What is DMAIC?
DMAIC is a data-driven, problem-solving approach that helps make incremental refinements and optimisations to products, designs and business processes. It was created in the 1980s as part of the Six Sigma methodology by Motorola engineer, Bill Smith. The Six Sigma approach was designed to drive continual improvement in manufacturing processes using data and statistics.
What are the different steps of the DMAIC process?
DMAIC has 5 interconnected steps: Define, Measure, Analyse, Improve and Control. Each phase is designed to have a cumulative effect: to build on information and data generated in the previous stages and to be repeated over multiple iterations.
- Define: The Define phase establishes what a problem is and what’s needed to accomplish a solution. This part of the process is where you clearly state your problem, the end goal and the scope it will take to achieve it. This phase helps you understand the process in its entirety and which elements are critical to quality, otherwise referred to as 'CTQs'. Inputs and outputs are usually outlined with a SIPOC diagram – SIPOC being an acronym for suppliers, inputs, process, outputs and customers. This information is typically captured in a project charter document, which sets out the shape of your DMAIC process.
- Measure: Once you’ve understood your process problem, you need to outline how you’re going to observe the changes you make to it. Of course, with a data-driven approach, having good data is essential to the DMAIC process. As such, the purpose of the Measure phase is to establish your current process performance and what data you’ll be analysing. From there, you can use a data collection plan to monitor your performance as you make changes and to compare at the end of the project.
- Analyse: You should now have a baseline of data that you can use to start making decisions about your process. As you might expect, the analyse phase is the perfect time to go through that data. Here, you and your team members will build a current process map using your data to understand where issues in your process begin. While some Six Sigma projects use more complicated tools for this, fishbone diagrams and Pareto charts are perfectly sufficient and commonly-used methods for conducting root cause analysis. Once you’ve identified several root causes, it’s time to involve your team. Have them vote to decide where the focus of your DMAIC process should be moving forward.
- Improve: Finally, it’s time to start making actual improvements to your process. In the Improve phase, work with your team to find creative solutions that can be implemented and measured within the DMAIC process. At this point, brainstorming and having effective meetings are critical for your team. Once you have solutions in mind, you need to test, fail-proof and implement them. Plan-Do-Check-Act or 'PDCA' cycles are a common method for this, combined with Failure Mode and Effects Analysis, or 'FMEA', to anticipate possible issues. This information should be drawn up in a detailed implementation plan, which you can then use to guide the application of solutions in your processes.
- Control: The last step in the DMAIC methodology can help you verify and sustain the success of your solutions for the future. In the Control phase, your team should create a monitoring and control plan to continually reassess the impacts of any implemented process changes. At the same time, you should create a response plan to act upon in the event that performance begins to fall again, and a new problem appears. Being able to look back on how improvements were conducted and what solutions were made can be an invaluable asset. In these moments, having proper documentation and version control on the improvement process is vital.
If you already know a little about DMAIC, you may have heard of a different initial step, known as the Recognise phase. This step involves selecting the right project or problem to tackle in the first place, as not all projects may require as rigorous an approach as DMAIC.
What are the advantages of DMAIC?
In industry, the appeal of Six Sigma and DMAIC was its ability to drive a lean manufacturing approach. The goal was eliminating waste, defects and over-production as much as possible – the name Six Sigma is actually derived from the statistical model used. In statistics, a standard deviation is also referred to as a 'Sigma' or σ. Manufacturers found that the more standard deviations there were between their average and acceptable quality limits, the less likely they were to go over those limits. Thus, Six Sigma, or six standard deviations, became the gold standard in defining production limits, reducing the number of defects and improving processes.
In actual statistical terms, having Six Sigma Quality means having a defect rate of 3.4 per million opportunities in your processes. Some industries, such as pharmaceutical or aeroplane manufacturers, may use an even higher sigma level in production due to the possible consequences of defects.
This statistical approach to process improvement, along with other methodologies such as Kaizen, was vital to driving improvement in manufacturing in the 90s and 2000s. But how does DMAIC apply to other industries today?
The core benefit of DMAIC is that it is an incredibly rigorous problem-solving approach. Experiments with new processes or changes to workflows can often end up not yielding definitive answers as to what’s changed, no matter your business. But by using DMAIC, a data-driven approach with defined objectives and stages designed to objectively assess whether changes to processes have been positive, you’re in a better position to assess measurable progress. Whether it’s identifying the real roots of your problems or the actual impacts of your experiments, DMAIC equips you with the data to do this.
DMAIC provides a standardised, structured framework for making change, and the clear 5-step approach keeps everyone abreast of every stage of the process. This enables you to produce documentation to summarise all decisions and progress made, so you can move seamlessly from step to step. And by having a definitive process, you can help guide teams in terms of what their current objectives are and where their work is headed.
One of the best things about DMAIC is that it’s a highly repeatable process designed to be ongoing. The Control phase of the process necessitates that process owners continue to monitor the impacts of optimisation. With a data monitoring plan already in place, your new process data naturally forms the baseline for a new Measure phase. Your response plan could also dovetail into a new DMAIC process in the case of initial improvements failing to sustain performance. As such, the DMAIC methodology is able to continually identify problems or impacts on performance, with a method to immediately target those issues.
In this way, the DMAIC framework provides a structured approach that can guide your organisation through repeated, incremental optimisations of business processes. No matter what sector you’re in or the size of your team or organisation, it’s a tested method for driving continual improvements in performance.
When should you use DMAIC?
As mentioned earlier, a common adjustment to DMAIC is the added Recognise phase: a period of identifying whether a problem is suitable for DMAIC. The reason for this step is that DMAIC is a process that requires a high degree of alignment and effort, as well as time to understand. An organisation needs to implement and practice DMAIC to see how it works best for their teams, their industry and with their organisational culture. At the same time, organisations need to consider where to best allocate their resources: which process or team stands to benefit most from DMAIC?
As such, it is not necessarily the go-to option for any issue or problem. If a process issue is identifiable, and a simple solution backed by compelling data is available, then a full DMAIC process may not be necessary.
However, when the process problem is complex or involves high risk – perhaps where performance decreases can’t be afforded – DMAIC is the preferable approach. Even if resource costs are higher, DMAIC ensures that procedure is followed and crucial steps are not skipped, increasing the chances of successful implementation.
Improving processes with DMAIC
It can be daunting to attempt a DMAIC project for the first time, especially if there are major problems that your organisation needs to solve. What’s critical to understand with DMAIC is that it’s not an out-of-the-box best practice approach for your organisation or business. Rather, it’s a discovery process that allows you to find the best practices for your teams and organisation through iteration and incremental improvement. While not all problems your organisation faces will require DMAIC, introducing it can start to drive real change and improvements to your processes, which, ultimately, can lead to measurable success for your company.