Yongxiang Li
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Yongxiang Li has a total of 4 co-authors. Here are the top co-authors by number of shared publications:
Kwok-Leung Tsui
Department of Industrial Engineering and Management Department of Statistics and Data Science Department of Biostatistics Department of Industrial, Manufacturing, and Systems Engineering Department of Industrial Engineering and Manage- ment Shanghai Jiao Tong University University of California at Los Angeles The University of Texas at Arlington Shanghai Jiao Tong University 200240 Shanghai
1 shared publicationXiaowu Dai
Department of Industrial Engineering and Management Department of Statistics and Data Science Department of Biostatistics Department of Industrial, Manufacturing, and Systems Engineering Department of Industrial Engineering and Manage- ment Shanghai Jiao Tong University University of California at Los Angeles The University of Texas at Arlington Shanghai Jiao Tong University 200240 Shanghai
1 shared publicationYu Tan
Department of Industrial Engineering and Management Department of Statistics and Data Science Department of Biostatistics Department of Industrial, Manufacturing, and Systems Engineering Department of Industrial Engineering and Manage- ment Shanghai Jiao Tong University University of California at Los Angeles The University of Texas at Arlington Shanghai Jiao Tong University 200240 Shanghai
1 shared publicationZiyu Liu
Shanghai Jiao Tong University
1 shared publication
Latest Publications
- Open Access
- 1 May 2026
- Published by IOP Publishing in Journal of Physics: Conference Series
- Vol. 3247 (1)
B-spline empowered functional Bayesian optimization for composite fuselage shape control
Composite fuselage shape control is a representative black-box optimization problem, where the fuselage shape is a high-dimensional functional response with strong spatial correlations and each evaluation is costly. Existing functional Bayesian optimization methods for this problem often suffer from prohibitive computational burdens due to high dimensionality and inadequate representations of spatial correlations in fuselage shapes. To address these limitations, this study proposes a B-spline empowered functional Bayesian optimization (BFBO) method, in which fuselage shapes are represented using low-dimensional B-spline control points and their correlations are modeled using a multivariate Gaussian process (MGP). Based on the predictive distribution of control points, a generalized chi-square expected improvement criterion is derived to guide the optimization process. Numerical results show that BFBO achieves improved performance in terms of convergence speed and optimization accuracy, demonstrating its effectiveness for fuselage shape control and its potential as a practical optimization framework for the design and manufacturing of engineering systems involving functional responses.
- 1 May 2026
- Published by IOP Publishing in Journal of Physics: Conference Series
- Vol. 3247 (1)
Open AccessB-spline empowered functional Bayesian optimization for composite fuselage shape control
Composite fuselage shape control is a representative black-box optimization problem, where the fuselage shape is a high-dimensional functional response with strong spatial correlations and each evaluation is costly. Existing functional Bayesian optimization methods for this problem often suffer from prohibitive computational burdens due to high dimensionality and inadequate representations of spatial correlations in fuselage shapes. To address these limitations, this study proposes a B-spline empowered functional Bayesian optimization (BFBO) method, in which fuselage shapes are represented using low-dimensional B-spline control points and their correlations are modeled using a multivariate Gaussian process (MGP). Based on the predictive distribution of control points, a generalized chi-square expected improvement criterion is derived to guide the optimization process. Numerical results show that BFBO achieves improved performance in terms of convergence speed and optimization accuracy, demonstrating its effectiveness for fuselage shape control and its potential as a practical optimization framework for the design and manufacturing of engineering systems involving functional responses.
- Open Access
Effect Decomposition of Functional-Output Computer Experiments via Orthogonal Additive Gaussian Processes
Functional ANOVA (FANOVA) is a widely used variance-based sensitivity analysis tool. However, studies on functional-output FANOVA remain relatively scarce, especially for black-box computer experiments, which often involve complex and nonlinear functional-output relationships with unknown data distribution. Conventional approaches often rely on predefined basis functions or parametric structures that lack the flexibility to capture complex nonlinear relationships. Additionally, strong assumptions about the underlying data distributions further limit their ability to achieve a data-driven orthogonal effect decomposition. To address these challenges, this study proposes a functional-output orthogonal additive Gaussian process (FOAGP) to efficiently perform the data-driven orthogonal effect decomposition. By enforcing a conditional orthogonality constraint on the separable prior process, the proposed functional-output orthogonal additive kernel enables data-driven orthogonality without requiring prior distributional assumptions. The FOAGP framework also provides analytical formulations for local Sobol' indices and expected conditional variance sensitivity indices, enabling comprehensive sensitivity analysis by capturing both global and local effect significance. Validation through two simulation studies and a real case study on fuselage shape control confirms the model's effectiveness in orthogonal effect decomposition and variance decomposition, demonstrating its practical value in engineering applications.
Open AccessEffect Decomposition of Functional-Output Computer Experiments via Orthogonal Additive Gaussian Processes
Functional ANOVA (FANOVA) is a widely used variance-based sensitivity analysis tool. However, studies on functional-output FANOVA remain relatively scarce, especially for black-box computer experiments, which often involve complex and nonlinear functional-output relationships with unknown data distribution. Conventional approaches often rely on predefined basis functions or parametric structures that lack the flexibility to capture complex nonlinear relationships. Additionally, strong assumptions about the underlying data distributions further limit their ability to achieve a data-driven orthogonal effect decomposition. To address these challenges, this study proposes a functional-output orthogonal additive Gaussian process (FOAGP) to efficiently perform the data-driven orthogonal effect decomposition. By enforcing a conditional orthogonality constraint on the separable prior process, the proposed functional-output orthogonal additive kernel enables data-driven orthogonality without requiring prior distributional assumptions. The FOAGP framework also provides analytical formulations for local Sobol' indices and expected conditional variance sensitivity indices, enabling comprehensive sensitivity analysis by capturing both global and local effect significance. Validation through two simulation studies and a real case study on fuselage shape control confirms the model's effectiveness in orthogonal effect decomposition and variance decomposition, demonstrating its practical value in engineering applications.