The Control phase of DMAIC ensures that we communicate our plans and document the new operations. But getting this right is more than just updating documents and creating training slide decks. Early in my career, I was disappointed when we’d do all the hard work to understand the process, develop new tools and procedures, and implement the new process only to see it not followed, or the organization fall back to old ways and prior performance levels. It was hard for me to accept that people revert to their old ways of working (where they are comfortable) or, worse, veer off into something completely unintended. The overarching problem with these projects was that process deployments lacked control. I’ll briefly explore why this is so difficult and the approaches I’ve learned over the years to ensure processes are controlled.
Every process has variation. When a new process is implemented, there is variation in training, in the gap between practical application and the intended design, in how the change is communicated, and among the operators themselves. Short of a completely automated process (and even those have variation), it is very difficult to eliminate variability and produce a consistently controlled output. This lack of control is an obstacle to the success of the new process. Success in the Control phase means minimizing and managing these variations so that we can sustain the new process.
The most common variation I observe is the result of inconsistent management and oversight. Years ago, I worked with a police department to improve some of their internal administrative processes. When it came to field work, they were excellent. They had well-documented processes that were closely followed by officers. The internal administrative work was different. Processes such as record processing varied by shift, by person, and by many other factors. One shift handled updating and filing records completely differently than another. There were identifiable best practices, but no one was responsible for enforcing them. Why was field work consistently followed while administrative work was not? It came down to oversight and emphasis. There were significant repercussions for not following processes in the field, internal affairs involvement and legal ramifications created strong incentives for control. Field work received extensive oversight, training, and leadership reinforcement. None of this existed on the administrative side. As a result, administrative processes varied and new processes were not sustained.
When deploying a new process, it is important to recognize that process change does not resolve process management issues, it often makes them worse. To address this source of variation, you must understand the management practices in the department where the change is occurring. If they lack process audits, leader standard work, and daily metric checks, these will need to be established or the process change won’t stick. Otherwise, you will be left wondering why the new process has reverted to old ways. In fact, it is best to address process management problems before starting a DMAIC project. I learned the hard way that if management allows variation in the current process, you should expect the same or worse in the future-state process.
Even excellent daily management practices are not enough to eliminate process variation completely. The more complex the process, the more likely people are to deviate from it, or the harder it becomes to find individuals who can perform the work effectively. While developing the solution, we need to think ahead and reduce opportunities for variation. This sets us up for a successful Control phase. Solutions like full automation or structured software workflows can enforce the new process in ways that significantly limit variation. The team needs to ask, “Is the new process more complex?” If the answer is yes, then current team members may not have the skills needed to manage that complexity. They may also experience fatigue that wasn’t previously a factor. All of these sources of variation need to be identified and addressed. When developing a solution, the team must anticipate potential errors or variations introduced by those performing or overseeing the process.
Error proofing is a powerful method for making it difficult, or even impossible, not to follow the process. In manufacturing, we often build fixtures that force operators to load or assemble parts in only the correct way. In administrative processes, we use software to guide workflow and require specific inputs at each step. This is the mindset the team needs to adopt. Visualizing the correct way to perform the work and visualizing potential errors helps the organization maintain the standard. All of these controls must be supported by strong daily management practices.
The control phase of DMAIC is crucial to ensuring that the process is sustainable. You must understand and address the sources of variation that prevent consistent adoption and adherence. Creating daily management practices is essential, but it will always be an imperfect and resource-intensive way to ensure consistency. Certain characteristics of your solution can either strengthen or weaken your ability to control the process. A new process should simplify people’s work. Complexity breeds variation, and variation is difficult to monitor and manage. As you develop new processes, ask yourself whether you are simplifying the work or making it more complex. If it is more complex, you will need to invest more time in workflow controls and error proofing.
If you’ve been following my recent newsletters, I’ve covered all five phases of DMAIC. Hopefully, these articles have helped you understand the intent of each phase and how to approach them successfully. If you haven’t read them all, I encourage you to go back and review the previous articles. DMAIC is not necessary for every problem. For those new to continuous improvement, it’s often best to keep things simple with Plan-Do-Check-Act (PDCA) and focus on engaging more people in problem-solving. But for those tackling highly complex challenges, DMAIC is a powerful methodology to help ensure success.