作者: Max D. Morris
DOI: 10.2307/1269043
关键词: Morris method 、 Factorial 、 Mathematics 、 Computational model 、 Latin hypercube sampling 、 Representation (mathematics) 、 Function (mathematics) 、 Elementary effects method 、 Fourier amplitude sensitivity testing 、 Algorithm
摘要: A computational model is a representation of some physical or other system interest, first expressed mathematically and then implemented in the form computer program; it may be viewed as function inputs that, when evaluated, produces outputs. Motivation for this article comes from models that are deterministic, complicated enough to make classical mathematical analysis impractical have moderate-to-large number inputs. The problem designing experiments determine which important effects on an output considered. proposed experimental plans composed individually randomized one-factor-at-a-time designs, data based resulting random sample observed elementary effects, those changes due solely particular input. Advantages approach include lack reliance assumptions relative sparsity inputs, monotonicity outputs with respect ad...