Adaptive Kriging-Based Optimization for Blended Wing Body UAV Design: Improving Conceptual Aircraft Development Through Surrogate Modeling

The conceptual design stage plays a decisive role in determining the overall performance, efficiency, and feasibility of an aircraft. For Blended Wing Body (BWB) Unmanned Aerial Vehicles (UAVs), designers must evaluate numerous interacting geometric, aerodynamic, structural, and flight-performance parameters before identifying an optimal configuration. Although multidisciplinary analysis (MDA) provides a comprehensive evaluation framework, repeatedly performing numerical simulations during optimization is computationally expensive. The reviewed study introduces an adaptive Kriging-based surrogate modeling framework that significantly improves optimization efficiency while maintaining prediction accuracy. By integrating multidisciplinary analysis with adaptive sampling strategies and statistical optimization, the proposed framework demonstrates how computational resources can be allocated more effectively during conceptual aircraft design. The research offers valuable insights into surrogate-assisted optimization for aerospace engineering and presents a practical methodology for accelerating the development of high-performance BWB UAVs.

Bibliographic Information

Item	Information
Article Title	Optimization in Conceptual Design of a BWB UAV Using an Adaptive Kriging-Based Surrogate Model
Authors	Xinshi Suo; Yu Wang; Zhouwei Fan; Tengzhou Xu
Journal	Engineering Reports
Volume	8
Issue	2
Publication Year	2026
Article Number	e70658
DOI	https://doi.org/10.1002/eng2.70658
Publisher	John Wiley & Sons Ltd.
License	Creative Commons Attribution (CC BY)
ISSN	2577-8196
Keywords	blended wing body; conceptual design; multidisciplinary analysis; sensitivity analysis; surrogate model; unmanned aerial vehicle

1. Research Background

Blended Wing Body (BWB) aircraft integrate the fuselage and wings into a continuous aerodynamic structure, offering several advantages over conventional aircraft configurations. The seamless geometry can improve lift-to-drag ratio, reduce structural weight, lower aerodynamic drag, decrease noise generation, and enhance fuel efficiency. While these characteristics make BWB aircraft attractive for future aerospace applications, their unconventional configuration also introduces substantial complexity during the conceptual design stage.

Unlike traditional aircraft, the aerodynamic, structural, mass-distribution, and flight-performance characteristics of a BWB UAV are strongly coupled. A modification to one design parameter may simultaneously influence several engineering disciplines, requiring multidisciplinary analysis (MDA) to evaluate each candidate configuration. Consequently, conceptual optimization becomes computationally demanding because thousands of numerical evaluations may be required before identifying an optimal design.

Previous studies have explored various surrogate modeling approaches—including Response Surface Methods, Support Vector Machines, Artificial Neural Networks, Polynomial Chaos Expansion, and Kriging—to reduce computational cost during engineering optimization. Among these methods, Kriging is particularly attractive because it not only predicts engineering responses but also quantifies prediction uncertainty, enabling adaptive sampling of the design space. The authors build upon this capability by developing an adaptive optimization framework that combines multidisciplinary analysis with Kriging-based surrogate modeling and sequential sampling strategies. Their objective is to improve optimization efficiency while preserving predictive accuracy throughout the conceptual design process.

The study develops a streamlined multidisciplinary analysis model integrating geometry generation, structural mass estimation, aerodynamic analysis, and flight-performance evaluation. Using this framework, three different optimization strategies are compared under an identical computational budget. Rather than relying solely on a static surrogate model, the proposed adaptive approaches dynamically refine the model through additional simulations, allowing computational resources to focus either on regions of high uncertainty or on regions with the greatest optimization potential. These strategies are statistically evaluated through repeated randomized optimization trials to determine their robustness and effectiveness. This scientific assessment is based exclusively on the uploaded research article.

2. Research Objective

To develop an adaptive Kriging-based surrogate modeling framework that improves optimization efficiency during the conceptual design of Blended Wing Body (BWB) Unmanned Aerial Vehicles.
To establish an integrated multidisciplinary analysis model combining geometry, structural mass estimation, aerodynamic evaluation, and flight-performance prediction.
To compare three optimization strategies consisting of a non-adaptive baseline approach, an exploration-oriented adaptive strategy, and an exploitation-oriented adaptive strategy under the same computational budget.
To evaluate the robustness of each optimization strategy through repeated randomized optimization experiments and statistical analysis.
To identify the most influential conceptual design parameters affecting the flight range of a BWB UAV using complementary sensitivity analysis methods.

