CONTROL CHARTS - PRELIMINARY ISSUES


CONTROL CHARTS ”PRELIMINARY ISSUES

Process control is the ultimate of a control chart. That means that the person who has ownership of the process has to recognize certain roles and responsibilities. Some of these roles and responsibilities are:

  • Ensuring that the process is designed to meet stakeholder requirements.

  • Dealing with barriers to getting process designs agreed upon and consistently implemented.

  • Ensuring that performance measurement is appropriate for the process.

  • Using measurement information to drive continual improvement in the process.

  • Communicating information clearly and constantly.

To be sure, these five points are not an exhaustive list, but they do represent some key issues for every process owner. These issues are as follows :

  • Process owners must change from silo thinking and reorient their attitude and thinking to encompass the process activities.

  • They must learn skills to influence, persuade, and negotiate the necessary outcomes of a data-driven process, rather than expect behaviors due to titles, strict past experience, and positions .

  • They must sponsor, support, and sustain the work of process improvement.

The last point is of particular interest because without the sponsorship, support, and sustaining efforts, improvements will fade away, and the status quo will return stronger than ever. So the question then becomes, how do we get the process owner "charged up" for this improvement?

The process owner must recognize, above all, that there are four main ingredients that will drive the process and its output to control and improvement. On recognition of these ingredients, the process owner will find himself or herself right in the middle of the improvement cycle. That process owner at that moment becomes the conductor, with a vision for the goal and articulation of customer satisfaction. These four ingredients are:

  1. The process includes ALL the people, equipment, materials, methods , measurement, environment. They all work together to produce the "expected" output. The process must have a quality output. To get all this, it must be efficient. The process needs a good design; it needs to be built correctly. It also needs to be operated correctly. The process control system works only when it aids the process.

  2. Information about the process is needed. Much can be learned by studying the process output. This is the product itself. Also, any intermediate output may be studied. The intermediate output describes the operating state, such as temperature or even cycle time. Get as much information as possible. Explain it. Use facts to decide what needs to be fixed. Is it the whole process? Or is it the output? Is it the intermediate output? Take timely actions; take good actions. If this is not done, information-gathering efforts have been wasted .

  3. Action on the process is future oriented. It is taken to prevent the making of out-of-specification products. This might mean a change in operation, a change in operator training, or maybe a change to incoming materials. Perhaps the most basic parts of the whole process need change. Sometimes such things as shop temperature or humidity can cause the problem. Actions should be watched. Check to see whether the actions are solving the problem. If they are not, further study must be made, rather than going to a different project. Maybe even more action should be taken. A process owner must want never-ending improvement.

  4. Action taken on the output is past oriented. It involves finding out-of-specification products and improving the reworking of defective units. This will go on until the process is fixed or until specifications are changed on the product.

An old saying reminds us that "a stitch in time saves nine." That is what this is all about. It is cheaper to prevent problems than to deal with a defective output. The major tool used in this effort is the process control chart .

We begin the process then by asking ”or at least being aware of ”the following:

  1. What characteristics are to be studied? [operational definition required]

  2. What testing or evaluation devices will be needed?

  3. Which chart will best accomplish the objective?

  4. What size sample should be drawn?

    • Subject to common causes only

    • Sensitivity to detection of special causes

    • Special reference distribution considerations

    • Cost: data acquisition

  5. How frequently should a sample be drawn?

  6. In what manner should the sample be drawn?

    • By whom?

    • In what way?

    • Where?

    • When?

  7. Implementation readiness?

    • 25 successive samples (minimum)

    • Statistical control established




Six Sigma and Beyond. Statistical Process Control (Vol. 4)
Six Sigma and Beyond: Statistical Process Control, Volume IV
ISBN: 1574443135
EAN: 2147483647
Year: 2003
Pages: 181
Authors: D.H. Stamatis

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net