The House of High Quality Articles for Everyone in the World

Feb 12, 2010

seven basic quality tools

Once the basic problem-solving or quality improvement process is understood, the addition of quality tools can make the process proceed more quickly and systematically. Seven simple tools can be used by any professional to ease the quality improvement process: flowcharts, check sheets, Pareto diagrams, cause and effect diagrams, histograms, scatter diagrams, and control charts. The key to successful problem resolution is the ability to identify the problem, use the appropriate tools based on the nature of the problem, and communicate the solution quickly to others.
Flowcharts describe a process in as much detail as possible by graphically displaying the steps in proper sequence. A good flowchart should show all process steps under analysis by the quality improvement team, identify critical process points for control, suggest areas for further improvement, and help explain and solve a problem.
Check sheets help organize data by category. They show how many times each particular value occurs, and their information is increasingly helpful as more data are collected. More than 50 observations should be available to be charted for this tool to be really useful. Check sheets minimize clerical work since the operator merely adds a mark to the tally on the prepared sheet rather than writing out a figure. By showing the frequency of a particular defect and how often it occurs in a specific location, check sheets help operator's spot problems.
Most quality problems result from a small number of causes. Quality experts often refer to the principle as the 80-20 rule; that is, 80% of problems are caused by 20% of the potential sources. A Pareto diagram puts data in a hierarchical order which allows the most significant problems to be corrected first. The Pareto analysis technique is used primarily to identify and evaluate non-conformities, although it can summarize all types of data. It is perhaps the diagram most often used in management presentations.
The cause and effect diagram is sometimes called an Ishikawa diagram after its inventor. It is also known as a fish bone diagram because of its shape. A cause and effect diagram describes a relationship between variables. The undesirable outcome is shown as effect, and related causes are shown as leading to, or potentially leading to, the said effect. This popular tool has one severe limitation, however, in that users can overlook important, complex interactions between causes. Thus, if a problem is caused by a combination of factors, it is difficult to use this tool to depict and solve it.
The histogram plots data in a frequency distribution table. What distinguishes the histogram from a check sheet is that its data are grouped into rows so that the identity of individual values is lost. Commonly used to present quality improvement data, histograms work best with small amounts of data that vary considerably. When used in process capability studies, histograms can display specification limits to show what portion of the data does not meet the specifications. After the raw data are collected, they are grouped in value and frequency and plotted in a graphical form. A histograms shape shows the nature of the distribution of the data, as well as central tendency (average) and variability. Specification limits can be used to display the capability of the process.
A scatter diagram shows how two variables are related and is thus used to test for cause and effect relationships. It cannot prove that one variable causes the change in the other, only that a relationship exists and how strong it is. In a scatter diagram, the horizontal (x) axis represents the measurement values of one variable, and the vertical (y) axis represents the measurements of the second variable.
A control chart displays statistically determined upper and lower limits drawn on either side of a process average. This chart shows if the collected data are within upper and lower limits previously determined through statistical calculations of raw data from earlier trials. The construction of a control chart is based on statistical principles and statistical distributions, particularly the normal distribution. When used in conjunction with a manufacturing process, such charts can indicate trends and signal when a process is out of control. The center line of a control chart represents an estimate of the process mean; the upper and lower critical limits are also indicated. The process results are monitored over time and should remain within the control limits; if they do not, an investigation is conducted for the causes and corrective action taken.