According to Six Sigma Global Institute, SSGI, DMAIC tools have become an indispensable part of the Lean Six Sigma Process because they provide us with a proven blueprint for improving process efficiency and then monitoring that process to assure the results are consistent over time. By following the guidelines of DMAIC, SSGI has found that project success rates almost always increase.
D – Define
A – Analyze
I – Improve
C – Control
Define is the process whereby we carefully and completely understand and then summarize a problem. In most situations it is useful to submit the summary in writing. But, every effort must be made to uncover the real cause or what we call the Root Cause of the problem. By taking the time to assure that the root cause has been isolated, we can be confident that real cause of the problem has been identified not just its symptom.
Often there is more than one root cause. One DMAIC tool that is very useful is an Ordered Histogram. A histogram is actually a bar chart like the one shown to the left. It identifies the possible sources of the problem and the frequency with these problems have been observed to occur. In an ordered histogram, the tallest bar is positioned on the left with the shortest on the right. In this example, the most frequent problem, wrong item sent, is followed by the next most frequent problem, damaged packaging. This chart helps to set priorities. The problem to the left is addressed first, then the next most frequent problem and so on.
When we define a problem In Lean Six Sigma we make every effort to measure it. We refer to these measurements as Metrics. They may include revenues, costs, time, number of staff required to operate a process, number of returns and so on. For example, the table shown to the left summarizes the reasons why items have been returned, such as Wrong Item Sent and the frequency for these categories. So Wrong Item Sent occurred on 332 occasions. By objectively defining and measuring these dimensions we have a concrete way of defining a problem and then comparing project efforts on a before and after basis. In the absence of metrics, project problems and the results from the Six Sigma project are less clear or even ambiguous. The DMAIC tools we use include database query tools to extract data from databases, questionnaires, interviews and surveys.
Analyze is the step where we study the problem in detail. Very helpful are DMAIC tools such as value stream and process maps that display the steps in a process and provide us with a visual description of the way in which the process actually performs and is expected to perform. They help identify where waste and delays occur, important factors as we struggle to improve process efficiency.
In this step we determine what it is that needs to be changed. Often we return to the value stream and process maps. Once decided, we then engage the appropriate resources and take steps to implement these changes. For example, in the order process shown below, delays between every step prevent the company from adopting a same-day shipping policy. By reducing these delays, the process efficiency improves, orders are shipped the day they are received, and, above all, customer value and satisfaction increases.
Once improved, a process needs continual monitoring to assure that performance does not deteriorate over time. Control is executed by using another DMAIC tool, Control Charts, like the one shown to the left. They have a center line to identify process targets. Upper Control Limits, UCL, and Lower Control Limits, LCL, are established to identify the points beyond which sample process results are beyond acceptable levels. For example, once an order fulfillment process has been improved, control charts are used to monitor the number of minutes between order receipt and shipment. If this number increases beyond the Upper Control Limit, the process needs to be examined and the root cause of the problem determined.
These are the Basic Tools of DMAIC. Together, according to SSGI, they provide us with a framework for addressing processing problems and opportunities to improve performance over a wide range of applications from product development to patient flows in hospitals. Indeed, these tools have proved so successful that they have found widespread use across industries and across the world.