Farokhi, Hamed
Bacarreza, Omar
Aliabadi, M.H. Ferri
Funding for this research was provided by:
Northumbria University
Article History
Received: 31 March 2019
Revised: 9 February 2020
Accepted: 3 March 2020
First Online: 15 May 2020
Compliance with ethical standards
:
: The authors declare that they have no conflict of interest.
: There are several ways through which the readers can check their simulation results against those presented in this study. First is using the tables provided in this study; in fact, the most important optimisation results of the present study are reported using tables including the optimisation inputs and outputs to make it easier for benchmark analysis and comparison purposes. Additionally, the raw data for constructing the results presented in Figs. InternalRef removed, InternalRef removed, InternalRef removed, InternalRef removed, InternalRef removed, InternalRef removed and InternalRef removed is provided as supplementary materials; this allows the readers to check their optimisation results against a wide range of results reported in this study. The size of raw data for Fig. InternalRef removed is very large due to the large number of points (10<sup>6</sup>) used to create the distribution; therefore, the data for that figure is not provided. Additionally, since the random normal distributions and truncated ones for Monte Carlo samplings (i.e. Figs. InternalRef removed and InternalRef removed) are different each time they are generated, the data for these figures is not provided.It should be noted that the readers might obtain slightly different results than those presented in this study due to the probabilistic nature of the optimisation as well as the differences in the algorithms used for surrogate model development. If a surrogate model is developed by a reader with a similar accuracy to the one used in this study, even if a different method is used, they should be able to obtain results close to those presented in this study. As explained in the manuscript, this study utilises a Monte Carlo sampling for robust optimisation analysis using 10<sup>5</sup> samples. Such large number of samples was intentionally selected (through a convergence analysis), to ensure a converged probabilistic optimisation analysis assuming random normal distributions.