Non-dominated Sorting Genetic Algorithm II Research Articles

A bi-objective approach for the multi-skilled worker assignment of a hybrid assembly line-seru production system

The flexibility and responsiveness of seru production have caught the attention of manufacturing and electronics industries. However, multi-skilled worker assignment poses a crucial and challenging decision-making problem for seru production systems. The existing literature on this problem for pure seru production systems primarily focuses on improving efficiency indexes, which often leads to an unbalanced workload among workers. To address this issue, this article investigates multi-skilled worker assignment for a hybrid assembly line-seru production system that comprises divisional serus and a short assembly line. To balance workload and optimize production efficiency, a bi-objective integer nonlinear programming model is developed. This model jointly optimizes worker-to-seru, worker-to-line, batch-to-seru, task-to-worker, and the processing sequence of each batch. A meta-heuristic method, combining Non-Dominated Sorting Genetic Algorithm II (NSGA-II) with Multi-Objective Simulated Annealing (MOSA), NSGA-II-MOSA, is designed to solve the model. The results of numerical experiments demonstrate that the proposed model and solving method can greatly reduce workload imbalance while maintaining production efficiency. Moreover, NSGA-II-MOSA provides better Pareto solutions than three well-known multi-objective optimization approaches.

Coupled and collaborative optimization model of impervious surfaces and drainage systems from the flooding mitigation perspective for urban renewal

Urban pluvial flooding mitigation is a significant challenge in city development. Many mature methods have been used to reduce the risk of flood. The optimal design of impervious surfaces (ODIS) is an adaptive solution to urban flooding from the perspective of urban renewal planning. However, existing ODIS models are limited because they do not consider the drainage systems. To address this issue, this study proposes an elastic and controllable optimization model based on assumptions about rainstorm and drainage capacity, nondominated sorting genetic algorithm-II (NSGA-II), multivariate linear programming (MLP) and soil conservation service curve number model (SCS-CN) in a case study of the old town of Guangzhou city, China. The model not only coupled the drainage systems, but also collaboratively optimized the impervious surfaces and the drainage systems. The results show that the proposed model achieved an optimized efficiency of 5.70 %, which is more than a tenfold improvement compared to existing ODIS models. The study emphasizes that the optimization of the drainage system should be the focus and the optimization of impervious surfaces should be supplementary, and different flood risk areas require different optimization strategies. Furthermore, transforming impervious surfaces into a “high-low-high” spatial distribution of impervious surface densities is the optimal design solution for impervious surfaces. In general, this study offers a novel perspective and approach to urban flooding mitigation, enabling comprehensive control of flooding from a global perspective.

Optimized design of small sized low noise magnetic shielding cylinder

Magnetic shielded cylinder (MSC) is crucial in various applications, especially in the realm of medical weak magnetic signal measurement. However, designing a small volume, low noise MSC is a multi-objective optimization problem, with the objectives conflict with each other. To address the complexity arising from numerous variables in the optimal design of MSC with a small volume and low-noise, this paper proposes an optimal design method based on non-dominated sorting genetic algorithm II (NSGA-II) to realize the optimal design of MSC with a high shielding factor, small volume and low-noise. Firstly, an analytical model for the shielding factor of the MSC is established, and a multi-objective optimization model is given. Then, the NSGA-II algorithm is employed to obtain a series of Pareto frontier optimal solutions. Finally, the efficacy of this method is evaluated through simulation and experimentation. The results reveal that the radial and axial shielding factors of the optimized MSC achieve maximum values of 8.6 × 107 and 9.5 × 106 respectively. Meanwhile, the average axial noise and radial noise of the MSC between 1 and 100 Hz are 22 fT/Hz1/2 and 20 fT/Hz1/2 respectively. Practice has proven that this method can be used to optimize the design of small-volume, low-noise magnetic shields to improve the performance of magnetic shields.

