What if your process has poor yields? What if it can’t produce defect free items or services? Or it takes too long? Or is there so much variation that it is impossible to plan, schedule, staff, or budget accurately? And what if you have no idea what’s causing the problem? That’s when Six Sigma DMAIC projects are required.
Six Sigma is a disciplined, statistical-based, data-driven approach and continuous improvement methodology for eliminating defects in a product, process or service. … Six Sigma can also be thought of as a measure of process performance, with Six Sigma being the goal, based on the defects per million.
Six Sigma DMAIC projects are what you need when you don’t know the root causes, factors, or drivers of performance. DMAIC are the five phases of Six Sigma projects and they are define, measure, analyze, improve, and control.
What Is Six Sigma?
Generally, Six Sigma is a problem-solving methodology that helps enhance business and organisational operations. It can also be defined in a number of other ways:
- A quality level of 3.4 defects per million opportunities
- A rate of improvement of 70 percent or better
- A data-driven, problem-solving methodology
- Real problem – Statistical analysis – statistical solution – Real Solution
- An initiative taken on by organizations to create bottom-line breakthrough change
Six Sigma Principles
Six Sigma is based on a handful of basic principles, and these principles create the entire Six Sigma arrangement. Here are Six Sigma’s fundamental principles:
- Y=f(X) + ε: All outcomes and results (theY) are determined by inputs (the Xs) with some degree of uncertainty (å).
- To change or improve results (the Y), you have to focus on the inputs (the Xs), modify them, and control them.
- Variation is everywhere, and it degrades consistent, good performance. Your job is to find it and minimize it!
- Valid measurements and data are required foundations for consistent, breakthrough improvement.
- Only a critical few inputs have significant effect on the output. Concentrate on the critical few.
- Every decision and conclusion has risk (ε), which must be weighed against the context of the decision.
The underlying premise of Six Sigma is the equation Y equals f of X or Y is a function of X. Y is the outcome and the Xs that belong in that equation are the causes, or factors, that impact the outcome. DMAIC is data driven. Data is used to understand, analyze, and determine the key Xs that have the biggest impact on the performance of Y. In other words what are the key Xs in the equation Y is a function of X? By knowing which Xs impact Y you can control the Xs to obtain the Y you want. Armed with this knowledge you can move from reactive, detection based interventions to proactive, prevention based routine excellence. In other words from firefighting to fire prevention.
Let’s say you are the owner of an eCommerce reatil store. Complaints have gone up. Refunds to customers increased as a result of complaints and many repeat customers don’t come back. Note that we actually do not know what the root causes of these problems are. So we should launch a Six Sigma DMAIC project to address the problem.
Here’s a summary of the five phases using a eCommerce problem as an example.
- Define. In the define phase the project is defined, the team is selected, and project is launched by management. What’s important or critical to customers is understood and the performance outcome to be improved or the Y in Y is a function of X is defined. In our example Y is the number of complaints. You want to reduce the number of complaints. The financial impact is a reduction in the amount of refunds and the potential loss of repeat customers.
- Measure. In the measure phase the size and scope of the problem is understood and performance of Y is measured. In our eCommerce we collect data to measure the number of complaints and the types of complaints. A Pareto (par-rate=o) chart of the complaint data shows that the highest number of complaints is on delivery lead times for the latest fashion accessories and it is persistently the biggest complaint across the whole site. So what do we do? We want to set up our process to proactively prevent poor delivery times to customers on new fashion acessories.
- Analyse. In the analyse phase we analyse data to determine the causes or factors that impact performance. In other words we diagnose and prove which X factors impact Y (we define Y). The more specific the Y the quicker the analysis as to which factors are the key Xs. Back to our eCommerce example what is Y in our Y equals f of X? Is it the number of complaints? Or can we get a more specific Y? Yes, we know that delivery times is the culprit for the majority of the complaints. Applying the Pareto of 80 20 rule we can get the biggest return on our efforts by focusing on delivery times complaints. So our focus Y is delivery complaints. Potentials Xs or potential causes or theories of our eCommerce problem are proposed and we test them using data to prove or disprove which ones are the key Xs or root causes. Say the root cause analysis shows three key Xs are causing poor crust quality. Inconsistent lead times for new accessories, insufficient new stock and delivery logicsts.
- Improve. In the improve phase solutions are developed to address the proven X factors so that Y can be improved. Solutions are developed, tested, piloted, and implemented to optimize the delivery quality. In our example let’s say we determine that vertical suppliers integrate into the value chain to speed up the process or change delivery companies .
- Control. In the control phase controls are established to ensure there are improvements or gains are sustainable. Controls and procedures are put in place so that employees know when and how to intervene to ensure superior performance. For our eCommerce store we make sure they are trained staff on the new procedures and control plans are implemented. Then we would hope that the number of complaints would drop drastically and, in turn, benefit financially from fewer refunds and higher retention of customers.
By determining the key X factors that impact Y we can be proactive in ensuring superior Y performance simply by controlling the key X factors.