Leaderboard

P Balakumaaran has highlighted the ease of using Shainin DOE along with the associated risks. Hence, his answer has been selected as the best answer.

DOE was first proposed by R.Fisher in 1920s. As all factors should be taken into account, Full Factorial Designs (FFD) will undoubtedly give the most accurate results. However, this method is not practically preferred, as it includes too much experiments, which is time consuming and costly. For example, when 4 factors with 2-levels, we have to manage 16 experiments (2^n = 24). But in practice, the number factors may not be as small as 4. For example; 15 factors with 2-levels need 2^15 = 32768 experiment to do. Therefore, to reduce the number of experiments, fractional factorial design has been developed. 1. Taguchi Method (TM) Genichi Taguchi simplified classical DOE by using orthogonal arrays (OA). Taguchi created new methods on the improvement of product and process, which includes, "Taguchi Loss Function" and product/process design with three approaches - "System, Parameter and Tolerance Design". He simplified the Fisher's DOE by using Orthogonal Arrays (OA). He used the signal/noise (SN) ratio to reduce variation in the experimental design. TM also used the SN ratio, which is used to predict the loss of quality, to maximize the robust design’s objective function. SN ratio takes the test results’ mean and variance. 2. Shainin Method (SM) The modern approach to the DOE is Shainin Methodology. This strategy is based on detection of the one, two or three dominant causes of the process variations by focusing on a problem response. Dorian Shainin developed this method to reduce the process output variability, It is simple, relatively easy to understand and implement, but uses the combination of powerful statistical techniques, to make it more reliable and faster to achieve results. In this method, the problem of the poor quality and causes of this problem are identified by the colors of Green, Red and Pink. These parameters, named Red X, Pink X and Pale Pink X, are ranked based on Pareto Principle. Green Y: Indicates special quality characteristics that are important to customers Red X: Indicates the dominant cause of the variation and it contains at least 50% of the causes of variation (Green Y) Pink X: Indicates the secondary cause to the overall variation. It includes 20-30% of the Green Y. Pale Pink X: Indicates the tertiary important reason. It causes to 10-15 % of the Green Y . With SM, the analysis variation can be reduced by 75% to 95% for the causes of the Green Y (Red X, Pink X and Pale Pink X). SM has mainly 12 techniques, of which, 9 are problem solving and 3 are controlling and preventing any repetition of the solved problems. Comparison between Taguchi DOE Vs Shainin DOE methods: Also, the Pro's and Con's of the 2 methods are listed below and it helps to choose the best appropriate method, based on the requirement Some of the risks associated with Shainin DOE method is listed below: 1. This method focuses only on the analysis of mean response and does not take into account the variability of 2 different responses. 2. It can help only upto 70%-80% reduction of the problem, as it focuses on Vital few. The impact of the remaining causes are to be accounted with further more iterations. 3. Grouping of the causes and progressive elimination method, may result in eliminating some significant causes.