Numerical Analysis and Validation of an Optimized B-Series Marine Propeller Based on NSGA-II Constrained by Cavitation

Constantly growing environmental concerns focused on reducing pollution, in addition to rising fuel costs in recent years, have led the maritime industry to develop and implement fuel-saving solutions. Among them is the optimization of marine propeller efficiency, as marine propellers are a crucial part of ship’s propulsion system. During the initial design stage, selecting the optimal propeller is considered a multi-objective optimization process. This research focused on maximizing propeller open water efficiency, while minimizing engine brake power constrained by thrust and cavitation. Optimization was applied to Wageningen B-series propellers and conducted using the Non-Dominated Sorting Genetic Algorithm II (NSGA-II). The algorithm selected optimum parameters to create the propeller model, which was then evaluated numerically through computational fluid dynamics (CFD) with a multiple reference frame (MRF) and under the SST k-ω turbulence model, to obtain the open water hydrodynamic characteristics. In addition, the cavitation effect was evaluated using the Zwart–Gerber–Belamri cavitation model. The numerical model results were validated through comparison with open water experimental data from the Netherlands Ship Model Basin for the Wageningen B-series propellers. The results showed errors of 3.29% and 2.01% in efficiency under noncavitating and cavitating conditions, respectively. Correct performance of the functions was shown, based on neural networks trained to estimate thrust and torque coefficients instead of polynomials. The proposed optimization process and numerical model are suitable for solving multi-objective optimization problems in the preliminary design of fixed-pitch marine propellers.

Predicting and optimizing reactive oxygen species metabolism in Punica granatum L. through machine learning: role of exogenous GABA on antioxidant enzyme activity under drought and salinity stress

BackgroundDrought and salinity stress have been proposed as the main environmental factors threatening food security, as they adversely affect crops' agricultural productivity. As a potential solution, the application of plant growth regulators to enhance drought and salinity tolerance has gained considerable attention. γ-aminobutyric acid (GABA) is a four-carbon non-protein amino acid that accumulates in plants as a response to stressful conditions. This study focused on a comparative assessment of several machine learning (ML) regression models, including radial basis function, generalized regression neural network (GRNN), random forest (RF), and support vector regression (SVR) to develop predictive models for assessing the effect of different concentrations of GABA (0, 10, 20, and 40 mM) on various physio-biochemical traits during periods of drought, salinity, and combined stress conditions. The physio-biochemical traits included antioxidant enzyme activities (superoxide dismutase, SOD; peroxidase, POD; catalase, CAT; and ascorbate peroxidase, APX), protein content, malondialdehyde (MDA) levels, and hydrogen peroxide (H2O2) levels. The non‑dominated sorting genetic algorithm‑II (NSGA‑II) was employed for optimizing the superior prediction model.ResultsThe GRNN model outperformed the other ML algorithms and was therefore selected for optimization by NSGA-II. The GRNN-NSGA-II model revealed that treatment with GABA at concentrations of 20.90 mM and 20.54 mM, under combined drought and salinity stress conditions at 20.86 and 20.72 days post-treatment, respectively, could result in the maximum values for protein content (by 0.80 and 0.69), APX activity (by 50.63 and 51.51), SOD activity (by 0.54 and 0.53), POD activity (by 1.53 and 1.72), CAT activity (by 4.42 and 5.66), as well as lower MDA levels (by 0.12 and 0.15) and H2O2 levels (by 0.44 and 0.55), respectively, in the ‘Atabaki’ and ‘Rabab’ cultivars.ConclusionsThis study demonstrates that the GRNN-NSGA-II model, as an advanced ML algorithm with a strong predictive ability for outcomes in combined stressful environmental conditions, provides valuable insights into the significant factors influencing such multifactorial processes.

Multi-objective optimization of micro planar combustor with tube outlet by RSM and NSGA-II for thermophotovoltaic applications

The pressure loss (Δp), output power (Qtot) and system efficiency (ηMTPV) are the most important performance parameters of micro combustor for thermophotovoltaic applications. However, the above performance paremetres are conflicted in the optimization of micro planar combustor with tube outlet for thermophotovoltaic applications. In this work, RSM (Response Surface Methodology) and Non-dominated Sorting Genetic Algorithm II (NSGA-II) is combined to make a trade-off among the above performance parameters. With the inlet velocity (vin) and equivalence ratio (φ), tube length (L9), tube diameter (D) and tube distance (L10) selected as design variables, central composite design is used to arrange simulation cases. The regression model about Δp is quadratic, while the regression models about Qtot and ηMTPV are linear. Then, the significance of each term in the regression models is determined and arranged by variance (ANOVA). Fianlly, according to the Pareto optimal solution obtained from NSGA-II, the optimal values of the objective function are Δp = 977.87 Pa, Qtot = 6.59 W and ηMTPV = 2.72 %, and the corresponding design variables are vin = 7.09 m/s, φ = 0.99, L9 = 6.71 mm, D = 1.75 mm and L10 = 3.97 mm, respectively.

