1. Repeat the test
This means that at least two tests should be carried out under the same test conditions. The purpose of this is to obtain an effective estimate of the test error by repeating the test. Simply put, the purpose of the experiment is to compare whether the output of the process is different under different factors. If there is a difference and the difference is greater than the random error, it is considered that this factor has a significant impact on the process output.
It should be noted that what we are talking about here is repeated experiments, that is, repeated experiments under the same test conditions (factor level combination), rather than repeated measurements. Of course, when repeated tests are used, the number of tests will increase greatly, resulting in an increase in test costs.
2. Random testing
It refers to arranging the test sequence in a completely random way during the test. The purpose of this is to eliminate the systematic influence and interference caused by the factors that have not been studied in the experiment on the test results. Suppose we want to investigate the influence of different hot oil temperatures on the fracture strength of products. If the test is arranged according to the order of oil temperature from low to high, if the working voltage changes from high to low, the working voltage will also affect the fracture strength, so it is impossible to distinguish whether the change of fracture strength is caused by oil temperature or voltage change. This problem can be avoided if the test order is disturbed.
3. Block testing
It means that the test area is divided into different groups according to the different test conditions and environments, so as to ensure that the tests in each group have little difference in the test environment and test conditions, while allowing large differences between different groups. Through block testing, the influence and interference of test factors on test results can be reduced. For example, in order to study the influence of cutting angle on shaft diameter machining, two raw materials provided by suppliers were used in the experiment. In order to eliminate the influence of different raw materials on the test results, it is necessary to divide the test into two groups according to the raw materials used in the analysis. Both random testing and block testing are aimed at eliminating the influence of factors other than testing on the test results. Generally speaking, in the experimental design, "the grouping of energy zones and the grouping that cannot be grouped are random".