An introduction to the DMAIC process


The DMAIC cycle is just like the PDCA cycle used to improve a process. Therefore it can sometimes be confusing to select the right technique, even for an experienced person.  So how do you choose?

The PDCA cycle is linked to lean while the DMAIC cycle is lined to six sigma.

The PDCA cycle is often used to improve in small steps while the DMAIC process is used to improve in a large step. Therefore the DMAIC process is often also referred to as break-through problem-solving. Because of this, there are often more resources needed when using the DMAIC process and the DMAIC process seem more complex and less straightforward than the PDCA cycle.

Another way of differentiating between the two is by looking at the complexity of the problem. Due to its simplicity, the PDCA cycle is used more frequently for solving problems on a daily basis.

DMAIC is more suited for complex problem solving and needs therefore someone with the right six sigma skillset. This person will have a six sigma belt to indicate his skill level. But we will come back to this later in more detail.

To be honest the cyles have a lot in common As you can see in the picture. This is normal since they have a common parent: the scientific method. The plan phase corresponds with the define, measure, and analyze phase, The do phase corresponds with the improve phase and the check and act phase corresponds with the control phase.  Due to the similarity, you can use both techniques for improving a process. But once you pick one you have to follow it to the end.

Now Lets look at the different steps in more detail:

The Define Phase

The Define phase is the first phase it the DMAIC cycle.

In the define phase, we have to identify the problem which we would like to solve through customer analysis. Often We start from a high-level description of the problem. We have to understand what exactly the problem is the customer experiences and break it down into customer requirements.

 We will also have to check that the project is aligned with the business requirements. We need to understand why solving this problem is important for the business. If we understand the problem from the business and customer perspective then We will be able to quantify the problem and we can also agree on how we will measure that the project is successful.

It is also important that during the defined phase we determine the scope of our project. If the scope is known we can determine which people we need to include in the team. If team members are new to the DMAIC process is could be opportune to train them. This phase often ends with the creation of a project charter which includes the summary of the project; what is the project planning, what is the budget and resources required. We will go more into detail on the project charter in another article.

 The Measure Phase

The measure phase is all about collecting data In order to better understand the problem. First We need to define which data we would like to collect. For this we can use the fishbone diagram.

secondly, we need to define how we would like to collect the data. In order words, we need to create a data collection plan. A  data collection plan ensures that everything that we need will be available for analysis later on. Generally, a data collection plan defines what needs to be measured and how, who needs to do the measurement, and finally where, when, and how many measurements need to be taken. Finally, it defines how the measurement needs to be recorded.

Before we start measuring we have to be certain that we can trust the data we would like to collect. In other words, we have to Assess the capability of the measurement system. For this, we can use a tool called measurement system analysis or short MSA. MSA is typically performed by green belts and black belts.

Do not forget to communicate to the different stakeholders why you are collecting the extra data, and train the people who need to take the measurements if needed.

And finally do not forget to monitor the data collection during the measurement phase. You do not want to find out after a month of measuring that the data they collected can not be used!

The Analyze Phase

The goal of the analyze phase is to find and validate the root cause of the problem which we are having. The data which has been collected during the measure phase is crucial for this. This data has to be transferred into information to prove or disprove the hypotheses as to why the problem exists. There are many statistical tools used in this phase to pinpoint the root cause of the problem.  One of the simplest and often used analyzes are graphical analyzes are  where data is put into a chart to better understand the problem and potential root causes

Another way to look at it is that in the analyze phase the practical problem which we have is being transferred into a statistical problem to better understand the variation and from where it comes.

Before we move to the improvement phase we have to validate that the identified root cause is the real root cause, often this is done by setting up a small test.

The root cause could occur in a different process step than where the symptoms of the problem are visible. Therefore make certain that you talk through the uncovered root cause with the people who are involved in the process step where the root cause is situated.

The Improve Phase

The improvement phase is the  fourth phase in the lean Six Sigma DMAIC cycle.

In this phase, we try to find a solution for the root cause we uncovered during the analyze phase. Just like the previous phases, also this phase is a team effort. The team has to brainstorm on ideas and select the best solution. Before implementing the solution on a full scale a test on a small scale; also called a pilot has to be done. During the pilot, data is collected to show that the solutions really solve the problems and creates a sustainable process. The pilot is also a good way to get the stakeholders involved. This will lower the risk of resistance to the chance and will give you valuable feedback.

The Control Phase

The last phase is the control phase. During the control phase, we continuously monitoring the changes we made in the process in order to be certain that the obtained results are sustainable. For this typically control charts are used. In order to have a sustainable change, we will need also to update all related process documentation suchs as control plans, training plans, workinstruction and audit plans.

This article was updated on November 30, 2022