An LSTM-stacked autoencoder multisource response prediction and constraint optimization for scaled expansion tubes

Deep learning is attracting increasing attention due to its excellent predictive power and its ability to be applied in traditional research areas. In this paper, we propose a multisource response prediction network architecture based on a long short-term memory (LSTM)-stacked autoencoder to predict key crashworthiness indicators and complete curve reconstruction. Taking an expansion tube as an example using an equivalent scaling research method, a scaled expandable tubular (SET) finite element model was established and verified using a quasistatic compression test and a full-size coupler and buffer system experiment. A design of experiments (DOE) approach was used to obtain a dataset for training the prediction network. Neural network hyperparameters are critical to network prediction accuracy, and after comparison, the multisource response prediction network architecture showed good computational efficiency and satisfactory prediction accuracy when appropriate hyperparameters were selected. Subsequently, multiobjective constraint optimization was performed using the nondominated sorting genetic algorithm-II (NSGA-II) based on a prediction network architecture, which greatly improved the energy-absorption structure optimization accuracy. The results are expected to provide a research methodology for solving complex engineering problems by establishing a new framework for deep learning algorithms combined with optimization methods.

Strategy and capacity optimization of renewable hybrid combined cooling, heating and power system with multiple energy storage

Combined cooling, heating, and power systems offer significant potential for integration with renewable energy sources, such as solar and geothermal energy, alongside energy storage devices. However, the effectiveness and feasibility of these systems depend crucially on the operational strategies and capacity planning for each component. Therefore, an innovative hourly dynamic simulation model that integrates solar energy, geothermal energy, and multiple energy storage devices is established. Using the multi-objective comprehensive evaluation method that combines Technique for Order Preference by Similarity to Ideal Solution methods and non-dominated sorting genetic algorithm-II including elite strategy, establish five dimensions of multi-objective optimization model, including energy-saving, economy, environmental protection, system independence and renewable energy utilization scale. Meanwhile, the system and reference system in a community in Beijing are used as models to analyze the equipment capacity under the three operating strategies. The findings reveal that, compared to the reference system, the new system attains optimal performance when following the electrical load strategy. Its comprehensive index reaches 69%, surpassing the thermal load and hybrid load strategies by 8% and 11%, respectively. When the target is set as “independence-renewable energy-cost,” the utilization rate of solar energy and geothermal energy reaches an impressive 79%, showcasing excellent renewable energy utilization. The presented model holds substantial significance for advancing research in the realms of operating strategies and equipment capacity planning within such systems.

Research on decentralized resource operation optimization of virtual power plant with 5G base station

Abstract The extensive construction and promotion of 5G base stations (5GBSs) have led to a surge in communication energy consumption, as 5G energy consumption is about three to five times that of 4G. To reduce the energy consumption of 5GBS, this article incorporates 5GBS into power demand side management and proposes a flexible resource collaborative optimization method that integrates 5GBS with virtual power plants (VPPs). Firstly, starting from the basic structure of 5GBS, the adjustable potential of 5GBS and the paths for participating in VPP optimization were analyzed. Secondly, with the minimum operating cost as the optimization objective and considering the constraints of 5GBS and various decentralized resources, a demand side decentralized resource collaborative optimization model for VPP integrating 5GBS was established. This model can be solved using the Non-dominated sorting genetic algorithm II (NSGA-II) algorithm. The results indicate that the participation of 5GBS in power demand side management can reduce their comprehensive energy consumption and operating costs.

Multi-objective optimization of heat exchanger network with disturbances based on graph theory and decoupling

Disturbance poses a significant challenge to highly coupled chemical systems. In this work, a multi-objective optimization framework is developed for reducing the impact of fluctuations on heat exchange networks (HENs) based on graph theory and decoupling. The framework bridges the gap between algebra and structure of HEN through a directed graph (Digraph) and defines the Excitation as a fluctuation impact indicator. The Non-dominated Sorting Genetic Algorithm II (NSGA-II) is improved for generating feasible solutions and global optimality, and the quasi-gradient is applied to select an equilibrium solution from the Pareto Front. Two case studies are conducted, and the equilibrium solution is selected based on quasi-gradient. For the literature case, the Excitation of the equilibrium solution and the total annual cost is reduced by 56.15% and 2.12%, respectively. For the coal-to-methanol system, the utility consumption and Excitation are 29.6% and 264.50% lower than the results of the Aspen Energy Analyzer, respectively.

Group technology empowering optimization of mixed-flow precast production in off-site construction.

