The Six Sigma Methodology – Hypothesis Testing has applicability in the execution of improvement efforts to processes via a Lean Six Sigma Define-Measure-Analyze-Improve-Control (DMAIC) roadmap or the enterprise as a whole.
With both efforts, it is important to first determine what measurement is to be improved through this undertaking. At the enterprise level, one could be interested in improving an overall corporate on-time delivery key performance indicator (KPI), while at a local process level a similar objective might be to improve on-time delivery of a product from a particular facility.
Next, for both efforts, it is important to track the particular metric in a format so that when improvement efforts occur they can be identified and quantified through the charting. Traditional forms of reporting are not adequate:
A 30,000-foot-level format for reporting provides a very desirable form of tracking, noting that the individuals chart in this form of a report-out has differences from traditional control charting. With 30,000-foot-level reporting, change can be identified through a shift of the control chart. Data from the recent region of stability can be used to estimate what is expected in the future, as noted in the statement at the bottom of the chart.
One should not react to each up-and-down of data movement in a region of stability. This variation is considered common cause; however, a measurement that is not within the statistically calculated upper and lower control limits (UCL and LCL), is considered to be special cause. In the above chart, a special cause downward shift occurred; i.e., a downward change occurred in the process.
With 30,000-foot-level reporting, statements at the bottom of the chart are made relative to the stability of the process. If a process has a recent region of stability, the process is said to be predictable. If the prediction response is not desirable, the process needs to be improved.
Six Sigma Methodology – Hypothesis Testing
When a process has a recent region of stability and improvement is desired, the Six Sigma methodology – hypothesis testing approach can be useful to determine where improvement efforts should focus.
For example, data from the recent region of stability, as identified by a 30,000-foot-level chart, can be used to statistically test for the following:
- Differences between machines
- Differences between facilities
- Differences between day of the week
- Differences between time of day
- Differences between suppliers
- Differences between customers
Insight from the Six Sigma methodology – hypothesis testing can be useful to gain insight to what could be done differently in a process. This form of thinking can transition organizations from firefighting to identifying and making improvement efforts so that the enterprise as a whole benefits.
Six Sigma Methodology – Hypothesis Testing: Resources
- Lean Six Sigma Book: Integrated Enterprise Excellence Volume III – A Management and Black Belt Guide for Going Beyond Lean Six Sigma and the Balanced Scorecard
- Enterprise Application of Hypothesis Testing book: Integrated Enterprise Excellence Volume II – A Leaders’ Guide for Going Beyond Lean Six Sigma and the Balanced Scorecard
- Article: Positive Metric Performance Poor Business Performance: How Does this Happen?
- Webinar: From Business Chaos to Competitive Advantage
- Training: Lean Six Sigma Green Belt, Lean Six Sigma Black Belt, Lean Six Sigma Master Black Belt training