MixedDoE-MATLAB is a program that creates a Design of (computer) Experiments (DoE) for combination of discrete and/or continuous variables in Matlab. The discrete part is composed either of full factorial or Taguchi-type Orthogonal array designs. For each point in the discrete part, a sliced design (Qian, 2015) is created for continuous variables. In this way, the continuous domain is uniformly covered but still, for different discrete points the continuous part is not overlapping. Two procedures for the continuous part are available - either Halton sequence, or the classical LHS design. In the same way, the possibility of independent testing data generation is available.

For educational and research purposes, license for using StableTrussOpt-MATLAB is free of charge. For commercial purposes, please contact program developers.

Copyright (C) 2024 Matěj Lepš, Anna Kučerová

In addition to Matlab it is essential to have a code for Least common multiple and regular Orthogonal array codes. Both algorihms found on internet are provided. This way, I would like to deeply thank their authors for providing them to the public.

Just very small test example

>> main 

Your design using JSON file, inspect examples for syntax

>> main 'JSONfile.json'

Documentation/user manual is available here.

Peter Z. G. Qian, Sliced Latin Hypercube Designs, Journal of the American Statistical Association, 107:497, 393-399 (2012). DOI: https://doi.org/10.1080/01621459.2011.644132

Matěj Lepš
Anna Kučerová

Financial support for this work was provided by the Technology Agency of the Czech Republic (TAČR), project number FW04020163 (Extension of the HiStruct platform by optimisation of global stability and analysis of design combinations).