Faced with immense pressure to reduce environmental impact, off-site construction (OSC) is considered a sustainable alternative to conventional practices. However, challenged by component diversity and a significant surge in demand, deficient or empirical-based scheduling management struggles to effectively harness the potential of mixed-flow precast production to improve efficiency, instead resulting in environmental impacts, and falling short of expected benefits in OSC projects. Therefore, this study addresses the conflict between efficiency and environmental impact arising from the application of mixed-flow precast production by integrating multi-objective optimization and group technology. A multi-objective optimization framework is proposed, incorporating grouping technology for mixed-flow precast production scheduling and aiming to minimize carbon emissions and reduce tardiness/earliness penalty. The non-dominated sorting genetic algorithm II (NSGA-II), adjusted by adaptive population initialization strategy and group technology, is introduced to solve this problem, striking a balance between sustainability and penalty costs. Through a real-case analysis, the proposed approach demonstrates an average reduction of 37.5% in carbon emissions compared to rule-based scheduling methods, a 30.1% reduction compared to previous research methods, along with over 10% reduction in tardiness/earliness penalty. This study enhances environmental benefits and efficiency from a production scheduling perspective and establishes an automated, practical method, fostering low-cost, high-efficiency green production for construction component enterprises, particularly for small and medium-sized enterprises, thereby promoting sustainable development in the construction industry.

Design of a wind-PV system integrated with a hybrid energy storage system considering economic and reliability assessment

Hybrid energy systems (HESs) have garnered significant attention as a sustainable solution to meet the world's growing energy demands while minimizing environmental impact. Achieving cost-effective and resilient HES solutions requires considering economic and reliability factors for optimal design. This research delves into the optimization and design of a wind-PV system integrated with a hybrid energy storage system using the Multi-Objective African Vultures Optimization Algorithm (MOAVOA) in both standalone and grid-connected modes. A comprehensive case study is conducted in Tabuk, Saudi Arabia, focusing on a microgrid and incorporating wind speed, solar radiation, and load data. The effectiveness of MOAVOA is verified using the Non-Dominated Sorting Genetic Algorithm-II (NSGA-II) and Multi-Objective Particle Swarm Optimization (MOPSO) methods in achieving the Pareto front of optimal system designs. Then, the mean, entropy, and criteria importance using the inter-criteria correlation (CRITIC) methods are utilized to assign weights to each objective in the obtained Pareto front. Finally, the Technique for Order Preference by Similarity to the Ideal Solution (TOPSIS) method is employed to select the optimal system design from the Pareto front. The results underscore the significance of the optimization approach in assessing the cost, dependability, and ideal configurations of the system. Furthermore, they highlight the diverse outcomes produced by different optimization techniques. While the entropy method yields more cost-effective designs, the mean and CRITIC methods produce comparable results. However, the mean and CRITIC approaches exhibit higher levels of reliability. For instance, in the grid-connected mode, the entropy weighting method yields an optimal system design with a total cost of M$ 42.2 and a reliability index of 91.73 %, while using the mean method, the selected system configuration has a total cost of M$ 42.98 and a reliability index of 92.89 %. The study emphasizes the benefits of diversifying renewable resources by considering different scenarios involving wind and solar generation. For example, in the wind-PV grid-connected system, the total cost is 22.65 % less than in the PV-only grid-connected system with a higher system reliability. The findings provide valuable guidance for system designers in selecting optimal optimization techniques and promoting the integration of renewable energy sources in hybrid energy systems. By presenting a comprehensive framework that incorporates economic and reliability assessments through the utilization of MOAVOA, this study contributes to the field of HES design.

Linear Cutting Performance Tests and Parameter Optimization of Poplar Branches Based on RSM and NSGA-II

To reduce the cutting force and cutting energy consumption during the operation of tree-climbing pruning machines for poplar trees, a linear cutting test bench device for branches was designed according to the growth characteristics of poplar branches and pruning forestry requirements in this study. Firstly, the cutting mechanical analysis of poplar branches was carried out to explore the significance parameters affecting the cutting force, and then the cutting performance test and parameter optimization of poplar branches was carried out through the response surface method (RSM). The test results indicated that cutting speed, tool edge angle and tool back angle had significant effects on the ultimate shear stress, cutting power consumption per unit area, and the branch damage rate of poplar branches, and the established regression equation demonstrated high goodness of fit. Meanwhile, a second-order regression mathematical model was developed between ultimate shear stress, cutting the power consumption per unit area of the cut and the branch damage rate, and the significance parameter. The non-dominated Sorting Genetic Algorithm II (NSGA-II) was used for multi-objective optimization computation to obtain the optimal combination of cutting parameters as 3.02 m/s for cutting speed, 15° for tool edge angle, and 3° for tool back angle. In this case, the ultimate shear stress, cutting power consumption per unit area, and branch damage rate of poplar branches were small, which were 346.63 kPa, 9.35 mJ/mm2, and 12.36%, respectively. Through the test verification, it can be seen that the relative error between the verification test and the predicted value of model was less than 7%. Moreover, under a cutting tool edge angle of 15°, the ultimate shear stress, cutting power consumption per unit area, and branch damage rate were, respectively, reduced by 17.29%, 14.98%, and 34.21% compared with those under a cutting tool edge angle of 20°, which verifies the validity and reliability of the test results and the research method. This study can provide data support and reference for the research and development of energy-efficient poplar tree-climbing pruning equipment and related branch-cutting performance tests.