3. Why This Research Matters

Accelerates conceptual aircraft design. The proposed framework substantially reduces the computational effort required during multidisciplinary optimization without sacrificing predictive accuracy.
Improves engineering decision-making. Adaptive surrogate modeling allows computational resources to be allocated strategically to the most informative regions of the design space.
Supports multidisciplinary aerospace engineering. By integrating geometry, mass estimation, aerodynamics, and flight-performance analysis into a unified workflow, the framework reflects the interconnected nature of modern aircraft design.
Enhances optimization reliability. The comparison of multiple adaptive sampling strategies, supported by repeated statistical validation, demonstrates a rigorous methodology for reducing dependence on initial sampling conditions.
Provides practical guidance for UAV designers. The parameter sensitivity analysis identifies the design variables with the greatest influence on flight range, helping engineers prioritize optimization efforts during early-stage development.
Contributes to computational engineering. The study illustrates how surrogate-assisted optimization can improve efficiency in engineering problems characterized by expensive numerical simulations and complex multidisciplinary interactions.
Offers a transferable optimization framework. Although demonstrated using a BWB UAV, the adaptive Kriging methodology has potential applications in other multidisciplinary engineering systems that require computationally intensive design optimization.

4. Research Methodology

The study employed a multidisciplinary engineering optimization framework that integrates conceptual aircraft design, multidisciplinary analysis (MDA), adaptive Kriging surrogate modeling, genetic algorithm optimization, and statistical validation. Rather than relying on repeated high-cost numerical simulations throughout the optimization process, the proposed framework constructs a surrogate model capable of accurately predicting aircraft performance while progressively improving its predictive capability through adaptive sampling. The entire methodology was designed to maximize the flight range of a Blended Wing Body (BWB) Unmanned Aerial Vehicle while maintaining a fixed computational budget.

Overall Research Framework

The optimization framework consists of five major stages:

Development of an integrated multidisciplinary analysis (MDA) model.
Construction of an initial Kriging surrogate model using Latin Hypercube Sampling.
Adaptive refinement of the surrogate model through sequential sampling.
Optimization using a Genetic Algorithm (GA).
Statistical validation and parameter sensitivity analysis.

This workflow enables expensive numerical simulations to be performed only when necessary while allowing the surrogate model to guide the optimization process efficiently.

Multidisciplinary Analysis (MDA)

The multidisciplinary analysis model forms the computational foundation of the optimization process. Instead of employing extremely high-fidelity simulations suitable for detailed aircraft design, the authors developed an engineering-oriented MDA framework appropriate for conceptual design, balancing computational efficiency with sufficient engineering accuracy.

The MDA model integrates four interconnected engineering modules:

Geometry Module – Automatically generates the BWB aircraft geometry using a parameterized representation of the center wing, inner wing, outer wing, and wingtip.
Mass Module – Estimates structural mass using a modified semi-empirical Howe method while calculating propulsion system, payload, subsystem, and fuel masses.
Aerodynamic Module – Computes lift and drag characteristics using a panel method combined with engineering drag estimation techniques.
Flight Performance Module – Calculates the theoretical maximum flight range by integrating aerodynamic characteristics, aircraft mass, propulsion performance, and fuel consumption throughout cruise flight.

This multidisciplinary framework allows every candidate aircraft configuration to be evaluated consistently before being incorporated into the optimization process.

Adaptive Kriging Surrogate Modeling

To reduce computational expense, the study employs a Kriging surrogate model based on Gaussian Process theory. Unlike conventional regression techniques, Kriging provides both a predicted response and an associated prediction uncertainty. This uncertainty information becomes the basis for adaptive sampling, enabling the optimization algorithm to determine where additional numerical simulations should be performed.

The adaptive sampling procedure creates a closed iterative cycle consisting of:

Model construction.
Prediction of candidate solutions.
Selection of additional sample points.
Execution of multidisciplinary analysis.
Surrogate model updating.

This iterative learning process progressively improves surrogate accuracy while minimizing unnecessary numerical simulations.

Optimization Strategies

Three optimization strategies were developed and compared using the same computational budget of eighty numerical simulations.

Strategy 1: Baseline Static Kriging

The first strategy constructs a surrogate model using eighty initial Latin Hypercube Sampling points and performs optimization directly without adaptive refinement. This approach serves as the benchmark against which the adaptive strategies are evaluated.

