Kaizen and PDCA: continuous improvement on the shop floor

Why it matters

Searching for a spanner at a station may take only a short time in each cycle, but a repeated search interrupts the flow and pulls the worker's attention away from the work being done. Kaizen is an approach of investigating such losses together with the people who do the work and building a better working arrangement. A small change can be valuable, but being small does not mean it may go unmeasured. Moving a tool is an activity. A reduction in search time under the same conditions is the result. If this distinction is not made, the number of completed adjustments rises while the production problem stays where it is. The subject of the improvement is defined as removing unnecessary work rather than making the worker faster.

The steps plan, do, check and act are used as the PDCA cycle; its international abbreviation, PDCA, describes the same sequence. In planning, the problem, the starting point, the target and the way to test it are written down. In doing, the change is tried within the defined limits. In checking, the result is compared with the starting point and with the target set beforehand. In acting, it is decided how the arrangement that works will be kept; if the trial fails, the explanation is reviewed. Changing the target retroactively according to the result is not checking. Why the target was chosen and under which conditions it applies must be clear before the trial.

When choosing a problem, the frequency, time and cost Pareto studies covered in M9 are used. The most frequent event does not always cause the longest loss or the largest expense. This lesson does not rebuild the ranking method; it keeps track of which loss measure the chosen problem rests on. A measure that is mandatory for safety or product conformity is not postponed merely because another loss is larger. Choosing a problem is also not choosing a solution. Finding the search loss important does not prove that a specific tool to be purchased will be effective. A separate, measurable expectation is needed for that tool.

A reduction in time alone is not a sufficient acceptance condition. The same product conformity, safe access and the requirements of the next job must also be preserved. If moving the spanner closer blocks a walkway, the trial is stopped. A drop seen on a day when easier products were run is not generalized to all products. The period in which the worker gets used to the new arrangement is recorded separately. A difference becomes meaningful when the starting point and the scope of the trial are equivalent. A good improvement record therefore carries not only the successful result but also the limits and the open questions. As the cycle produces new information, the subject of the next trial becomes clearer. The figure and the short quiz scenarios in this lesson also fall within the synthetic data of the examples. This dataset was generated synthetically for training; it does not represent a real business, employee or customer study.

Formula