This readme describes the supplementary R source files and the set of experimental data, 
which was used in the article 
"Optimization via Multimodel Simulation - 
A New Approach to Optimization of Cyclone Separator Geometries"

Bart16eAllData.csv: This CSV file contains the data used in this study.

analyticalModelEvaluation.R: This R script calculates the values necessary to
reproduce Eq. 6 and parts of Table 5.

bart16eEda.R: This R script was used to generate the boxplots (Fig. 4) and to calculate 
means and standard deviations of the 3D-printing data (used in column M-P in Table 5).

bart16eOptimizationViaMultimodelSimulation.R: This R script was used to build the 
metamodel and to run the optimization step (S-9) on the metamodel F*, as mostly discussed in 
Section 7. 
The R functions, which are necessary for building the metamodel, are implemented in the R 
package SPOT. The open source R software package SPOT can be downloaded from CRAN with
the following command from within R: 
install.packages("SPOT")

cyclone.R: This file contains the functions for the analytical modeling of a cyclone dust separator.
It is used in the above *.R scripts. While SPOT also contains these functions, it is added here
as it contains some changes to the interface that are not included in the CRAN version
of SPOT at the time of article submission.