Strategy 2: Exploration-Oriented Adaptive Sampling

The second strategy begins with forty initial samples and allocates the remaining computational budget to regions exhibiting the greatest prediction uncertainty. New samples are selected where the surrogate model is least confident, thereby improving global model accuracy across the design space.

Strategy 3: Exploitation-Oriented Adaptive Sampling

The third strategy also starts with forty initial samples but concentrates additional simulations around regions predicted to contain high-performance solutions. Rather than exploring uncertain regions uniformly, this strategy continuously refines promising design areas identified by the optimization algorithm.

Optimization Algorithm

The optimization objective was to maximize the theoretical flight range of the BWB UAV. A Genetic Algorithm (GA) was employed as the global optimization technique operating on the Kriging surrogate model. Design variables included major geometric parameters together with payload mass and available fuel mass, all constrained within predefined engineering limits.

The optimization process maintained identical computational resources for all three strategies, ensuring a fair comparison of optimization efficiency and solution quality.

Statistical Validation

Recognizing the stochastic nature of optimization, each sampling strategy was independently repeated fifteen times using randomized initial samples. This experimental design enabled statistical comparison of optimization performance while minimizing the influence of random sampling bias.

The authors further employed two-tailed statistical significance testing to compare optimization results among the three strategies, providing quantitative evidence regarding the effectiveness of adaptive surrogate refinement.

Parameter Sensitivity Analysis

Following optimization, the study conducted parameter sensitivity analysis using two complementary analytical approaches. This analysis quantified the relative contribution of each design variable to aircraft flight range and identified the parameters that exert the greatest influence on optimization performance.

5. Key Findings

Adaptive Surrogate Modeling Improved Optimization Efficiency

The proposed adaptive Kriging framework successfully reduced the computational burden associated with conceptual aircraft optimization while maintaining high prediction accuracy. Instead of relying exclusively on repeated multidisciplinary simulations, the adaptive surrogate progressively learned the design space through strategically selected additional simulations.

Exploitation-Oriented Sampling Produced the Best Overall Performance

Among the three optimization strategies evaluated, the exploitation-oriented adaptive sampling strategy consistently produced the most reliable optimization results. Statistical comparisons demonstrated that concentrating computational resources near promising design regions generated better optimization performance than either static sampling or uncertainty-driven exploration.

The representative optimal configuration achieved approximately a 14.5% improvement in flight range compared with the baseline aircraft configuration.

High Prediction Accuracy Was Maintained

The optimized surrogate model demonstrated excellent agreement with multidisciplinary numerical simulations. For the representative optimal solution, the prediction error between the Kriging model and the numerical simulation was only 0.48%, indicating that the surrogate accurately represented the engineering response throughout the optimization process.

Adaptive Sampling Reduced Dependence on Initial Samples

The repeated statistical experiments showed that adaptive refinement substantially reduced optimization variability caused by different initial sample distributions. Consequently, adaptive sampling produced more robust optimization outcomes than the conventional one-shot surrogate modeling approach.

Fuel Mass Was the Most Influential Design Variable

Sensitivity analysis identified fuel mass as the single most influential parameter affecting the maximum flight range of the BWB UAV. Among the geometric variables, center wing span exhibited the strongest influence on aircraft performance, highlighting its importance during conceptual design.

Multidisciplinary Integration Improved Engineering Decisions

The integration of geometry generation, structural mass estimation, aerodynamic analysis, flight-performance prediction, surrogate modeling, adaptive sampling, and evolutionary optimization created a comprehensive design framework capable of supporting engineering decision-making during the conceptual development of BWB UAVs.

Adaptive Resource Allocation Enhanced Computational Efficiency

Although all three optimization strategies used the same total computational budget, adaptive allocation of numerical simulations significantly improved optimization quality. Rather than distributing computational effort uniformly, the proposed framework directed simulations toward the most informative or promising regions of the design space, increasing optimization efficiency without increasing computational cost.

6. Scientific Contribution

Introduces an adaptive surrogate-assisted optimization framework. The study integrates multidisciplinary analysis, adaptive Kriging modeling, and evolutionary optimization into a unified conceptual aircraft design methodology.
Demonstrates the effectiveness of adaptive sampling. The research provides quantitative evidence that adaptive surrogate refinement produces more robust optimization outcomes than conventional static surrogate modeling.
Provides a rigorous statistical comparison of optimization strategies. Multiple independent optimization trials and statistical hypothesis testing strengthen the reliability of the reported conclusions.
Advances conceptual BWB UAV design methodology. The proposed framework efficiently addresses the multidisciplinary coupling among aircraft geometry, structural mass, aerodynamics, and flight performance.
Identifies influential conceptual design parameters. The dual-method sensitivity analysis offers practical engineering guidance for prioritizing variables during early aircraft development.
Demonstrates transferable computational methodology. Although developed for Blended Wing Body UAVs, the adaptive surrogate modeling framework may be applied to many multidisciplinary engineering optimization problems involving expensive numerical simulations.