A new integrated remanufacturing process planning and scheduling model under uncertainties using extended non-dominated sorting genetic algorithm-II

Remanufacturing, with its environmental and economic implications, is gaining significant traction in the contemporary industry. Owing to the complementarity between remanufacturing process planning and scheduling in actual remanufacturing systems, the integrated remanufacturing process planning and scheduling (IRPPS) model provides researchers and practitioners with a favorable direction to improve the performance of remanufacturing systems. However, a comprehensive exploration of the IRPPS model under uncertainties has remained scant, largely attributable to the high complexity stemming from the intrinsic uncertainties of the remanufacturing environment. To address the above challenge, this study proposes a new IRPPS model that operates under such uncertainties. Specifically, the proposed model utilizes interval numbers to represent the uncertainty of processing time and develops a process planning approach that integrates various failure modes to effectively address the uncertain quality of defective parts during the remanufacturing process. To facilitate the resolution of the proposed model, this study proposes an extended non-dominated sorting genetic algorithm-II with a new multi-dimensional representation scheme, in which, a new self-adaptive strategy, multiple genetic operators, and a new local search strategy are integrated to improve the algorithmic performance. The simulation experiments results demonstrate the superiority of the proposed algorithm over three other baseline multi-objective evolutionary algorithms.

Adaptive Joint Multiobjective Operating Parameters’ Optimization for Active Direct Methanol Fuel Cells

The operating parameters of the active direct methanol fuel cell (DMFC) are essential factors affecting its power delivery performance. Different operating parameters lead to variations in the amount of methanol crossover in a DMFC, which might cause overpotential and cathode catalyst poisoning. Due to the complexity of the DMFC system, changes in operating conditions, and correlations among these parameters, it is challenging to maintain output power density while reducing the negative effects of methanol crossover. This paper proposes an adaptive joint optimization method for fuel cell operating parameters. The principle operating parameters are selected by the orthogonal tests, which include an adaptive numerical simulation and multiobjective optimization regarding cell output power density and methanol crossover. The selected parameter combinations are verified by an evaluation model that quantifies the influences of operating parameters on the active DMFC power density and its methanol crossover, where the nonlinear mapping function for the two optimization objectives is obtained. The nondominated sorting genetic algorithm‐II (NSGA‐II) is applied to rapidly obtain the optimal combination. The results show that with the optimal parameters, the maximum power density is increased by 16.7% and the methanol crossover is reduced by 35.1%.

Design and multi-objective optimization of combined air separation and ORC system for harnessing LNG cold energy considering variable regasification rates

Regasification of liquefied natural gas (LNG) releases significant cooling potential, but improper usage can easily lead to wastage of resources and environmental pollution. The study developed an integrated air separation system and Organic Rankine Cycle (ORC) power generation system to effectively harness the inherent cold energy within LNG. Two distinct optimization algorithms, the Non-Dominated Sorting Genetic Algorithm II (NSGA-II) and the Particle Swarm Optimization (PSO), were employed to determine the optimal operational parameters. The results showed the superior performance of PSO over NSGA-II, as demonstrated by the exergy efficiency, net output power, and levelized cost of energy at 78.12 %, 2.29*1011 kW, and 0.0665 $/kWh, respectively. Additionally, an assessment of various factors influencing the system's performance was conducted. The performance of both conventional and proposed systems was studied under varying regasification rates. The findings indicated that incorporating the cold recovery and storage unit enhanced the efficient utilization of LNG cold energy, resulting in an increased valley-to-peak ratio of the total power output from 0.03 in the conventional system to 0.5. The proposed system exhibits alignment with the energy consumption demands of users, and this study is anticipated to furnish a theoretical foundation for harnessing the cooling capacity of liquid hydrogen.

