We have all been there from one time to another in a large meeting room, hearing a problem from everyone’s perspective. Needless to say, when you are in the mix, it is good to start analyzing. To gain an understanding of what is going on with the process, lets start with the basics: who, what, when and where. Answering these questions in detail is the data stratification model, and it clarifies the collected information to reveal the root cause of the problem.
Six sigma teams use the data stratification model whenever data is collected. Because it quickly clarifies who is associated with the problem, what type of problem is occurring, when the problem is happening, and where the problem is occurring. The purpose of the model is to help the team clarify the problem and its impact. The information is also very useful when analyzing the problem to find the root cause.
The process of data stratification is to break it down into layers (AKA strata). The questions who, what, when, and where represent the layers.
- The who layer – The question who is intended to define who is associated with the process problem. This information could be further subdivided by:
- Vendor
- Department
- Individual
- Customer Type
- The what layer – The question what defines the type of problem which is occurring. This information can be further categorized by:
- defect category
- type of complaint
- reason for a complaint.
- The when layer – The question when is pretty much self explanatory it clarifies when. This information can be further clarified by:
- time of day
- day of week
- month of year
- fiscal quarter.
- Shift
- The where layer – The where question clarifies where a problem occurs. This can be clarified by:
- facility
- region
- location on product
- location in service.
By asking the key questions from the data stratification model, the who, what, when, and where of a process problem are further brought into focus.
While process data may not be available to answer all the model's questions, the team should nevertheless consider using the model whenever it collects data. Answering only some of the questions can still greatly benefit an investigation. Here is an example, for quantifying the problem surrounding OR charges.
Data Stratification – Late OR Charges |
||
Factors |
Definitions |
Examples |
Who: |
The question "Who?" is intended to define the person or organizational entity that is associated with the process problem. |
|
What: |
The question "What?" is intended to define the type of problem that is occurring in the process. |
|
When: |
The question "When?" is intended to define the time frame in which the problem is occurring in the process. |
|
Where: |
The question "Where?" is intended to define the location in which the problem is occurring in the process. |
|
Subscribe and Share!
Did you enjoy this article? Your feedback is very important! I'd like to invite you to keep up to date with the latest posts from Anticlue. We offer several venues. If you have some questions, help can be found here.- Become a Facebook Fan
- Subscribe to Anticlue
- Follow me on Twitter
- Add to Technorati Favorites
- Digg this post



