Learn to find the reasons for problems through data. As soon as you learn to put potential causes into your problem in visually diagrams/graphs through statistical methods, you’ll quickly see where optimization or new initiatives may be needed. A root cause analysis streamlines the problem-solving process and provide the basis for further development and optimization of processes.
In other words, you learn to find and argue for the root cause of a problem.
On this course, you learn to optimize through insight into probability and probability distributions. You also learn to use tools like hypothesis testing and regression analysis.
You need tools and theory for how to find reasons for the problems you work with through data, and how to communicate the statistics to your colleagues.
You could be employed in a development function, work with quality assurance or improvement of processes in general. You need to conduct statistical analyses that can provide help developing and optimizing your products or processes.
If you are in the process of training as Six Sigma Green or Black Belt this module will be one of the compulsory modules. It helps you to use the data approach in your practical DMAIC project.
What you will learn:
After completing this course, you can understand and use the most common statistical distributions to analyze data. You can develop and test hypotheses through historical data and controlled trials – and can select the correct test to confirm or reject hypotheses.
You can decide what kind of regression, which further helps to demonstrate the root causes. You can model historic data through model reduction, so your data is easier to analyze.
Content in bullets
- Probability and probability distributions – types and application
- Hypothesis testing – hypotheses and choice of test
- One sample t-test
- Two sample t-test
- Chi-square test
- Parred t-test
- None normal distributed data
- Normality Test – check if the data are normally distributed
- Linear regression – correlation between few variables
- Multiple regression – correlation between multivariable
- Binary logistic regression
You should be aware that the course requires that you have participated on Intro to statistics and SAS JMP / Minitab, if you have not previously worked with the statistical programs SAS JMP or Minitab. We recommend using SAS JMP, but if you or your company already work in Minitab, we will, of course, support this choice.
Before the start of the course, you need to have either SAS JMP or Minitab downloaded on your own PC.
This course can be extended with the courses in statistical Design of Experiment, also called DOE (Design Of Experiment) – a statistical tool to test which factors effects a problem the most:
In these modules, more sophisticated statistical tools are built. Just as you will be able to use the Design of Experiment to optimize your actual input.
If you would like further focus on developing performance and processes, you can seek more methodical inspiration on the module Design for Six Sigma, DfSS or on the module Quality Function Deployment, QFD, where the actual customer requirements for a new process are mapped, operationalized and the building blocks for a future process are created.