DMAIC is Six Sigma’s signature framework for process improvement. It provides a structured way of improving a process.
A DMAIC improvement process:
- Is data-driven.
- Is followed in a strict sequence.
- Uses all five steps.
- The Five DMAIC Stages
The five stages we use are:
- Define: Collate what we already know about the existing process.
- Measure: Collect further data about the existing process.
- Analyze: Identify the core problems that we’ll address.
- Improve: Plan, test, and implement solutions.
- Control: Set up supports to ensure that successful solutions are sustainable.
Note how this is reminiscent of the Deming wheel or Shewhart Wheel: Plan-Do-Study-Act).
Should you use DMAIC?
- Some projects aren’t suitable for this process. For example:
- It’s already very clear what the problem is and how to solve it.
- There’s no or little available data on the process to be improved.
- Managers and leaders do not support improvements to this process.
- The process does not directly impact key performance indicators.
- Measuring process improvements would be difficult or impossible.
History of DMAIC
Michel Harry & Bill Smith created “MAIC” – the methodology that evolved to become DMAIC.
Harry includes the following strategy elements in the traditional approach to Six Sigma:
- (R) Recognize the true state of your business
- (S) Standardize the systems that prove to be best-in-class
- (I) Integrate best-in-class systems into the strategic planning framework.
In the Define phase, you collate a lot of information you already have available. You’ll:
- Understand the project, including its purpose and scope.
- Map the current process.
- Determine whether the process is a good candidate for DMAIC.
- Detail customer expectations.
- Estimate timelines and costs.
You’ve mapped the existing process, understood the project, and decided that this is a good DMAIC candidate. You’ve listed customer expectations and estimated the times and costs involved.
Your next phase involves a lot of measurements. You need to have baseline figures to assess progress accurately in later phases.
During this phase, you will:
- Identify the data that you need to collect.
- Decide what measurements to use.
- Figure out what methods to use to collect your measurements.
- Determine the level of variation that you’ll be dealing with.
- Collect the data as per previous points.
In the Analyze phase, you work with the data that you collected in the Measure phase. You’ll:
- Identify defect causes.
- Analyze these to pinpoint the root cause.
You’ve identified the root cause of your issue in the Analyze phase. Now you need to come up with a solution. You’ll:
- Pull in people who perform or oversee the process.
- Brainstorm potential solutions.
- Determine criteria for selecting a solution.
- Weigh potential solutions against the criteria.
- Pick a solution.
- Test the chosen solution.
- Measure the results and compare them to the Measure phase data.
Once you’re happy that the chosen solution will improve the process, it’s time to implement the Control phase. This is where you actually implement the said solution, but there are some other tasks too:
- Document the solution.
- Collect data about how the solution is working in production.
- Put supports in place to ensure the solution is permanent, not temporary.
- Set up a plan to deal with any issues that might arise.
- Plan handover to the operations personnel.