7. Industrial Implications

Accelerates aircraft conceptual design. Adaptive surrogate modeling substantially reduces computational time during preliminary aircraft development.
Supports aerospace digital engineering. The framework aligns with modern digital engineering practices that integrate simulation, optimization, and data-driven decision making.
Improves engineering productivity. Engineers can evaluate numerous conceptual configurations using significantly fewer multidisciplinary simulations.
Reduces development costs. Lower computational requirements translate into reduced engineering resources during conceptual aircraft optimization.
Enhances UAV performance. More efficient optimization enables designers to identify aircraft configurations capable of achieving greater flight range without increasing computational expenditure.
Facilitates multidisciplinary collaboration. The integrated framework supports coordinated evaluation across geometry, aerodynamics, structures, propulsion, and flight performance disciplines.
Provides a reusable optimization methodology. The adaptive Kriging framework can potentially be extended to the conceptual design of other aerospace systems, advanced aircraft configurations, and multidisciplinary engineering products requiring computationally intensive optimization.

8. Research Limitations

The study presents an efficient surrogate-assisted optimization framework for the conceptual design of Blended Wing Body (BWB) Unmanned Aerial Vehicles (UAVs). Nevertheless, as acknowledged by the research methodology and scope, several limitations should be considered when interpreting the findings and applying the proposed framework to broader engineering problems.

Conceptual design focus. The multidisciplinary analysis (MDA) model was developed specifically for the conceptual design stage. Consequently, several engineering aspects commonly addressed during preliminary and detailed design—such as structural verification, aeroelasticity, manufacturability, and certification requirements—were beyond the scope of this investigation.
Engineering-level analysis. The multidisciplinary framework combines engineering estimation methods with numerical analysis to achieve computational efficiency. While appropriate for conceptual optimization, the methodology does not replace the higher-fidelity analyses typically required during subsequent aircraft development stages.
Linear aerodynamic assumptions. The aerodynamic model relies primarily on a panel method supplemented by engineering drag estimation. As noted by the authors, this approach is suitable for cruise-condition conceptual design but cannot accurately capture highly nonlinear aerodynamic phenomena such as flow separation at high angles of attack.
Simplified flight mission. Flight performance evaluation focuses on cruise-range optimization. Fuel consumption during takeoff and landing is neglected, and the optimization objective is restricted to maximizing theoretical flight range rather than evaluating complete mission performance.
Limited optimization objective. The optimization considers only maximum flight range. Other important aircraft performance metrics—including endurance, maneuverability, stability, controllability, payload flexibility, operational robustness, and lifecycle performance—were not simultaneously optimized.
Restricted design variables. The optimization includes selected geometric parameters together with payload mass and fuel mass. Numerous additional aircraft design variables that may influence multidisciplinary performance remain outside the optimization framework.
Computational budget constraint. All optimization strategies were intentionally evaluated using an identical computational budget of eighty multidisciplinary simulations. Although this facilitates objective comparison among the three adaptive strategies, different computational budgets may produce different optimization behaviors.
Configuration-specific validation. The case study investigates a single conceptual BWB UAV configuration. Additional validation involving different aircraft sizes, mission profiles, propulsion systems, and configuration classes would further demonstrate the general applicability of the proposed methodology.

9. Future Research Opportunities

The proposed adaptive Kriging framework establishes a strong foundation for future research in multidisciplinary aerospace optimization. Several opportunities exist to extend the methodology and further improve conceptual aircraft design.

