Beyond the Buzzwords: Applications of Machine Learning in Lean Six Sigma
Presenter: Cheryl Pammer, Sr. Advisory User Experience Designer, Minitab, Inc., San Diego, CA, USA
Co-Presenter: Charles Harrison, Statistician, Minitab, Inc. San Diego, CA, USA
Keywords: Machine Learning, Predictive Modeling, Preventative Maintenance
Industry: Automotive, Chemical, Manufacturing
As we collect more and more observational data from our processes, we need new tools to provide meaningful insights into this information. We will discuss how to use modern-day machine learning techniques, such as Classification and Regression Trees (CART), alongside traditional lean six sigma tools to analyze, improve, and control your processes.
Through the use of case studies based on real experiences, you will learn the basics around machine learning techniques and move beyond classical regression analysis to build predictive models that extract value from complex datasets.
In the first case study, you will learn how to quickly detect the root cause of an out-of-control process condition when no assignable cause is immediately apparent. Specifically, we will use machine learning techniques to determine which variables are the largest contributors to the process drifting out of control and then improve the process using this information.
In the second case study, you will see how to use data from machine sensors to predict when failures are likely to occur. This information is then used during the project’s control phase as part of a preventative maintenance plan.