Replication Data for: Accounting for Non-normal Distribution of Input variables and Their Correlations in Robust Optimization

Published: 19 October 2020| Version 1 | DOI: 10.17632/h5grkr6p32.1
O Nejadseyfi


ExperimentalData.mat includes the data obtained via tensile tests in three directions from 49 different samples gathered from different coils. These are the raw data used to describe the input to the optimization problem. Principal component analysis is done after subtracting their mean values. OptimizationDOE.mat includes the DOE made by LHS sampling in combined noise and design variable space, SimulationResults.mat includes the finite element simulation results that are performed on DOE points. These data are used to build the Kriging metamodel for optimization using the above-mentioned material data. WindData.mat includes the collected wind data from a weather station described in the manuscript.



Robust Optimization