CFE-CMStatistics 2024: Start Registration
View Submission - CFECMStatistics2024
A1606
Title: Sliced designs for computer experiments using sequences with zero autocorrelation function Authors:  Omar Alhelali - RMIT university (Australia) [presenting]
Abstract: Computer experiments have become increasingly important because they save time and money compared to physical experiments. A special type of experimental design, called sliced designs, is discussed, which expands on the idea of sliced Latin hypercube designs. The focus is on developing unique sliced orthogonal designs for computer experiments that can be used in first- and second-order models. Each slice forms a smaller, orthogonal design that's well-suited for computer simulations. The construction method proposed relies on using sequences like T-sequences, Golay sequences, and disjoint amicable sequences, which have zero autocorrelation functions. This method allows creating the first infinite families of these designs.