Six Sigma is one of the most well known methods to transform whole organizations into a data-driven and continuously improving company. At the heart of Six Sigma is the well-known Define Measure Analyze Improve Control (DMAIC) Cycle, which is integral in order to achieve organizational wide continuous improvements.
The main goal of the Define phase is to select an appropriate project and define the defect at hand. The project’s focus is crucial and usually assessed from the customer’s perspective as well as main painpoints specific to the organization’s product portfolio. In general, the customer’s voice and the organization’s pain points will be captured in corporate KPIs such as energy per ton of production, asset utilization, overall yield or product end quality. The detailed nature of the KPIs depends on the industry and respective endmarket requirements demanded by the customer. In order to focus on the right projects that are connected to improvements of strategic KPIs, appropriate prioritization is key, especially since traditionallyconducted Six Sigma projects can take a couple of months or longer. The results of the define phase should be captured in a project charter. The project usually is carried out by a team of process stakeholders with Six Sigma belts and process experts who know the production process in great detail.
The Measure phase aims to assemble measurable data about the process performance (Y’s relevant to the project outcome) in order to define the baseline performance as well as the project scope based on the baseline performance. Another part of the Measure phase is a measurement system analysis to ensure that the measurements are trustworthy. In the case of companies from process industry, all measurements recorded are stored in a historian system and can be accessed via standardized connectors.
In the Analyze phase it is all about determination and verification of the root causes of the process variation of interest: “Connecting the X’s to the Y’s”. The root causes should be sorted by impact and a selection of vital root causes needs to be performed based on the important and often forgotten principle “correlation is not causation”. The commonly used tools used in this phase are a variety of analytical hypothesis tests based on specific statistical assumptions such as normal distribution. The aim here is to choose the right hypothesis test that fits the data structure and problem at hand so previously stated hypothesis can be (in)validated.
During the Improve phase, focus is put on identification of changes/actions to take so that the defects and variation caused by the root cause is reduced. The selected changes/actions should be implemented while measurable improvement w.r.t. baseline performance is continuously evaluated. Within the process industry, those changes/actions are usually taken during the running production and an effect should be visible in the data recorded in the process historian
Since Six Sigma is a method for continuous improvement of processes the Control phase plays a vital part in the DMAIC framework. Here, the continued improved performance is at the center of attention. Ideally, a Six Sigma project does not just stop when finished but the results live within the organization for continuous sustained improvement. Additionally the results should be checked for applicability elsewhere in the organization such as other plants which produce the same product.
Low-code and No-code AI & Analytics Solutions Yield High-Value Use Industrial Use Cases13 September 2024 - 15:19
Building a Foundation for AI & Analytics with an Industrial Data Fabric30 August 2024 - 23:12
TrendMiner’s 2024.R2 Release; A Complete Platform to Facilitate Event Analytics and Operationalize Industrial Data8 August 2024 - 16:38