How Do I Select a “Good” Lean Six Sigma Project?
Choosing the right Lean Six Sigma project is critical to the success of the project. Choosing a project incorrectly could lead to a project that stalls or fails to produce the projected outcome. Knowing the right criteria will help to avoid project failure.
Before we discuss the criteria of a good Lean Six Sigma project, let’s discuss important red flags of a “BAD” project (one that will most likely fail).
Red Flags of Potential Project Failure:
If your project fits any one of these criteria, re-evaluate whether the project in a good candidate for a Lean Six Sigma Belt project.
- Project does not directly affect one or more organizational/site/departmental Key Performance Indicator(s) (KPIs)
- Little or no Buy-in exists for the project at the executive level nor throughout the functions that the project will touch (this usually leads to “turf wars”)
- The scope of the project is too large (i.e., we have been tasked with “solving world hunger”)
- The potential Belt will be the only person on the team (this will lead to the Belt biasing the solution)
- The proposed project is to justify a pre-determine improvement (in other words, you already know the “fix”)
- The Customer Requirement or Specifications are not known in metrics (we will not be able to reach a goal unless we know what the goal is)
- We have little to no control over the process to change inputs or to redesign the process.
- There is no accessible data from the process input(s), output(s), or the data doesn’t exist.
What are the Criteria of a GOOD Lean Six Sigma Project?
Project aligns directly with strategic direction of the organization and should have a direct effect on one or more organizational/site/departmental Key Performance Indicators (KPIs).
Companies tend to arbitrarily pick projects based on where they see the “pain” presently instead of letting the data point to the next most important project. Projects should be identified based on their effects of a company’s key performance indicators (or KPIs).
By establishing KPIs at each level of the organization, implementing a project hopper so that everyone in the organization can identify problems and prioritizing those problems to their effect on the KPIs, the organization will be able to identify the next best project.
The functions which the project touches are “Bought-in” to the project and it’s benefits and will therefore support the Belt in project completion.
If you do not have buy-in from your leadership and the functions that the project touches, then you will most likely encounter resistance. Leadership and the functions that the project touches must understand their role in the success of the project. They need to understand how they will benefit from the project.
In an effective implementation of Lean Six Sigma this buy-in is achieved through training at each level of the organization. Trainings like the “Executives and Champions Workshop” help leadership understand their role in Lean Six Sigma. Other trainings like the “Lean Six Sigma White Belt Certification Training” helps everyone in the scope of the implementation understand their role.
The scope of the project is focused on a small sample of the larger problem (i.e., instead of trying to solve world hunger, we focus on solving hunger in a small village first).
The biggest roadblock to a successful Lean Six Sigma project is taking on a project that is too large in scope.
For example, let’s say you were tasked by your leadership with a Lean Six Sigma project to “Reduce scrap by 50%”. What this problem doesn’t tell you is that the company has 200 different scrap codes (reasons used to identify scrap) across 100 products. This is a monumental scope that will lead to a project that will have little chance of succeeding.
Using that same mandate by leadership to “reduce Scrap by 50%”, we could use a Pareto Chart and scrap data to focus on the one product and scrap code that is causing us the most scrap. The pareto chart becomes a prioritization of products and scrap codes to address in subsequent projects. Planning these multi-generational projects will give leadership a better understanding of the timeline to achieve the goal of 50% reduction in scrap.
The Belt works with a team of Subject Matter Experts in the process. The Belts acts as the “to be the Guide on the Side” using the tools and methodology of Lean Six Sigma as the to the team of SMEs solve the problem.
It happened often in our public Lean Six Sigma courses that an individual comes to class with a potential Belt project where they are the only SME and the only person on the team. This is not a “show-stopper” but this scenario can lead to failure of the project to lead to improvement of the true root cause.
If the potential Belt is the only person on the team and the SME then the Belt will have an increased chance of biasing the solution towards his/her preconceived idea of what the solution should be.
To truly understand the process that is being investigated, the Belt must remove his/her biased from the analysis, let a team of SMEs (other than the Belt) provide the data and analyze that data using the tools and methodology of Lean Six Sigma.
The Belt and team should not “already know the solution” to the problem. A good Lean Six Sigma project has a solution that must be discovered through the process of analysis.
If you already know what the solution is, then why spend time and resource on the Lean Six Sigma project.
Six Sigma is a methodology to answer a question that we don’t yet know the answer. We let the tools and methodology help us to understand why the problem is occurring. From this analysis, we then use the experience and knowledge of the SMEs to develop an effective solution. The improvement in most cases is not known until the Improve phase of the DMAIC process.
We should understand the Customer Requirement or Specifications in a measurable term.
If you do not know where the “finish line” is, how can you finish the race. In the same sense, if you do not know your customer requirements in measurable terms, how can you determine that you have reached the goal of your Lean Six Sigma project.
For example, if you been tasked with “reducing the lead time to process an order”, you need to know 1.) the current time it takes to process an order (we can determine this through historical data) and 2.) you need to know what is the time required by the customer to process an order. Without these two inputs, we will not know where the starting line and the finish line are.
The process the team is investigating should have measurable and controllable inputs (X’s) and output(s) (Y’s). There should be accessible data on the measurements of the inputs (X’s) and output(s) (Y’s).
If our Lean Six Sigma project does not have measurable inputs then we cannot use statistics to understand: 1.) our baseline performance, 2.) the statistically significant inputs to the process, 3.) whether our Improvements can achieve the chartered goal, 4.) and, how to effectively control the process.
What does DATA for a Lean Six Sigma project look like?
The data we should collect helps us to understand the variables in the equation: Y = f (X1, X2, X3, Xn…). The Y in this equation is the metric that we are trying to improve. The Xs are the measurable inputs that create the Y. The “f” (or function) is the effect that the Xs have on the Y (we determine this through statistical tools in the Analyze Phase).
For example, we have been assigned a project to improve the company’s capability to fill bottles with liquid. Our current capability is 390,000 defects per million opportunities (DPMO) and we fill a million bottles every three days. Our goal is 125,000 DPMO.
The process to fill a bottle full of a liquid has the following “X’s”:
- Fill Date (Discrete)
- Fill Time (Discrete)
- Fill Operator (Discrete)
- Fill Line (Discrete)
- Fill Shift (Discrete)
- Fill Weight (Variable)
- Bottle Type (Discrete)
- Liquid Type (Discrete)
These measurable inputs (X’s) combine to produce an Output (Y): Bottle Fill Height
We can use a multi-vari sheet to help us to analyze the bottle filling process.
The Multi Vari Sheet could be an Excel spreadsheet where each row has a measurement for each of the “X’s” and each “Y” (if more than one “Y”).
Here is a picture of a muti-vari sheet using the “bottle filling” scenario above:
In our “Bottle Filling” scenario, each time a bottle is filled, there is a measurement on a row in the data for each “X” that creates the “Y.”
If we take a sample of 100 bottles, we should have 100 rows of data. This data helps us to analyze the process using statistical testing methods like ANOVA, t-tests, correlation & regression, chi-squared, etc.
I hope this article will increase you or your organizations chance of selecting the right projects to create a sustainable Lean Six Sigma effort in your organization.