Latin hypercube designs (LHDs) with space-filling properties are widely usedfor emulating computer simulators. Over the last three decades, a wide spectrumof LHDs have been proposed with space-filling criteria like minimum correlationamong factors, maximin interpoint distance, and orthogonality among the factorsvia orthogonal arrays (OAs). Projective geometric structures like spreads,covers and stars of PG(p-1,q) can be used to characterize the randomizationrestriction of multistage factorial experiments. These geometric structures canalso be used for constructing OAs and nearly OAs (NOAs). In this paper, wepresent a new class of space-filling LHDs based on NOAs derived from stars ofPG(p-1, 2).
展开▼
