CENTRAL COMPOSITE DESIGN IN RESPONSE SURFACE METHODOLOGY APPLICATIONS

摘要

Experimental optimization and process characterization can be carried out in several ways. Most common is the one variable at a time (OFAT) approach. This approach is an experimental technique in which only one factor is varied in any experiment, the remaining factors being held constant. It fails to look for interactions among the factors and in locating the true optimum when interaction effects are present. The multivariable design of experiments (DOE) is a powerful approach for discovering a set of design or process variables, which are most important to the process and then determine at what levels these variables must be kept to optimise the response characteristic of interest. DOE, in contrast to the one factor method (OFAT), advocates the changing of many factors simultaneously in a systematic way. Before an experiment can be conducted, the experimental design must be carefully planned to ensure that experimental objectives can be accomplished. The values of the independent variables define the experimental conditions or the design of the experiment. Experimental design refers to the formal plan in which the experiment will be set up and conducted and the plans for data collection and analysis. In many process development and manufacturing applications, the number of potential input variables is large. Screening designs are therefore used in the initial phases of a study when you wish to investigate the main effects of several factors simultaneously. Response surface methodology is then used to determine how a response is affected by a set of statistically significant variables over some specified region or expected operating ranges. The central composite design can fit a full quadratic model and it is the most popular of the many classes of response surface designs.