NSGA-II Algorithm-Based Structural Parameters of Electric Pulse Rock-Breaking Electrode Bit Multi-Objective Optimization

Abstract High-voltage electric pulse (HVEP) electrode bit has a considerable influence on the drilling and rock-breaking (RB) efficiency. HVEP electrode bit was systematically studied to optimize the structural parameters in order to improve RB efficiency. This paper analyzed the impact of main structural parameters on electric field strength (EFS) and depth of penetration (DOP) during high-voltage electric pulse drilling. A structural optimization method integrating back propagation (BP) neural network and genetic algorithm for HVEP electrode bit was proposed. The method mapped the complex nonlinear relationships among electrode distance, electrode cone angle, electrode grounding span, etc., and EFS and DOP by establishing a BP neural network model, and adopted the non-dominated sorting genetic algorithm-II (NSGA-II) to optimize the main structural parameters. The simulation data showed that the combined BP neural network/non-dominated sorting genetic algorithm-II (BP-NSGA-II) was an effective tool for optimizing the injection molding process. The multi-objective optimization of the structural parameters of the HVEP electrode bit based on the NSGA-II algorithm was crucial to direct the choice of the process parameters of the HVEP electrode bit, boost the RB efficiency, and lower the energy loss during drilling.

Research on Multi-Objective Optimization of High-Speed Solenoid Valve Drive Strategies under the Synergistic Effect of Dynamic Response and Energy Loss

Under high-frequency operating conditions, the high-speed solenoid valve (HSV) experiences energy loss and heat generation, which significantly impacts its operational lifetime. Reducing the energy loss of an HSV without compromising its opening response characteristics poses a significant challenge. To address this issue, a finite element simulation model of an HSV coupled with a current feedback model is constructed to investigate the synergistic effects of dynamic response and energy loss. Prediction models for the opening response time, HSV driving energy, and Joule energy using a back propagation neural network (BPNN) are established. Furthermore, a multi-objective optimization study on the current driving strategy using a non-dominated sorting genetic algorithm II (NSGA-II) is conducted. After optimization, although there was a 6.24% increase in the opening response time, both HSV drive energy and Joule energy were significantly reduced by 15.67% and 22.49%, respectively. The proposed multi-objective optimization method for an HSV driving strategy holds great significance for improving its working durability.

Design and Maintenance Optimisation of Substation Automation Systems: A Multiobjectivisation Approach Exploration

Substation automation systems (SAS) are critical infrastructures whose design and maintenance must be optimised to guarantee a suitable performance. In order to provide a collection of solutions that balance availability and cost, this paper explores the optimisation of the design and maintenance of a section of SAS. Multiobjective evolutionary algorithms are combined with discrete event simulation while the performance of two state-of-the-art multiobjective evolutionary algorithms is studied. On the one hand, the nondominated sorting genetic algorithm II (NSGA-II), and on the other hand, the S-metric selection evolutionary multiobjective optimisation algorithm (SMS-EMOA). Such a problem is solved from 2 and 3-objective approaches by attending to the multiobjectivisation concept. The robustness of the methodology is brought to light, and benefits were observed from the multiobjectivisation approach. Decision-makers can employ this knowledge to make informed decisions based on economic and reliability criteria.

Deployment of battery-swapping stations: Integrating travel chain simulation and multi-objective optimization for delivery electric micromobility vehicles

As the delivery sector increasingly relies on electric micromobility vehicles (EMVs), the urgency for efficient battery-swapping infrastructure becomes critical. This study develops a comprehensive framework for predicting battery-swapping demand for delivery EMVs (DEMVs) based on an activity-based travel chain simulation model and devises a multi-objective optimization model for the strategic placement of battery-swapping stations. The simulation model integrates submodules such as the EMV generation and attraction model, Agent-based EMV travel chain, and EMV and battery-swapping behavior model to capture the nuanced travel patterns and battery-swapping demand. Leveraging the Non-dominated Sorting Genetic Algorithm II (NSGA-II), the study optimizes the network design of battery-swapping stations considering both construction and travel costs. A case study in Nanjing City, representative of the diverse delivery sector's operations, substantiates the simulation's accuracy, maps out the spatiotemporal distribution of swapping demand, and analyzes the Pareto optimal set derived from the optimization model. Sensitivity analysis focuses on the facility planning model, assessing how uncertainties in swapping demand and battery charging rates within stations impact operational efficacy. This research melds demand forecasting with infrastructure optimization, providing actionable insights for the planning and management of battery-swapping stations.