Control limit robust

Specifications How good do the numbers have to be Write specifications Pick methods to meet specifications Consider sampling, precision, accuracy, selectivity, sensitivity, detection limit, robustness, rate of false results Employ blanks, fortification, calibration checks, quality control samples, and control charts to monitor performance Write and follow standard operating procedures... 

Such data are shown in Table 3 and Fig. 6. Upper and lower control limits are calculated based upon n = 2 and A2 = 1.880. Thus, for 10 lots there will be 9 data points to plot, which results in a robust analysis of the quality control data for the product. Unlike a normal control chart, when you decide to use RSD values to create the quality control chart, the lower control limit (LCL) is more desirable than the upper control limit (UCL) simply because lower RSD values reflex a tighter dispersion around the mean. 

The s.d. is calculated on the basis of the difference between two succeeding log, transformed counts. This method of calculating the s.d. is preferred when only one sample is examined for each observation on the graph [39]. It will result in a more robust estimate of the s.d. compared to the usual way of calculating the s.d. It also means that this estimate of the s.d. is likely to be less affected by variations in counts due to assigned (systematic) errors [38] and results in more robust control limits. From the mean and standard deviation the following control chart limits are calculated on the logarithmic scale ... 

Process characterization defines process capability and facilitates prospective process validation at the production scale [40]. Full process characterization is valuable in maintaining smooth manufacturing operations and minimizing lost batches, and it provides supporting information for lot release justification for atypical batches [15]. Its goals are to identify key operating and performance parameters, define control limits for key process parameters, demonstrate robustness of the commercial process, and provide technical information about the process [68]. The steps involved in process characterization include risk assess-... 

The MPC control problem illustrated in Eqs. (8-66) to (8-71) contains a variety of design parameters model horizon N, prediction horizon p, control horizon m, weighting factors Wj, move suppression factor 6, the constraint limits Bj, Q, and Dj, and the sampling period At. Some of these parameters can be used to tune the MPC strategy, notably the move suppression faclor 6, but details remain largely proprietary. One commercial controller, Honeywell s RMPCT (Robust Multivariable Predictive Control Technology), provides default tuning parameters based on the dynamic process model and the model uncertainty. 

It is inherently safer to develop processes with wide safe operating limits that are less sensitive to variations in critical safety operating parameters, as shown in Figure 4.3. Sometimes this type of process is referred to as a forgiving or robust process. If a process must be controlled within a very small temperature band in order to avoid... 

Linear control theory will be of limited use for operational transitions from one batch regime to the next and for the control of batch plants. Too many of the processes are unstable and exhibit nonlinear behavior, such as multiple steady states or limit cycles. Such problems often arise in the batch production of polymers. The feasibility of precisely controlling many batch processes will depend on the development of an appropriate nonlinear control theory with a high level of robustness. 

Analytical methods, particularly those used by accredited laboratories, have to be validated according to official rules and regulations to characterize objectively their reliability in any special field of application (Wegscheider [1996] EURACHEM/WELAC [1993]). Validation has to control the performance characteristics of analytical procedures (see Chap. 7) such as accuracy, precision, sensitivity, selectivity, specificity, robustness, ruggedness, and limit values (e.g., limit of detection, limit of quantitation). 

---