Examining robustness of model selection with half‐normal and LASSO plots for unreplicated factorial designs

Jang, Dae‐Heung; Anderson‐Cook, Christine M.

doi:10.1002/qre.2156

Cited by 6 publications

(5 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…32 The factorial analysis method is an efficient factor screening method, which can analyze the complex interaction effect between factors. The mathematical model used in the factorial design method can be expressed as 33 (a) (b)…”

Section: Factorial Analysis Methodsmentioning

confidence: 99%

Multi-objective crashworthiness optimization of vehicle body with negative Poisson’s ratio structure based on factorial analysis

Zou

Dai

Zhao

et al. 2020

Proceedings of the Institution of Mechanical Engineers, Part D:

View full text Add to dashboard Cite

To improve vehicle side crashworthiness, this paper first introduces the negative Poisson’s ratio structure to the traditional B-pillar and proposes a negative Poisson’s ratio B-pillar. Then, the performance of the negative Poisson’s ratio B-pillar is studied in detail by comparison with a traditional B-pillar and honeycomb B-pillar. Aiming at the problem that the side crashworthiness is also significantly affected by the side structure parameters of vehicle body, the factorial analysis theory is adopted to screen out the side structure parameters with significant effect. Based on this, by combining the optimal Latin hypercube design and response surface model, a multi-objective optimization design is conducted for those structure parameters based on non-dominated sorting genetic algorithm II. Finally, the normal boundary intersection method is adopted to seek the Pareto optimal solution, and the simulation results show that compared with the traditional B-pillar, the negative Poisson’s ratio B-pillar optimized by non-dominated sorting genetic algorithm II has better comprehensive crashworthiness. The results of this paper can provide some basis for the design and optimization of vehicle side crashworthiness.

show abstract

Section: Factorial Analysis Methodsmentioning

confidence: 99%

Multi-objective crashworthiness optimization of vehicle body with negative Poisson’s ratio structure based on factorial analysis

Zou

Dai

Zhao

et al. 2020

Proceedings of the Institution of Mechanical Engineers, Part D:

View full text Add to dashboard Cite

show abstract

“…Here, the half-normal plot [32] method is used for the data processing of USFD. Under the assumption that the error is normal, independent, and of the same variance, the estimators of each factor are independent of each other.…”

Section: Screening Of Significant Influencing Factorsmentioning

confidence: 99%

Optimization of the Tribological Performance and Service Life of Calcium Sulfonate Complex—Polyurea Grease Based on Unreplicated Saturated Factorial Design

Zhang,

Mo,

Liu

et al. 2023

Lubricants

View full text Add to dashboard Cite

In order to further extend the service life of calcium sulfonate complex–polyurea grease (CSCPG) while ensuring its tribological performance, this article starts with the production of raw materials and the preparation process of the grease and explores the factors that significantly affect the tribological performance and service life of CSCPG based on unreplicated saturated factorial design (USFD). The Kriging prediction model is used along with the optimization objectives of friction coefficient and service life, and nondominated sorting genetic algorithm II (NSGA-II) was used for a multi-objective optimization solution. The tribological and service life tests were conducted before and after optimization. The results show that the viscosity of the base oil and the content of the nano-solid friction reducers have a significant impact on the tribological properties of CSCPG. The content of polyurea thickeners and antioxidants, as well as the thickening reaction temperature, have a significant impact on the service life of CSCPG. When the friction coefficient and service life are optimized as objectives and are compared to the initial group, the friction coefficient of CSCPG could be reduced by 5.3%, and the service life could be extended by 3.8%. The Kriging prediction model based on USFD has high accuracy and can be used to guide the preparation and performance optimization of CSCPG.

show abstract

“…Recently, Jang and Anderson‐Cook suggested the LASSO influence plot for exploring the impact of individual observations on the results for observational datasets with high correlation between explanatory variables. In addition, the plots can be used for unreplicated factorial designs . In this paper, we describe how these plots can also be used to increase understanding of the robustness of results in analysis with SSD, where by definition the possible factors must be correlated with each other.…”

Section: Introductionmentioning

confidence: 99%

“…In addition, the plots can be used for unreplicated factorial designs. 11 In this paper, we describe how these plots can also be used to increase understanding of the robustness of results in analysis with SSD, where by definition the possible factors must be correlated with each other. Although these designed experiments have known structure and correlation patterns, the fact that so many factors are being evaluated can impact the stability of results.…”

Section: Introductionmentioning

confidence: 99%

Assessing robustness of factor ranking for supersaturated designs

Jang

Anderson‐Cook

2018

Quality & Reliability Eng

Self Cite

View full text Add to dashboard Cite

Supersaturated designs can potentially be a beneficial tool for efficiently exploring a large number of factors with a moderately sized design. However, because more factors are being considered than there are runs, the stability of the identified factors depends heavily on effect sparsity and the lack of highly influential observations. A helpful tool for the analysis of supersaturated designs is least absolute shrinkage and selection operation (LASSO), which is useful when the effects of many explanatory variables are sparse in a high‐dimensional dataset. To understand the impact of individual observations on the selected factors, the LASSO influence plot was created. This paper describes an application of this plot and its variants that can be used to identify influential points, increase understanding of the impact of individual observations on model parameters, and the robustness of results in analyses with supersaturated designs. These graphical methods can serve as a complement to other regression diagnostics techniques in the LASSO regression setting.

show abstract

Examining robustness of model selection with half‐normal and LASSO plots for unreplicated factorial designs

Cited by 6 publications

References 7 publications

Multi-objective crashworthiness optimization of vehicle body with negative Poisson’s ratio structure based on factorial analysis

Multi-objective crashworthiness optimization of vehicle body with negative Poisson’s ratio structure based on factorial analysis

Optimization of the Tribological Performance and Service Life of Calcium Sulfonate Complex—Polyurea Grease Based on Unreplicated Saturated Factorial Design

Assessing robustness of factor ranking for supersaturated designs

Contact Info

Product

Resources

About