The Deming Wheel is more properly called the Shewhart Cycle and now commonly referred to as continuous improvement or, kaizen, for the Japanese word meaning improvement.
By any term, it is a process all of us need to be doing and far too few are. Occasionally when I ask managers if they have embraced a continuous improvement program, they respond that they have but it is not a formal plan, but most often the response is a simple no.
I suggest that unless you have formalized a continuous improvement program with established procedures you are, at very best, not realizing the full value of the tool.
While the concept of continuous improvement seems simple enough — and it is — there are basic principles that are not necessarily intuitive and often overlooked. Being aware of and following these basic principles can improve the effectiveness of the process.
Before we look at what to do and how to approach it, let’s look at what continuous improvement is — it might surprise you. It is process oriented thinking, not results oriented thinking. That may sound like an insignificant differentiation but read it again. The focus is process improvement and not reaching a defined result — remember this is a journey, not a destination. This concept of a journey is the crucial part of embracing the cultural belief that the process is never ending — for if you think you have arrived the process ultimately fails.
Seven Requisites of a Successful Continuous Improvement Program:
- Gain top management commitment and involvement.
- Establish an organization dedicated to promote continuous improvement.
- Appoint the best available personnel to manage the continuous improvement program.
- Train that person or persons in the process of continuous improvement.
- Conduct organizational training on exactly what continuous improvement is and how it will work.
- Establish (and formalize) a step-by-step process for continuous improvement projects.
- Gain top management commitment and involvement.
Lacking the commitment of top management will doom the success of the program — don’t waste your time or resources if that is not certain. Management must be committed to not only supporting the program — they must demonstrate involvement.
As in all of these types of programs we tend to think of them in terms of being applicable to the shop — but they do not stop at the office wall — this is an organizational program that applies equally across all disciplines of the organization.
The Process
Yes — to be sure — yet another of the endless and at times annoying catchy words and phrases! Add to the list of possible names for the process PDCA Cycle — or Plan, Do, Check and Act.
We need to lay some ground work before we go on. Walter Shewhart, who is credited with developing the PDCA method, was an advanced thinking statistician for the former Bell Labs in the 1930s. There are statistics involved in any continuous improvement program, and your lead people need to have a basic understanding of statistical analysis — if they don’t, you must provide the training necessary.
I have shared in the past my conditional support for statistical measures. They can take on a life of their own and the reporting and displaying of the results become more important than understanding and leveraging the value of the information contained in the data.
That said, it is both necessary and invaluable to evaluate your current status as a base line. It is against the base line that you will demonstrate improvement in your future state resulting from your continuous improvement program.
Let me reinforce again that this is a process oriented program and not results based. The future state statistical analysis is to verify change — not to verify some preconceived goal. Your goal is to improve — period. The future state as defined by the statistics will become your base line for the next improvement project in that area.
The Process Defined
Plan: Determine where your problems (and your opportunities) lie. Start small in one area. You will have plenty of low hanging fruit but resist using a shotgun approach. Be aware that multiple concurrent projects can distort your statistical analysis and contribute to invalid conclusions. Once you have identified a project, establish a small focused group, use cause and effect diagrams, flowcharting and other established methods to define the obstacles and plan the changes. Base line the current state performance while planning.
Do: This is the implementation phase. Often it is best to implement the change in one area as a test, when the solution is verified then apply it across a wider area. Often conflicts will surface at this stage; resolve them before proceeding.
Check: Verify improvement and check for introduced problems. Introduced problems resulting from the changes could be in the same area or even a different department.
Act: Once verified, fully implement the changes, continually verifying the results until certain the results are repeatable and sustainable. Standardize the new process changes.
Return to step one and plan the next improvement project.
The How
Here are the basic steps needed for continuous improvement:
This assumes that you have a continuous improvement manager identified (this need not be a full time job for small organizations—this could be as little as 30 minutes a day).
- Identify a project that offers a high certainty of visible results — you do not want to have a failure on your first attempt — there will be plenty of time for that in the future as your willingness to take risk grows.
- Determine current performance by base lining. Make sure your base line defines the targeted area of improvement.
- Obtain commitment of both management and the people in the target area. Define the improvement objective — but in terms of process change not “X” number more widgets. This might be an improvement in quality (scrap/rework reduction), an incremental but undefined reduction in cycle time or ergonomic improvements.
- Organize the team. This will include the program manager, the area personnel and supervisor and often members from other areas in the plant as “fresh eyes.” Occasionally, outside help may be beneficial.
- Identify the causes of the current performance limitations.
- Define potential solutions and test to determine if they will accomplish the improvement objective.
- Document an improvement plan that defines exactly how and by whom the changes will be implemented.
- Identify and overcome (where possible) unwarranted resistance to the change. There will always be resistance, particularly at the beginning of this journey. Use persuasion whenever possible but be aware that on occasion you may have to move the resistance out of the way.
- Implement the change.
- Put in place controls to maintain the changes, monitor and verify the results.
- Acknowledge and reward the success. Encouragement of the people, who have made the improvements, however small, is an important component. The success needs to be acknowledged and publicized across the organization with equal credit going to the entire team. Reward the team — but within reason. A pizza party and extended lunch for the team that had a major accomplishment might be far more appropriate in your organization than some flashy award that can cause ill feelings across the entire organization.
- Notice the PDCA diagram is a circle — begin again!
I cannot close without a comment on failure — it will happen. Accept that whenever you do something new or different, failure is a possibility. DO NOT go in search of the guilty. Understand why there was a failure, fix the cause, learn from it and move on. If you have followed all the steps the failures will surface during the test phase. Thus the results are only uncomfortable — and possibly embarrassing — not catastrophic.
Obviously, I cannot do justice to this process in the space limits of this article. There are many excellent books available — one I will recommend is Out of the Crisis by W. Edwards Deming, MIT 1989.
Don’t follow the path — make one — begin the trip and enjoy the journey!