process and is well-documented as the modifiable areal unit problem (MAUP). MAUP involves aggregating BSUs to obtain optimal zones satisfying specific criteria and constraints. This paper proposes an improved spatial aggregation methodology to develop a freight traffic analysis zone system by applying the multiobjective optimization concept using a genetic algorithm. The decision variables, namely, (i) Freight trip density; (ii) Number of establishments; (iii) Employment density; and (iv) Compactness, are chosen to represent the elements of freight, passenger traffic, and land use. The problem is formulated as a multiobjective network partitioning problem. The four objectives aim to create zones with better homogeneity and compactness. It is solved using a genetic algorithm with a weighted distance metric approach to prioritize the four objectives. Results show that zones resulting from the improved methodology are superior to the existing zones in terms of homogeneity of decision variables and compactness. The findings are expected to help the decision-making process of urban, transportation, and land-use planners in selecting appropriate freight traffic zone delineation strategies for a given region.

Credit portfolio optimization: A multi-objective genetic algorithm approach

The algorithm for optimization of a credit portfolio has not been fully demonstrated. This paper fills the gap in the literature by presenting a general approach for optimizing a credit portfolio by minimizing the default risk of the entire portfolio. Default risk is measured with quadratic weighting and a matrix containing information about the default intensity of two stocks and the correlation in default between them. The default correlation and the default intensity are represented with a novel bivariate intensity model. A multi-objective genetic algorithm is introduced to optimize a credit portfolio with the purpose of overcoming limitations in the analytical method and improving the efficiency of optimization. The algorithm can be applied to a portfolio's credit risk management, which is particularly crucial for investors and regulars in emerging markets.