Reply to “Some Observations on the Paper ‘Optimal Experimental Design for Discriminating Numerous Model Candidates—The AWDC Criterion’ ”

Michalik, Claas; Stuckert, Maxim; Marquardt, Wolfgang

doi:10.1021/ie1015574

Cited by 1 publication

(1 citation statement)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Two criteria are usually proposed in the literature, namely, the sum over all model pairs, Donckels et al (2009), Schwaab et al (2006), Chen and Asprey (2003), Hunter and Reiner (1965) and the worst-case method, in which a max-min approach is applied, Cooney and McDonald (1995), Bambach et al (2013). As the motivating example of Michalik et al (2009) shows, the first criteria may lead to a bad solution, since the proposed experiment can originate a big separation between two particular models, at the same time that some of them are very close, giving an overall good result but an inefficient model selection method. Hence, in this paper, we will use as a discrepancy measure the max-min approach.…”

Section: Introductionmentioning

confidence: 99%

A global optimization method for model selection in chemical reactions networks

Blanquero

Carrizosa

Jiménez-Cordero

et al. 2016

Computers & Chemical Engineering

View full text Add to dashboard Cite

Model inference is a challenging problem in the analysis of chemical reactions networks. In order to empirically test which, out of a catalogue of proposed kinetic models, is governing a network of chemical reactions, the user can compare the empirical data obtained in one experiment against the theoretical values suggested by the models under consideration. It is thus fundamental to make an adequate choice of the decision variables (e.g. initial concentrations of the different species in the tank) in order to have maximal separation between sets of concentrations provided by the theoretical models, making then easier to identify which of the theoretical models yields data closest to those obtained empirically under identical conditions.In this paper we illustrate how global optimization techniques can be successfully used to address the problem of model separation, as a basis for model selection. Some examples illustrate the usefulness of our approach.

show abstract