Hướng dẫn sử dụng Minitab cụ thể, chi tiết. Có phần câu hỏi kiểm tra lại kiến thức. Hướng dẫn tất cả các công cụ trong minitab Hữu ích cho người mới làm quen, tiếp xúc với phần mềm liên hệ : dsnghia96gmail.com nếu có thắc mắc về bài giảng
Minitab: Course Introduction What Is Minitab? Minitab is a statistical software package designed for Six Sigma practitioners It was developed at the Pennsylvania State University in 1972 by researchers: ● ● ● Barbara F Ryan Thomas A Ryan Jr Brian L Joiner What Is Minitab? Simplification of input process Greater focus on identifying trends and patterns Automation of calculations Analysis and interpretation of data and results Creation of graphs Find solutions to the problem at hand Why Minitab? Provides a quick and effective solution for complex Six Sigma projects Has a very user-friendly interface Has several features that help Six Sigma practitioners work with data and statistics Minitab over Other Tools Minitab Stats Package for Social Sciences (SPSS) Microsoft Excel Minitab • Used for process improvement, quality management, and Six Sigma • Contains excellent support and infrastructure blogs • Is easy to diagnose and correct errors • Has inbuilt functions, easy-to-make graphs, and automated analysis of complex data Stats Package for Social Sciences (SPSS) • Is used in research in the field of social sciences • Does not offer any blog-based support • Is easy to diagnose and correct errors • Offers automated analysis of data and easy-tomake graphs MS Excel • Performs complex analysis by building macros and using formulas • Uses spreadsheets to organize data with formulas and functions • Has support functionality • Is cumbersome while analyzing and diagnosing an error • Has very few inbuilt functions • Does not support many Six Sigma tools Target Audience Students who need help in understanding the concepts of statistics and in applying the different methods to solve problems ● Innovation, transformation, and change leaders ● Professionals managing Lean Six Sigma teams ● Lean Six Sigma practitioners involved in highimpact, transformational projects Target Audience Professionals involved in process control, quality, and improvement Aspirants for data analytics, research, process engineering, and reengineering initiatives Aspirants for Lean improvement, waste reduction, production, and service efficiency Factorial Design Experiment The goal is to test the effect of all the three factors and determine the one with greatest effect on MPG Response: Fuel Efficiency Factors Levels Levels Car Speed (X1) 55 60 Octane Rating (X2) 85 90 Tire Pressure (X3) 30 35 Experiment runs: 2^3 = 2*2*2 = Demo: Full Factorial Design • You can see the Mileage column which is the result of various combination of factors and respective levels Demo: Factorial Regression • We shall consider R Square for the study, which is 34.11 and does not suggest a strong model • None of the p values are below α level of 0.05 which means none of the factors (or interactions) are significantly impacting the Mileage except the three interactions of all three factors whose value is higher than α level of 0.05, but it is on border line so can still be considered a significant factor Demo: Main Effects Plot for Mileage As it can be seen from Main effects plot, Car Speed at 55, Octane Rating at 90, and Tire Pressure at 30 is the optimum setting for the given factors Demo: Interaction Plot for Mileage It is evident from the interaction plots as below: ● Car Speed and Octane Rating interact for a mileage of close to 25.5 ● Car Speed and Tire Pressure interact for a mileage of close to 25.5 ● However, no interaction is seen for Octane Rating and Tire Pressure, but they come close to mileage of 25 Demo: Optimization Plot ● The optimization plot shows how different experimental settings affect the predicted responses for a stored model ● It shows the effect of each factor (columns) on the responses or composite desirability (rows) ● The vertical red lines on the graph represent the current factor settings ● The numbers displayed at the top of a column show the current factor level settings (in red) The horizontal blue lines and numbers represent the responses for the current factor level ● In the current setting, we see that with Octane Rating at 90, Car Speed at 55 and Tire Pressure at 30, we get a mileage of 27.4680 Demo: Response Optimization Minitab creates the response optimization model which has optimum setting for all factor levels to get the optimum mileage Basis the prediction model and the response optimization setting, Minitab created the prediction for following factor levels Key Takeaways DOE is a structured method for determining relationship between input factors, affecting the process output by conducting experiments The full factorial approach is used to determine the factors that have a statistically significant effect on the response variables The response optimizer tool states how different experimental settings affect the predicted responses for a stored model Knowledge Check Knowledge Check A matchstick company wants to improve their quality of sticks by studying parameters such as optimised chemical deposit, thickness, moisture content of stick In this case, the company should employ _ method A Design of experiment B Analysis of variance C Main effects plot D Correlation analysis Knowledge Check A matchstick company wants to improve their quality of sticks by studying parameters such as optimised chemical deposit, thickness, moisture content of stick In this case, the company should employ _ method A Design of experiment B Analysis of variance C Main effects plot D Correlation analysis The correct answer is A Design of Engineering (DOE) is a structured method for determining relationship between input factors, X, affecting the process output, Y, by conducting experiments DOE is a pre-emptive experiment done to Knowledge Check Which of the following is true of "One factor at a time" approach? A All variables or factors are held constant in an experiment B Value of every factor is changed simultaneously C When a factor is altered, all the other factors are held constant D None of the options Knowledge Check Which of the following is true of "One factor at a time" approach? A All variables or factors are held constant in an experiment B Value of every factor is changed simultaneously C When a factor is altered, all the other factors are held constant D None of the options The correct answer is C According to the "One factor at a time" approach, one change and it affects one factor at a time, and other factors are kept constant Knowledge Check The _ plot tells us how different experimental settings affect the predicted responses for a stored model A Main effects B Interaction C Optimization D Scatter Knowledge Check The _ plot tells us how different experimental settings affect the predicted responses for a stored model A Main effects B Interaction C Optimization D Scatter The correct answer is C The optimization (a Minitab Response Optimizer tool) plot tells us how different experimental settings affect the predicted responses for a stored model ... Features of Minitab Minitab is a much more proficient tool to perform an in-depth exploration Industry experts prefer Minitab over other conventional tools Standout Features of Minitab The features... learners can refer to Introduction to Minitab Learning Objectives By the end of this lesson, you will be able to: Outline the importance of Minitab Use the Minitab to perform various analyses List... Sigma practitioners work with data and statistics Minitab over Other Tools Minitab Stats Package for Social Sciences (SPSS) Microsoft Excel Minitab • Used for process improvement, quality management,