Uniform designs

A single evaluation set is the simplest and most rapid validation scheme. A fraction—usually between 50% and 90% of the total number—of the available samples constitutes the training set, while the remaining objects form the evaluation set. The subdivision may be arbitrary, random, or performed by way of a uniform design, such as the Kennard and Stone and the duplex algorithm (Kermard and Stone, 1969 Snee, 1977), which allows two subsets to be obtained that are imiformly distributed and representative of the total sample variability. 

Hence, when cut into veneers, either in the direction shown by the dottad lines, or obliquely or longitudinally, it will yield a number of slabs or layers of the same uniform design. The thin rods of glass employed for this purpose are composed of very cosily fusible glass of all possible colors and shades, and arc prepared in manufactories erected expressly for this purpose. 

Phase 3 - The subjects were issued 100% cotton denim coveralls of uniform design and weight. These subjects were divided into two groups, one wearing Scotchgard treated uniforms and one wearing non-treated coveralls. During this study phase, respirators were not worn. As in the first and second phase, pads and urines were collected daily. 

As mentioned above, some of the primary optimal conditions determined in the first and second stages cannot yet be confirmed as really optimal, because they are the boundary values tested. In order to optimize these conditions further, some supplementary experiments were carried out in wider ranges to examine the influences of the pH, the reaction temperature, and the concentration of TiCl4 solution on the mean size of the product, for which the uniform design technique was employed. The experimental conditions and the results are listed in Table 15.8, where the values for the average sizes of particles are calculated from the X-ray spectrums. 

Fang, K. T. (1980). The uniform design Application of number-theoretic methods in experimental design. Acfa Mathematicae Applicatae Sinica, 3, 363-372. 

Unlike W plasma etch back process, the typical W CMP process usually removes the adhesion layer such as Ti/TiN or TiN during the primary polish. As a result, during the over polish step there is some oxide loss. Since the oxide deposition, planarization CMP (oxide CMP), and tungsten CMP steps are subsequent to each other, the oxide thickness profile could become worse further into the process flow. Therefore, the across-wafer non-uniformity of the oxide loss during W CMP process is one of the very important process parameters needs to be optimized. To determine the effect of the process and hardware parameters on the polish rate and the across-wafer uniformity, designed experiments were run and trends were determined using analysis of variance techniques. Table speed, wafer carrier speed, down force, back pressure, blocked hole pattern, and carrier types were examined for their effects on polish rate and across-wafer uniformity. The variable ranges encompassed by the experiments used in this study are summarized in Table I. 

This analysis employs uniform design and economic bases. 

If we are to understand the structure of a crystal lattice fully and are interested both in the mutual position of the atoms and the magnitude and nature of the forces, we must take into account the interval rule and the other properties of the substance in addition to the above described geometrical considerations and try to merge them into a uniform design. 

The total load divided by the cross sectional area is the uniform design load, p. 

For consumer and end users of polymer materials an indication of the chemical composition of the polymer is not sufficient. A polymeric material can be modified to be tailor made to a wide range of processing and performance requirements. Consumer and end users require a uniform designation for commercially available products. The designation should indicate which basic properties characterise the corresponding product. 

In all about 200 trademarks are registered. Apart from this, alkaline cells are offered in four different classes. The manufacturers attempt to make these classes differentiable by using certain labels, but a uniform designation has not been introduced. As has already been mentioned, choosing a product is a complex problem, with the consumer mainly making a decision on the brand and price. 

---