Optimization statistically designed experiment

The development of on-line sensors is a very costly and time-consuming process. Therefore, if one has available a dynamic model of the reactor which predicts the various polymer (or latex) properties of interest, then this can be used to guide one in the selection and development of sensors. Ideas from the optimal statistical design of experiments together with the present model expressed in the form of a Kalman filter have been successfully used (58) to select those combinations of existing or hypothetical sensors which would maximize the information that could be obtained on the states of the polymerization system. Both the type of sensors and the precisions necessary for them are easily investigated in this way. By changing the choice of the measurement matrix and... 

We now illustrate some recent examples of chemometrical modeling of IPC systems for the sake of clarity. In the framework of a quality by design approach, statistically designed experiments were used to optimize the IPC condition for the analysis of atomoxetine and impurities and demonstrate method robustness. 

When up-scaling a reaction the use of statistically designed experiments may facilitate the optimization of the reaction [474] when the different parameters of the reaction... 

Myers, R.H. Montgomery, D.C. Response surface methodology process and product optimization using designed experiments. Wiley Series in Probability and Statistics, Wiley Interscience, 1995 248 pp. 

Within the last several years HPLC separations have been optimized in terms of the most appropriate mobile phase composition for a particular set of solutes by exploring the whole plane of solvent selectivities using this solvent classification scheme with a minimal number of measurements in statistically-designed experiments. For reversed phase HPLC systems, the selectivity triangle is often defined by methanol, acetonitrile, and tetrahydrofuran with water as the diluent (37). 

The approach in material development is to systematically perform statistically designed experiments. Techniques include fractional factorial designs, Grecko-Latin squares, and self-directed optimization. Data collected is statistically evaluated to determine primary and combined effects of the... 

The main purpose of the process optimization program is to identify the reaction parameters that would provide the best possible yield and quality in each step. To achieve quality and yield repeatably and reproducibly, planned optimization process parameters and strict controls on critical operating parameters are a must in each step of the process. Process optimization is done by using either the traditional one-at-a-time approach or statistically designed experiments, depending on the nature of interactions between parameters. This is ensured by smdying each key parameter involved in any reaction and identilying the optimal conditions. The parameters that are always studied are ... 

In statistically designed experiments, a broader screening set of experiments frequently answers the first question, a more focused experimental matrix generates the model for the second, and that matrix delivers the optimization to answer the third question. It is important to put statistical methods for chemical process optimization in a suitable context When one designs a new jet passenger plane or a fighter, one has to spend a lot of money and time in obtaining reliable model information. ... 

An experimental design was produced using Design-Expert software from Stat-Ease, Inc. Statistically designed experiments were produced D-optimally for the mixture variables. Nineteen different formulations and six replicates were needed... 

Statistical Design I. In order to optimize the conversion of MCI to MF we chose to use a statistically designed experimental approach. This approach is particularly useful when a large number of variables are involved over a rather large reaction space. Essentially a statistically designed experiment is produced by defining the reaction space with variables that may have some bearing on a desired result. This result must be quantifiable and is referred to as the response surface. The completed "experiment" then describes the response of interest as a function of the variables, i.e. a surface in "n" dimensional space. 

Statistical design (experiments) A technique for optimizing the information that is obtained from the least number of experiments. Useful for establishing process parameter limits. Also called Factorial design. See also Parameter windows. 

Because all the variables that influence the properties of the final product are known, one can use a statistical design (known as a one-half factorial) to optimize the properties of the GPC/SEC gels. Factorial experiments are described in detail by Hafner (10). For example, four variables at two levels can be examined in eight observations. From these observations the significance of each variable as related to the performance of the gel can be determined. An example of a one-half factorial experiment applied to the production of GPC/SEC gel is set up in Table 5.2. The four variables are the type of DVB, amount of dodecane, type of methocel, and rate of stirring. 

This work describes one approach for optimizing recovering systems using a simulation package in conjunction with standard statistical techniques such as designed experiments, multiple correlation analyses and optimization algorithms. The approach is illustrated with an actual industrial process. 

A further reduction of experimental effort may be achieved by the selection of special designs developed, for example, by HARTLEY [1959], BOX and BEHNKEN [1960], WESTLAKE [1965], and others. In these designs the ratio of experiments to the number of coefficients necessary is reduced almost to unity. (This situation is somewhat different from regression analysis or random selection of experiments where, in principle, k experiments or measurements are sufficient to estimate k parameters of a model. In experimental design the optimized number of experiments is derived from statistical consideration to encompass as much variation of the factors as possible.)... 

Jane Chang is an Assistant Professor in the Department of Applied Statistics and Operations Research at Bowling Green State University. Her research interests are in optimal experimental design, the design and analysis of microarray experiments, and multiple testing in two-level factorial designs. 

---