Optimal Experimental Design in the Presence of Nested Factors

Abstract

A common occurrence in practical design of experiments is that one factor, called a nested factor, can only be varied for some but not all the levels of a categorical factor, called a branching factor. In this case, it is possible, but inefficient, to proceed by performing two experiments. One experiment would be run at the level(s) of the branching factor that allow for varying the second, nested, factor. The other experiment would only include the other level(s) of the branching factor. It is preferable to perform one experiment that allows for assessing the effects of both factors. Clearly, the effect of the nested factor then is conditional on the levels of the branching factor for which it can be varied. For example, consider an experiment comparing the performance of two machines where one machine has a switch that is missing for the other machine. The investigator wants to compare the two machines but also wants to understand the effect of flipping the switch. The main effect of the switch is conditional on the machine. This article describes several example situations involving branching factors and nested factors. We provide a model that is sensible for each situation, present a general method for constructing appropriate models, and show how to generate optimal designs given these models. Supplementary materials for this article are available online.

Hierarchical optimal designs and modeling for engineering: A case‐study in the rail sector

“…More recently, in Reference 12 the authors investigate optimal designs for the fixed‐effects estimation as well as for prediction of random effects in hierarchical linear models, while in Reference 13 optimal designs for hierarchical random effects models applied in the field of precision medicine are taken into account. Moreover, by specifically considering the design of experiments with a particular type of factors, it is relevant to note that in Reference 14 the design and analysis of computer experiments are addressed in the presence of nested and branching factors, while more recently, in Reference 15 the authors investigate D‐optimal designs when nested and/or branching factors are present.…”

Comparison of conditional main effects analysis to the analysis of follow‐up experiments for separating confounded two‐factor interaction effects in 2IVk−p fractional factorial experiments

“…One situation where CME analysis has been proven to be effective is when there is a nested factor whose effect is likely to be different depending on the level of the factor within which it is nested (see Goos and Jones). In that case, again the forward selection of CME analysis is recommended.…”

Orthogonal designs with branching and nested factors

“…Hence, Hung et al (2009) incorporated LHD into the design with branching and nested factors and proposed a corresponding design called branching Latin hypercube design (BLHD). Goos and Jones (2019) discussed the modelling of data from experiments with branching and nested factors as well as the optimal design of such experiments. Chen et al (2019) considered the case where branching factors and nested factors are both qualitative and proposed two‐layer sliced Latin hypercube designs (SLHDs) to suit such situations.…”