Uniform design: design, analysis and applications

Abstract

A brief introduction is given to explain the fundamental idea of uniform design and measure of uniformity. A new variable selection procedure is proposed for the analysis of data obtained from uniform designs. This new procedure is distinguished from the traditional ones in the way that it simultaneously deletes insignificant variables and estimates the coefficients of significant variables. This procedure possesses an oracle property, which means that it performs as well as if the true model were known in advance. An example is given to illustrate the application of the uniform design and this variable selection procedure.

Application of uniform design in optimisation of three stage ageing of Al–Cu–Mg alloys

“…Thin foils for transmission electron microscopy (TEM) were prepared using a standard procedure: samples of 400 μm this study because its principle is to replace the complete combination of experimental parameters by using relatively fewer experiment trials uniformly distributed within the parameter space. These experiment trials are determined using the numbertheoretic method [17,18] and mathematically proved to be a good approximation of the complete combination of experimental parameters [19]. For each of the three variables (factors), different treatment levels were incorporated ( Table 2).…”

Isolation of a novel deep-sea Bacillus circulus strain and uniform design for optimization of its anti-aflatoxigenic bioactive metabolites production

“…In order to enhance the yield of anti-fungal substances, many researchers used response surface methodology to optimize cultivation media and cultivation conditions [20][21][22]. Uniform design was first proposed by Wang yuan and Fang kaitai in 1978 [23]. Compared to the response surface methodology, uniform design is superior in that when dealing with multi-factors and multi-levels, fewer number of essential experiments are needed to achieve satisfactory results [23].…”

“…Uniform design was first proposed by Wang yuan and Fang kaitai in 1978 [23]. Compared to the response surface methodology, uniform design is superior in that when dealing with multi-factors and multi-levels, fewer number of essential experiments are needed to achieve satisfactory results [23]. In this research, we firstly used the uniform design method to optimize the concentrations of different nutrient components in ISP2 medium and the fermentation conditions for production of anti-aflatoxigenic metabolites by the FA13 strain.…”

A new uniform evolutionary algorithm based on decomposition and CDAS for many-objective optimization

“…The uniform design method has been successfully implemented in science, engineering and industries [35][36][37][38][39][40]. The fact that the uniform design performs better at estimating nonlinear problems than other designs has been shown in the literature [41].…”