“What is the minimum sample size?” may be the most common question a statistician or a lean six sigma practitioner will be asked. Everyone assumes that this is a question with a straightforward answer, but it truly a complicated question to answer. Surprisingly the answers to this question are more based on business considerations than …
Statistical analyses is used in various capacities in the 9-step Integrated Enterprise Excellence (IEE) business management system.
In step 2 of the IEE system, an organizational value chain compiles the results of what’s and how’s of past occurrences (i.e., processes) for the creation of statistically determined predictive 30,000-foot-level performance statements for functions throughout an organization.
In steps 4 of the IEE system, the entire enterprise can be assessed statistical to uncover causalities which provide insight to what areas of the business should be further investigated for improvement opportunities.
In step 7 of the IEE system, a targeted business improvement area can be assessed statistically for the uncovering of causal events which provide insight to what might be done differently to improve. Statistical Design of Experiments (DOE) can also be used to provide insight to which factor levels provide an improved process-output response.
The topic of autocorrelation data analysis came up during my Lean Six Sigma Master Black Belt course last week and in two coaching sessions the prior week. Since this is an uncommon topic, it surprised me. I am using the blog to record the answers provided in class and in coaching sessions. Autocorrelation Data Analysis: …
When determining an approach for assessing incoming part quality, the analyst needs to address the question of process stability. If a process is not stable, the test methods and confidence statements cannot be interpreted with much precision. Process control charting techniques can be used to determine the stability of a process.
Organizations need a systematic approach for risk containment when quality, delivery, and design product and service issues occur. Such a system should also help them to recover quickly from errant decisions made by executives, operations personnel, and the quality department. This article describes how well-chosen metrics can help mitigate these risks if the measurements contain good tracking and reporting methods that lead to the most appropriate action.
Time series analysis and its applications is a topic that practitioners can can much when addressing process improvement efforts. I was talking with a Master Black Belt candidate, in the process of being certified, who asked about other statistical tools that a practitioner should learn. Since the candidate is in a company that has a …
At Smarter Solutions, our Lean Six Sigma (LSS) Black Belt (BB) and Master Black Belt (MBB) courses cover analysis phase tools well beyond the new body-of-knowledge (BOK) published by ASQ. Of course we teach the standard, one variable-at-a-time tests: t-tests, f-test, chi-square test, simple regression, and one factor ANOVA. But as Lean Six Sigma has become a commodity that everyone is teaching, the BOK has become watered down. Our belt programs do not cover the non-parametric tests until you reach the MBB level training. Smarter Solutions chose to teach the distribution theory along with the one variable-at-a-time tests in order to show the belt that the non-parametric tests are not really needed. When a BB or MBB learns of these new tools, he will need an advanced tool selection diagram until he becomes familiar with the tools. This advanced tool selection diagram is attached at the end of this document. You should be able to use this decision tree to answer the questions and identify the proper analytical tool to apply. It also has the Minitab Commands needed to open the tool menus.
The stereotypical Lean Six Sigma DMAIC project has a nice process that produces data at a rapid enough rate that you can evaluate improvements during the project, but what options do you have if that is not the case? Using Monte-Carlo simulations can be a powerful way to estimate future performance without having to actually pilot test or prototype the improvements.
Qualitative analysis for decision making when conducting attribute analyses can be deceiving. For this situation, an attribute measurement involves a yes or no response.
What do you do with data that is non-normal and non-transformable data? Read and learn how this quick and simple Minitab resource helps you do just that!
Important Box Cox transformation considerations that should be addressed before applying this data-transforming technique are described in the PDF article below.