Details of a PDSA Cycle
by Karen Roberts
Posted on 29 Nov 2018
One of the simplest and most commonly models used for planning and problem solving, is the PDSA Cycle or the Deming Cycle. This well-known model for quality is also referred to as the Shewhart Cycle. Walter A. Shewhart discussed the concept in his book Statistical Method from the Viewpoint of Quality Control. It can also be referred to as The PDSA Cycle, where the "S" stands for "study." The "S" was introduced by Deming to emphasize the
importance of learning in process improvement.interchangeably. The "C" stands for check. Either check or study. Makes no difference; just make sure you look at the data you've collected in an objective way.
Plan:
The planning phase of the improvement cycle should contain;
• The specific objective and the questions to be answered
• The prediction of the results of the cycle
• The plan for carrying out the cycle
• The plan for the data collection
Once the methods have been identified, the team should decide who has the responsibility for collecting and studying the data. The team should maintain controls over the collection of data, especially, if the responsibilities are outside of the team. That is, they should be aware of such things as the training that is needed, any problems that occur, and whether the method of measurement has been studied. The team must also decide when, where, and by whom the plan will be implemented.
Consideration should be given to whether the plan can be carried out on a small scale so the impact can be studied with minimal disruption. For example, a change might be made and studied on one assembly line first before being implemented throughout the entire plant.
The team should review the plan to assure that it is consistent with their charter. Special attention should be given to the boundaries for the activities contained in the charter.
Do:
This phase begins when the plan for the cycle is put into place. In any collection of data, there are many opportunities for error and many opportunities for special causes to occur. Plotting of the data as they are obtained (i.e.. in order of time) is vital to recognize problems.
The team should be informed about and note things that occurred during the data collection that were not part of their plan. For example, the operators might not have been adequately trained or the procedures used not standardized. This would also include unusual or unexpected situations that occurred, especially things that went wrong. For example, a power failure may have occurred or an observation may have been missed. When studying the data, the team should consider these occurrences. If it is not possible to take action on such items during the present cycle, they should be considered in future cycles.
Study:
In this phase, the data are analyzed to help answer the questions which were posed during the planning phase. Since most data from well planned studies can be analyzed using simple graphical methods, many teams will be able to analyze their own data. There will be times, however, when help from a statistician or someone else with special expertise is required. The information gained from the analysis of the data is combined with the "current knowledge".
It is in the study phase of the improvement cycle that learning takes place. To assist the team in assessing what was learned, the results of the analysis of the data should be compared with the predictions made during the planning phase. The predictions were based on the team's current knowledge. If the analysis of the data confirm the predictions, the degree of belief in the current knowledge is increased.
Act:
In this phase of the improvement cycle, the team decides whether changes are to be made to the process or product. If there are any changes, the team should plan and implement them.
What was learned in a cycle may not result in sufficient knowledge to make a change. For example, the cycle may have identified an important quality-characteristic for use in developing a cause and effect diagram, confirmed existing knowledge, or identified new technology that could impact on the design of a new product. In some cases, a cause may be identified which does impact a quality-characteristic but the cause and effect system may not be sufficiently understood to take action.
Following a process, regardless of whether it's a PDSA Cycle or some other model connected with quality, the main thing is that you're reading this article and thinking about how to improve your organization. Good on ya!