Predictor dimension

Another process monitoring technique which is based on a statistical method called correlation technology (CT), has been offered commercially by Algoryx [16]. CT is based on the idea that although the relationships between the causes (machine settings) and effects (dimensions) may be too difficult to discern, the relationships between physical dimensions of the finished parts should always be related and consistent within a normal process window. The technique is based on the use of proprietary algorithms to characterize the dimensional relationships and to identify a lead predictor dimension. ... 

Plastic injection molding is an example where part dimensions are related. If a plastic injection molded part has five critical dimensions and is manufactured in a 16-cavity mold, the molding press will produce 80 dimensions for each machine cycle (shot). It is possible to determine the single predictor dimension that is the statistically best predictor of the remaining 79 dimensions. 

Determining the predictor dimension that is the statistically best predictor of all other dimensions can be (and usually is) very computationally intensive. It is not unusual to have to calculate 20,000-30,000 correlation coefficients to determine... 

When the predictor dimension is known, all other dimensions are also known. It is then not necessary to measure or do SPC or process capability analysis on all of the predicted dimensions as shown in Figure 16-19. 

In a similar fashion, many companies perform process capability (Cpk) analyses. Returning to our example of a plastic injection molded part with 5 critical dimensions that is manufactured in a 16-cavity mold, one would do 80 process capability analyses with prior technology. With the Lean Process Capability technology afforded by Correlation Master, one would have to do only process capability analysis on the predictor dimension. 

Figure 16-20. Unconstrained critical dimensions never have to be measured when the predictor dimension is conforming.

Fig. 38.13. Loadings of predictor (instrumental) variables on the first two PLS dimensions.

The relative effects of supercitical carbon dioxide density, temperature, extraction cell dimensions (I.D. Length), and cell dead volume on the supercritical fluid extraction (SFE) recoveries of polycyclic aromatic hydrocarbons and methoxychlor from octadecyl sorbents are quantitatively compared. Recoveries correlate directly with the fluid density at constant temperature whereas, the logarithms of the recoveries correlate with the inverse of the extraction temperature at constant density. Decreasing the extraction vessels internal diameter to length ratio and the incorporation of dead volume in the extraction vessel also resulted in increases in SFE recoveries for the system studied. Gas and supercritical fluid chromatographic data proved to be useful predictors of achievable SFE recoveries, but correlations are dependent on SFE experimental variables, including the cell dimensions and dead volume. 

Based on Henderson et al. s work, an abbreviated version of the ISSI has been developed by Unden and Orth-Gomer (1989) which contains two dimensions availability of social integration (AVSI) and availability of attachment (AVA). The AVSI measures superficial relationships and support. It simply focuses on the quantity or number of relationships which would be a weak predictor of future psychological health. However, the AVA measures the availability of deep relationships with close friends, family members and spouses who could provide emotional support. The AVA purports to be a good predictor of resistance to psychological stress. 

Time is the fundamental predictor in PK/PD models and therefore deserves some special attention. In NONMEM it is possible to specify the time points for observations and dosing events in the form of date and clock time. This is very convenient as this is the form in which the data is often stored in clinical databases. These dates and clock times are converted to decimal times in the preprocessing stages of the execution of a NONMEM run, and it is these decimal times that are used by NONMEM in the minimization procedure and that are provided in the tabulated output. From a plotting point of view, dates and clock times are not easy to work with. Except for cases with diurnal variations and/or annual rhythms (2,3), the extra dimension offered by dates and clock times are unnecessary and may actually make it harder to visualize the data in an informative way. 

Whether based on past performance or current performance, predictors must not only be accurate, they must also be relevant to the actual tasks to be performed. In other words, the criteria against which candidates performance is assessed must match those of the tasks to be performed (Swezey 1981). One way to ensure that predictors match tasks is to use criterion-referenced assessment of predictor variables for employee selection. To do this, one must ensure that the predictors match the tasks for which one is selecting candidates as closely as possible on three relevant dimensions the conditions under which the tasks will be performed, the actions required to perform the task, and the standards against which successful task performance wiU be measured. While it is better to have multiple predictors for each important job task (at least one measure of past performance and one of current performance), it may not always be possible to obtain multiple predictors. We do not recommend using a single predictor for multiple tasks—the predictor is likely to be neither reliable nor valid. 

Bosak, J., Coetsee, W. J., Cullinane, S. (2013). Safety climate dimensions as predictors for risk behavior. Accident Analysis and Prevention, 55, 256-264. 

Kath, L. M., Marks, K. M., Ranney, J. (2010b). Safety climate dimensions, leader-member exchange, and organizational support as predictors of upward safety communication in a sample of rail industry workers. Safety Science, 48, 643-650. 

Experience in sales situations has been shown to be a predictor of customer orientation, as Franke and Park show in a meta-study (Franke and Park 2006). They claim experience with sales leads to a greater ability to identify ways to help satisfy customer needs (Pranke and Park 2006, p. 696]. The same argumentation holds for the assertion that use experience will also foster customer orientation. By using product themselves, employees get to know customer needs and problems directly in a first-hand experience. They also built implicit need knowledge. This, in turn, will help to better understand customer needs, which is an important dimension of customer orientation (Brown et al. 2002]. Homburg et al. (2009] provide empirical support for the positive correlation between customer orientations of employees and need knowledge. 

---