Regularities in data from factorial experiments

Li, Xiang; Sudarsanam, Nandan; Frey, Daniel

doi:10.1002/cplx.20123

Cited by 110 publications

(74 citation statements)

References 52 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The methods assume that the direction of the effects of the active variables tend to be monotonic. This regularity was seldom investigated in literature, and was proven to be valid when the system regularities of Bergquist et al (2011) and Li et al (2006) are present, as explained in Al Rashdan (2014. In addition, the methods evaluated the validity of this assumption at an early stage of their application.…”

Section: Introductionmentioning

confidence: 88%

Parametric evaluation of an SMR design domain

Rashdan

Tsvetkov

2015

Annals of Nuclear Energy

View full text Add to dashboard Cite

Section: Introductionmentioning

confidence: 88%

Parametric evaluation of an SMR design domain

Rashdan

Tsvetkov

2015

Annals of Nuclear Energy

View full text Add to dashboard Cite

“…As a first approximation, however, it is sufficient to estimate only the linear (i.e., main) effects because nonlinear and interaction effects are often much smaller than main effects; see, for example, pages 192, 230, 272, 314, and 329 in Montgomery (2009) and pages 33-34 in Li et al (2006). For this reason a linear regression-based model is adequate.…”

Section: Statistical Modelmentioning

confidence: 99%

On the effect of model parameters on forecast objects

Marzban

Jones

et al. 2018

Geosci. Model Dev.

View full text Add to dashboard Cite

Abstract. Many physics-based numerical models produce a gridded, spatial field of forecasts, e.g., a temperature "map". The field for some quantities generally consists of spatially coherent and disconnected "objects". Such objects arise in many problems, including precipitation forecasts in atmospheric models, eddy currents in ocean models, and models of forest fires. Certain features of these objects (e.g., location, size, intensity, and shape) are generally of interest. Here, a methodology is developed for assessing the impact of model parameters on the features of forecast objects. The main ingredients of the methodology include the use of (1) Latin hypercube sampling for varying the values of the model parameters, (2) statistical clustering algorithms for identifying objects, (3) multivariate multiple regression for assessing the impact of multiple model parameters on the distribution (across the forecast domain) of object features, and (4) methods for reducing the number of hypothesis tests and controlling the resulting errors. The final "output" of the methodology is a series of box plots and confidence intervals that visually display the sensitivities. The methodology is demonstrated on precipitation forecasts from a mesoscale numerical weather prediction model.

show abstract

“…Since main and low-order interaction effects are part of high-order effects, knowledge of the latter ones is more important. Interactions of level 4 and above are highly unlikely [17]. Hence, a half fractional factorial design is chosen, which is a specific type of DoE that reduces the number of tests by 50% compared to testing all the possible combinations of factors and levels and that is able to map interaction effects up to level 3.…”

Section: Methodsmentioning

confidence: 99%

Tensile Properties of Inkjet 3D Printed Parts: Critical Process Parameters and Their Efficient Analysis

Mueller

Kim

Shea

et al. 2015

Volume 1A: 35th Computers and Information in Engineering Conference

View full text Add to dashboard Cite

To design and optimize for capabilities of additive manufacturing processes it is also necessary to understand and model their variations in geometric and mechanical properties. In this paper, such variations of inkjet 3D printed parts are systematically investigated by analyzing parameters of the whole process, i.e. storage of the material, printing, testing, and storage of finished parts. The goal is to both understand the process and determine the parameters that lead to the best mechanical properties and the most accurate geometric properties. Using models based on this understanding, we can design and optimize parts, and fabricate and test them successfully, thus closing the loop. Since AM materials change rapidly and this process will have to be repeated, it is shown how to create a cost and time efficient experimental design with the one-factor-at-a-time and design of experiments methods, yielding high statistical accuracies for both main and interaction effects. The results show that the number of intersections between layers and nozzles along the load-direction has the strongest impact on the mechanical properties followed by the UV exposure time, which is investigated by part spacing, the position on the printing table and the expiry date of the material. Minor effects are found for the storage time and the surface roughness is not affected by any factor. Nozzle

show abstract

Regularities in data from factorial experiments

Abstract: This article documents a meta-analysis of

Cited by 110 publications

References 52 publications

Parametric evaluation of an SMR design domain

Parametric evaluation of an SMR design domain

On the effect of model parameters on forecast objects

Tensile Properties of Inkjet 3D Printed Parts: Critical Process Parameters and Their Efficient Analysis

Contact Info

Product

Resources

About