Orthogonal Array Experimental Design

In this study, an orthogonal array experiment was designed to optimize the synthesis of photocatalyst. This chapter provides readers with an important foundation to understand the terms of the experimental design (DOE) and practical applications. As described later in this article, actually using DOE can dramatically reduce optimization procedure, catalyst synthesis, or other aspects of time and effort. This section describes the general process of experimental design and some general design methods.

By using the orthogonal array experimental design proposed by Taguchi it is possible to examine the influence of many different parameters on the performance characteristics of the coagulation experiment group. Once the parameters that influence the controllable process are determined, it is necessary to determine the level at which these parameters should be changed. To determine the level of the variable to be tested, you need to understand the process in detail, such as the minimum, maximum and current values ​​of the parameter. If the difference between the minimum and maximum of the parameter is large, you can increase the test value further or test more values. If the range of parameters is narrow, you can reduce the value to be tested or approach the value of the test. In the example of the previous jacket temperature, 60 levels at 1 degree interval would be an exorbitant cost.

There are three basic types of experimental design. These include pre-experimental plans, actual experimental designs and quasi-experimental plans. The extent to which researchers assigned subjects to conditions and groups has differentiated the type of experimental design. This module will focus on different kinds of actual experimental design. The actual experimental design is characterized by a random selection of participants and a random assignment of participants to the group in the study. Researchers can also fully control external variables. Therefore, you can confidently determine that the influence on the dependent variable is directly attributable to the operation of the independent variable. For these reasons, true experimental design methods are often considered the best type of research planning method. There are several types of actual experimental design.