Attribute/variable Technique

The only statistical techniques which need control are those used to determine the acceptability of a product or service or the capability of a process that produces the product or service. Any activity where you rely on statistical evidence rather than physical measurement is an activity which should be governed by these requirements. The use of recognized techniques is important to the confidence one has in the result. It is similar to the use of measuring equipment that has been calibrated against known standards of accuracy. Unless you actually check every product, measure every attribute or variable you cannot be 100% certain. But that is costly and you can be 99.99% certain by using statistical techniques 99.99% may be sufficiently accurate for your needs. 

Techniques for multivariate input analysis reduce the data dimensionality by projecting the variables on a linear or nonlinear hypersurface and then describe the input data with a smaller number of attributes of the hypersurface. Among the most popular methods based on linear projection is principal component analysis (PCA). Those based on nonlinear projection are nonlinear PCA (NLPCA) and clustering methods. 

In a four-part article, Porter [130] provided a comprehensive review of tablet coating technology, with emphasis on contemporary practice. More specifically, a recent review [131] discusses characterization techniques for the aqueous film coating process and provides a useful influence matrix between process variables and final product attributes. 

Since the separation process in CEC has a number of attributes similar to those of HPLC, the most important variables affecting the separation are the same for both of these techniques. However, in HPLC mobile phase, flow and separation are independent variables. Therefore, the most important operational variables are the analyte-sorbent interactions that can be modulated by the chemistry of the packing, composition of the mobile phase, and temperature. In contrast, the CEC column has a dual role as it serves as both (i) a flow driving device and (ii) separation unit at the same time. Although the set of variables typical of HPLC is also effective in CEC, their changes may affect in one way or another both column functions. Therefore, optimization of the separation process in CEC is more complex than in HPLC. 

All of the compounds measured In the monitoring program are listed In the report by Thrane (VI). Table I lists the compounds which were selected as variables for the cluster analysis. Feature (l.e. attribute) selection for the cluster analysis was partially based upon the results of a principal component analysis (Henry, 12). Additional features were Included If (1) the compound occurred In relatively large concentrations, or (2), If a compound was known to have adverse health effect. Wind direction, wind speed, and temperature were recorded as ordered variables. The chemical measurements were taken at five locations. Descriptions of those sites and of the methods and techniques used to collect the data are described in detail in the report by Thrane. 

Process characterization represents the methods used to determine the critical unit operations or processing steps and their process variables, that usually affect the quality and consistency of die product outcomes or product attributes. Process ranging represents studies that are used to identify critical process or test parameters and their respective control limits, which normally affect the quality and consistency of the product outcomes of their attributes. The following process characterization techniques may be used to designate critical unit operations in a given manufacturing process. 

Flame is probably the oldest plasma known to humanity and flame treatment is one of the oldest methods used by industries for the modification of polymeric surfaces. Flame treatment is very often used to treat bulky objects. It is mainly employed to enhance the ink permeability on the polymer surface. Though a very simple set-up (comprising of a burner and a fuel tank) is required for this technique, a very high degree of craftsmanship is needed to produce consistent results. Oxidation at the polymer surface brought about by the flame treatment can be attributed to the high flame temperature range (1000-1500 °C) and its interaction with many exited species in the flame. For an efficient flame treatment, the variables like air-to-... 

Figure 106 takes lithography as an example and presents in diagram form the variables affecting success at the stages of pre-press, preparation of plate, choice of substrate, water quality, ink, printing proper, and variables attributable to the operator. Similar diagrams could be compiled for each of the other techniques. 

Samples collected with these devices may inadequately represent the pore water in its natural occurrence because of problems inherent in the technique (Litaor, 1988). This limitation may be additionally influenced by the complex nature of the soil, whose heterogeneity highly affects the chemical concentrations in pore water. Hence, Rhizon samplers, with their small cross-sectional area, may not adequately integrate for spatial variability (Amoozegar-Fard et al., 1982 England, 1974 Haines et al., 1982), and may represent point samples with qualitative rather than quantitative attributes (Biggar and Nielsen, 1976). 

You can use Design of Experiments (Technique 50) to help you determine specific attribute combinations you should test during the simulation. Design of Experiments allows you to identify interactions caused by changing two or more variables simultaneously. 

Normalized variables and nondimensional parameters may be identified from equation (114) [73], and an iterative method of solution may be devised [50] and applied [73] to obtain an approximate formula for — Po o Po o- This formula exhibits the attributes of the burning velocity for the heterogeneous regime that were discussed in Section 11.6.1. Thus a more formal development leads to the qualitative results that have been inferred from physical reasoning. It is instructive to have seen a representative sequence of approximations involved in a formal development. The iterative method [50] has been applied successfully [77] with fewer approximations, through use of techniques for numerical integration, to derive... 

In this Report of catalyst-preparation technology we have placed particular emphasis on catalyst design as opposed to preparation. A properly designed catalyst should have the essential attributes of activity, stability, selectivity, and regenerability. These can be related to the physical and chemical properties of the catalyst, which in turn can be related to the variable parameters inherent in the method used for the preparation of the catalyst. In the past much of the literature on supported catalysts has not included this information. In part this was due to the lack of techniques for physically and chemically characterizing supported catalysts. Many advances have been made in recent years in this area, as described in Chapter 2, so that the design of supported catalysts has become a feasible activity. 

Clustering attempts to find natural groups as determined by the metric used (Euclidean distance, cosine distance, or categorical attributes) which are blind as to treatment. These types of techniques are particularly useful for discovering new relationships among variables and for the derivation of measurement data based on natural differentiations and not the bias of the observer. The use of these techniques to determine assessment and measurement endpoints has been extensively discussed (Landis et al. 1994). 

Further statistical analyses can be used to determine the relative influence that any factor or set of factors has on the total variation (global uncertainty). One of these methods is the analysis of variance (ANOVA). This is an important technique for analyzing the effects of categorical factors on a response. However, the assumption of normality of the data has to be checked prior to the use of ANOVA to decompose the variability in the response variable between the different factors. Depending upon the type of analysis, it may be important to determine (a) which factors have a significant effect on the response, and/or (b) how much of the variability in the response variable is attributable to each factor (as described in the statistical software STATGRAPHICS, Vs 5.0). 

---