Statistical material selection reliability

Haugen, E. B. 1982b Modern Statistical Materials Selection - Part 2 Random Variables and Reliability. Materials Engineering, 96, August, 49-51. 

Ibis chapter will discuss various reliability issues. The discussion will kept at a philosophical level rather than getting into a statistical analy The statistical analysis is best left to others equipped with the training, toi and the data. Hopefully, this material will give the reader some comn sense insight into the various considerations involved in the selection applicadon of reliable compressors, their drivers, and auxiliary systems. 

The criteria for the choice of the CRM are not different from the criteria to select the material for the preparation of a laboratory reference material for method development, statistical control charts etc. The difference lies in the availability of adequate CRMs from reliable suppliers and the level of compromise which the analyst must make between an ideal situation and the reality of what is on offer. Massart and co-workers have proposed a principle component analysis to help select the best adapted CRMs available on the market to verify AAS analysis of foodstuffs [10], Their approach took into account the analytes as well as the matrix composition. Besides the fact that they highlighted a lack of sorts of CRM, in particular those having a fatty matrix, they demonstrated that such a statistical approach can help in the most appropriate selection of materials. Boenke also proposed a systematic approach for the choice of materials to be certified for mycotoxins [11] and which could be followed by potential users. The selection of the CRM by the analyst should include a certain number of parameters this can cover the following properties to fulfil the intended purpose level of concentration of the analytes ... 

The outcome of the previous risk assessment is a qualitative match, based on perceptions, which needs to be confirmed and quantified during a subsequent process development stage. A mechanistic description of the process, as introduced in Sect. 8.1.5, can be used as a powerful tool to establish the design space with minimum need for experimentation at final scale. However, this type of description is not always readily available, and the use of a statistical approach constitutes a pragmatic alternative. The limitation of the statistical approach is its reliance on scale-dependent experimentation and the difficulty in extrapolating relationships to other scales or equipment. Selection of the most adequate approach (mechanistic or statistical) depends, therefore, on the (1) existence/reliability of the mechanistic understanding, (2) availability of material for experimentation, and (3) flexibility required on the design space. In this section, a statistical approach will be illustrated in order to quantify relationships between process parameters and product attributes the three involved steps are described in Fig. 8.22. 

In order to characterize clusters with different sizes, a resolution on an atomic scale is necessary. The available TEM instrument at TUM was able to confirm successful deposition of cluster materials, and allowed for simple statistical analysis of larger (unselected) clusters [1, 3]. However, a reliable analysis with focus on even smaller sizes towards discrimination of selected cluster with different sizes was not possible due to low resolution in the lower size range. Therefore, acquired TEM images are not explicitly shown. 

---