Integrate higher-fidelity Computational Fluid Dynamics (CFD) simulations into the adaptive surrogate framework to improve aerodynamic prediction accuracy during later design stages.
Develop multi-fidelity surrogate models that combine engineering estimation methods, panel methods, and high-fidelity numerical simulations within a unified optimization environment.
Expand the multidisciplinary framework to include structural optimization, aeroelastic analysis, vibration characteristics, thermal management, manufacturing constraints, and aircraft certification considerations.
Perform multi-objective optimization involving flight range, endurance, payload capacity, fuel efficiency, aerodynamic efficiency, structural weight, operating cost, and environmental performance simultaneously.
Investigate additional adaptive sampling strategies, Bayesian optimization techniques, active learning algorithms, and machine learning approaches for surrogate-assisted optimization.
Evaluate the framework using different classes of unmanned aircraft, conventional aircraft configurations, hybrid-wing-body concepts, and other advanced aerospace systems.
Conduct experimental validation or higher-fidelity numerical verification of optimized conceptual designs to further assess surrogate prediction accuracy.
Investigate adaptive optimization under uncertainty by incorporating manufacturing tolerances, operational variability, and uncertain environmental conditions into the design process.
Explore automated digital engineering workflows that integrate parametric geometry generation, multidisciplinary analysis, artificial intelligence, and cloud-based optimization for next-generation conceptual aircraft design.

10. Potential for Public Policy Citation

Although this study focuses primarily on aerospace engineering methodology, its findings have broader relevance for public policies supporting advanced manufacturing, digital engineering, and sustainable aviation technologies. Efficient surrogate-assisted optimization reduces computational resources required during aircraft development, supporting national strategies that encourage digital transformation and innovation in engineering design.

The proposed framework also aligns with governmental initiatives promoting intelligent manufacturing, simulation-driven engineering, and advanced computational technologies for high-value industrial sectors. By enabling more efficient conceptual design of fuel-efficient aircraft configurations, the methodology may indirectly contribute to policies aimed at reducing energy consumption and improving environmental sustainability within future aerospace systems.

Furthermore, the research provides an example of how artificial intelligence-inspired surrogate modeling and adaptive optimization can be integrated into engineering workflows, offering useful references for research funding agencies, aerospace research institutions, technology development programs, and organizations responsible for promoting digital engineering capabilities.

11. Who Should Read This Paper?

Aerospace engineers involved in conceptual aircraft design and multidisciplinary optimization.
Researchers working in surrogate modeling, computational optimization, and engineering design automation.
Aircraft designers developing Blended Wing Body (BWB) configurations or advanced unmanned aerial vehicles.
Graduate students studying aerospace engineering, computational engineering, multidisciplinary analysis, or design optimization.
Researchers interested in adaptive sampling, Gaussian Process regression, and Kriging surrogate models.
Professionals developing digital engineering, model-based systems engineering, and simulation-driven product development workflows.
Scientists working on artificial intelligence applications in engineering optimization.
Government research organizations and industrial innovation centers supporting advanced aerospace technologies.

12. Final Thoughts

This study presents a carefully designed adaptive surrogate-assisted optimization framework for the conceptual design of Blended Wing Body Unmanned Aerial Vehicles. By integrating multidisciplinary analysis, adaptive Kriging surrogate modeling, evolutionary optimization, and statistical validation, the authors demonstrate that optimization efficiency can be substantially improved without sacrificing prediction accuracy.

One of the study's principal strengths lies in its rigorous comparison of three optimization strategies under identical computational budgets. Rather than relying on a single optimization experiment, the repeated randomized trials and statistical analyses provide convincing evidence that exploitation-oriented adaptive sampling produces more reliable optimization performance than conventional static surrogate modeling or uncertainty-driven exploration alone.

Equally important is the study's contribution to engineering methodology. The proposed framework illustrates how surrogate models can intelligently allocate computational resources during multidisciplinary optimization, enabling engineers to evaluate complex conceptual aircraft designs more efficiently. The dual sensitivity analysis further enhances the practical value of the research by identifying the design parameters that most strongly influence flight range.

Although developed specifically for Blended Wing Body UAVs, the underlying optimization philosophy extends beyond aerospace engineering. The integration of adaptive surrogate modeling, multidisciplinary simulation, and statistical optimization provides a transferable computational framework that may benefit numerous engineering disciplines where expensive numerical simulations limit design exploration.

Overall, this research represents a valuable contribution to conceptual aircraft design methodology and demonstrates how adaptive surrogate-assisted optimization can support faster, more reliable, and more computationally efficient engineering decision-making during the early stages of aerospace product development.

13. Suggested Citations

UNP–Teknomekanik Style

Suo, X., Wang, Y., Fan, Z., & Xu, T. (2026). Optimization in Conceptual Design of a BWB UAV Using an Adaptive Kriging-Based Surrogate Model. Engineering Reports, 8(2), e70658. https://doi.org/10.1002/eng2.70658

APA 7th Edition

Suo, X., Wang, Y., Fan, Z., & Xu, T. (2026). Optimization in conceptual design of a BWB UAV using an adaptive Kriging-based surrogate model. Engineering Reports, 8(2), e70658. https://doi.org/10.1002/eng2.70658

IEEE

X. Suo, Y. Wang, Z. Fan, and T. Xu, "Optimization in Conceptual Design of a BWB UAV Using an Adaptive Kriging-Based Surrogate Model," Engineering Reports, vol. 8, no. 2, Art. no. e70658, 2026, doi:10.1002/eng2.70658.

