2. Design of Experiment (DOE): Plan to arrange a batch of experiments, carry out these experiments under the set conditions in strict accordance with the plan, obtain new data, and then analyze them to obtain the required information, so as to obtain the best improvement method. Experimental design has now formed a relatively complete theoretical system, and the experimental design scheme can be roughly divided into three levels. The first-level experimental design is the most basic experimental design scheme, including partial factor design, total factor design and response surface design (RSM). The second-level experimental design includes Taguchi design (robust parameter design) and mixed design. With the development of modern industry, these two levels of experimental design scheme can no longer meet the more demanding and personalized experimental design scheme, so the third level of experimental design scheme was born, including nonlinear design, space filling design (uniform design), extended design, tolerance design, customized experimental design and so on. Among these experimental design methods, the customized experimental design method is particularly worth mentioning. The traditional experimental design scheme is relatively fixed. When the actual problem deviates from the model of the experimental design scheme, the experimenter often needs to correct the problems he has studied to match these traditional experimental design methods. However, customized experimental design is just the opposite, allowing experimenters to reasonably modify the model of experimental design method to make it suitable for the problem to be solved.
Customized experimental design method can be said to be a revolution in the field of experimental design. It allows experimenters to customize the number and weight of response variables (y), the constraints of test factors, the influence to be considered in the test model, and even the number of tests. Experimental design is one of the most important methods in design for six sigma, but it can't be realized without the support of professional Six Sigma software. JMP As far as the function of experimental design is concerned, among the above three levels of experimental design methods, the six sigma software on the market can only support the first and second levels of experimental design schemes, but cannot support the third level of experimental design schemes. In contrast, JMP software can well complete all the above three-level experimental design schemes. In particular, the support for customized experimental design can be said to be a major feature of JMP, which enables experimenters to customize models to meet the needs of practical problems. In the later stage of experimental design scheme, the simulation function integrated in JMP software can also help experimenters to simulate the design scheme, thus minimizing the risk of failure of the new scheme. I consulted six professionals in the quality management industry of Sima. It is understood that JMP has great advantages in graphical interface and supporting the implementation of Six Sigma quality management (such as statistical process control (SPC), conventional regression and variance analysis, etc.). ).
3.QFD (house of quality) method is a tool to help implementers transform customer requirements into specific characteristics of products, which is developed from seven dimensions: customer requirements and importance, engineering measures, relationship matrix, indicators and importance of engineering measures, correlation matrix, market competitiveness evaluation and technical competitiveness evaluation.