Understanding Variation
When analyzing time ordered data, you need to look at the variation, how does the data change from point to point. Specific patterns in the variation can provide clues as to the source and nature of the process problems.

What is variation?
No two anythings are exactly alike.

• How a process is done will vary from day to day
• Measurements or counts collected on process output will vary over time
• Quantifying the amount of variation in a process is a critical step towards improvement
• Understanding the causes of variation helps us decide what actions are most likely to lead to lasting improvements 

Special Cause and Common Cause
Special Cause variation means something different happened at a certain time or place. Common cause variation is always present to some degree in the process. The goal is to minimize the variation. It's very important to distinguish between special and common cause variation as each requires a different strategy.

Special Cause Strategy
The strategy for Special Cause signals is:

• Get timely data
• Take immediate action to remedy any damage
• Immediate search for a cause. Find out what was different on that occasion. Isolate the deepest cause you can affect.
• Develop a longer term remedy that will prevent that special cause from reoccuring. Alternatively, if results are good, retain that lesson for the future.

Common Cause Strategy
The strategy for Common Cause signals is to improve a stable process. The process is stable but it does not meet customer needs.

Improving a Stable Process
Common causes of variation can hardly ever be reduced by attempts to explain the difference between individual points if the process is in statistical control. All the data is relevant. A process in statistical control usually requires fundamental changes for improvement. Using the DMAIC* method can help you achieve fundamental changes in the process.

Effective improvement relies on being able to distinguish common cause variation from special cause variation.

• If you treat special cause variation like common causes then you lose an oppurtunity to track down and eliminate something specific that is increasing variation in your process.
• If you treat common cause variation like special causees then you will most likely end up increasing your variation.
• Taking the wrong action not only doesn't improve the situation but usually makes it worse. 

Useful links:

SPC Check

SPC Report

Pareto Report

 

*DMAIC Method
The DMAIC method was created with the development of continual improvement, quality, safety or environmental level improvement in mind. In point of fact, it is an improved PDCA* cycle. The quality is the area where the cycle achieved main progress and implementation.

DMAIC Description
The method defines 5 phases for successful implementation of change or a project control for improvement.

• D – Define – In the first phase the targets are specified, information is collected, expected final status is being described, a team is established. The process to be improved is described. As a part of the process description (start & end of the process, inputs & outputs, etc.) is a scope of process itself. The plan including specific actions for solving the problem is clearly defined.
• M – Measure – Consecutive steps which shall be achieved and which results in the fulfillment of the specified goals are important during the improvement cycle. Evidence of the achieved goals is possible only when defined measurements and quantifiable indicators are specified at the beginning. Only in that way is it possible to distinguish speculation from reality.
• A – Analyze – Detected information will be analyzed in detail and real potential for improvement shall be identified. A root cause analysis of problems, noncompliances, dissatisfaction, etc is a baseline. At the same time the question, if the real original problem is being solved, is fully probed.
• I – Improve – Is a baseline for real root cause elimination. New parameters of a process are set up, a process is being optimized. Everything is done to improve customer's (either internal or external) satisfaction. Customer's costs improvement or contribution shall be a part of an improvement. Each solution can be tested in a pilot test.
• C – Control – If the problem is successfully solved or an improvement has been achieved, the last and end step shall be done – to implement every changes, to standardize them into processes or system. We shall verify if the changes are properly asserted and they are a part of every day's activities. Determination of the time period for achieved results or profits from improvement monitoring is suitable.

DMAIC Applicability
The method is possible to use for:

• Whatever problem solving
• New change implementation
• Achievement of better and pre-set results
• Customer satisfaction

DMAIC phases are often repeatable, by which the spiral of continual improvement and better results achievement can be spun again and again.

 

*PDCA Cycle
PDCA was originally prepared for effective solution and improvement of activities, processes and systems in a production process. It can be used as a simple methodology for changes implementation. Quality is the area where the cycle achieved the main development and practical usage. PDCA should be a part of knowledge of every individual who works in the areas of quality systems, ecological systems and safety assurance.

PDCA Description
The PDCA methodology consists of the following steps:

• P – Plan – The cycle starts with collection of information and a description of the problem which shall be solved. It is a baseline for a plan preparation. A plan should contain activities specified as tasks what shall be done in order to solve the problem.
• D – Do – Once the plan is prepared the next step is to implement described activities.
• C – Check – Follows achieved results monitoring and their comparison with plan. It is an inspection if an original problem was really solved.
• A – Act – In case result differs from expectation and the problem is not solved, then look for the problem's cause. You should focus a new plan on cause's elimination in such a case. If a problem is successfully dispatched than the last and final step should be to implement and standardize all needed changes into processes or the system. In addition, it should be probed if changes are correctly exercised in every day activities.

PDCA Usage
It is possible to use the cycle for solution on whatever problem solving or changes implementation. The four basic steps of PDCA can be continually repeated. PDCA repetition spins the spiral of continual improvement.