Harvard

Suo, X., Wang, Y., Fan, Z. & Xu, T. 2026, 'Optimization in Conceptual Design of a BWB UAV Using an Adaptive Kriging-Based Surrogate Model', Engineering Reports, vol. 8, no. 2, e70658. https://doi.org/10.1002/eng2.70658

Vancouver

Suo X, Wang Y, Fan Z, Xu T. Optimization in Conceptual Design of a BWB UAV Using an Adaptive Kriging-Based Surrogate Model. Engineering Reports. 2026;8(2):e70658. doi:10.1002/eng2.70658.

Chicago Author–Date

Suo, Xinshi, Yu Wang, Zhouwei Fan, and Tengzhou Xu. 2026. "Optimization in Conceptual Design of a BWB UAV Using an Adaptive Kriging-Based Surrogate Model." Engineering Reports 8 (2): e70658. https://doi.org/10.1002/eng2.70658.

MLA 9th Edition

Suo, Xinshi, et al. "Optimization in Conceptual Design of a BWB UAV Using an Adaptive Kriging-Based Surrogate Model." Engineering Reports, vol. 8, no. 2, 2026, article e70658. Wiley, https://doi.org/10.1002/eng2.70658.

14. Editorial Note

This article is an independent scholarly review prepared for Engineering Research Insights. The review summarizes and interprets the scientific contributions of the published research while preserving the original findings reported by the authors. The scientific discussion presented here is intended to improve accessibility for researchers, graduate students, engineers, educators, and industry practitioners interested in multidisciplinary aircraft optimization and surrogate-assisted engineering design.

Readers are strongly encouraged to consult and cite the original peer-reviewed article when conducting research, preparing academic manuscripts, or developing related engineering applications. Proper citation of the original publication acknowledges the authors' contributions and supports responsible scholarly communication.

15. SEO Meta Description

Comprehensive review of Optimization in Conceptual Design of a BWB UAV Using an Adaptive Kriging-Based Surrogate Model, highlighting multidisciplinary analysis, adaptive Kriging surrogate modeling, aircraft conceptual design, optimization strategies, sensitivity analysis, key findings, engineering contributions, industrial implications, and future research opportunities.

16. SEO Keywords

Adaptive Kriging, Kriging surrogate model, surrogate modeling, blended wing body UAV, BWB aircraft, conceptual aircraft design, multidisciplinary analysis, multidisciplinary optimization, UAV optimization, aircraft optimization, aerospace engineering, adaptive sampling, Gaussian process regression, genetic algorithm, engineering optimization, flight range optimization, conceptual UAV design, digital engineering, Engineering Reports review, Engineering Research Insights.

17. Recommended URL Slug

optimization-conceptual-design-bwb-uav-adaptive-kriging-surrogate-model-review

18. Original Research Article

Official Publisher Page:
https://onlinelibrary.wiley.com/doi/10.1002/eng2.70658

DOI:
https://doi.org/10.1002/eng2.70658

19. Copyright and Fair Use Statement

This review is an original editorial work created for educational and scholarly communication purposes. It does not reproduce substantial portions of the original article. Scientific interpretations are based exclusively on the published research paper, while bibliographic metadata have been verified against the official publisher's webpage. Copyright of the original research article remains with the respective authors and the publisher. Readers should access the original publication through the official Wiley website for the complete scientific manuscript.

20. Closing Remarks

Adaptive surrogate-assisted optimization is becoming an increasingly important component of modern aerospace engineering. This study demonstrates how adaptive Kriging modeling, multidisciplinary analysis, and evolutionary optimization can be integrated into a computationally efficient framework that supports conceptual aircraft design while maintaining excellent predictive accuracy. Beyond its application to Blended Wing Body Unmanned Aerial Vehicles, the proposed methodology illustrates broader opportunities for accelerating multidisciplinary engineering optimization in computationally intensive design environments. As digital engineering and intelligent optimization continue to evolve, frameworks such as the one presented in this study are expected to play an increasingly significant role in the development of next-generation aerospace systems.