IV.A.1 The seven classic quality tools
What are the Seven Classic Quality Tools?
Excerpt From The Certified Manager of Quality/Organizational Excellence Handbook
1) Flowchart - A flowchart is a map of the sequence of steps and decision points in a process. An example of a flowchart is shown in Figure 13.1. Flowcharting a process is a good starting point for a team, as it helps the group gain a common understanding of the process flow. Each team member typically has a perspective based on his/her own role, but may not have a full understanding of the entire process. A flowchart can also reveal missing, redundant, or erroneous steps.
2) Check Sheet - The check sheet (sometimes called a tally sheet) is used for gathering information to analyze. As with most of the tools, it facilitates the use of facts, rather than just opinions, in talking about and solving problems. Check sheets are used to gather data on frequency of occurrence (see Figure 13.3). Data from several check sheets can be organized into a Pareto chart for final analysis.
3) Cause-and-Effect Diagram - The cause-and-effect (C-E) diagram (also called an Ishikawa diagram after its developer Kaoru Ishikawa, or alternatively called a fishbone diagram) provides a way of collecting and organizing what may be a long list of potential causes that might contribute to a particular problem. In a C-E diagram, the problem (the effect) is stated in a box at the right side of the chart, and likely causes are listed under major categories that can lead to the effect. The four M’s (manpower, machinery, methods, and materials) are categories of problem sources typically used for manufacturing problems, although other headings such as environment and measurements are sometimes added. For service processes, the four P’s (people, policies, procedures, and plant) are often used. Figure 13.4 shows a C-E diagram of some of the reasons that outpatient clients may not be able to locate the X-ray department in a hospital.
4) Pareto Chart - Vilfredo Pareto, an 1800s Italian economist, noted that 80 percent of the wealth in Italy was held by 20 percent of the population. Juran later applied the Pareto principle for other applications, pointing out that 80 percent of the variation in a process is caused by roughly 20 percent of the variables; he labeled these variables the vital few, as opposed to the trivial many (changed later to useful many) that had much less overall impact.
Pareto analysis could also be used by a supervisor within each of the higher-cost work centers to pinpoint the particular machines that are the primary sources of trouble. Note that although the y-axis of a Pareto chart often represents the frequency of a problem, it is also important to do a similar analysis based on costs, since not all problems have equal financial impact.
5) Control Charts - Control charts are a refinement of the original run chart, which does not include control limits. A control chart serves two vital purposes as a data-gathering tool: (1) it shows when a process is being influenced by special causes, creating an out-of-control condition, and (2) it indicates how a process behaves over time.
Control charts should be examined for nonrandom patterns of data points; such an analysis of patterns can indicate what source of variation is most likely creating the condition (for example, which of the causes in the C-E diagram is most likely influencing the process). Patterns could reflect wildly fluctuating values, sudden process jumps or shifts, a gradual trend, or increased variation. Each of the causes in the C-E diagram is likely to create only a particular type of pattern, making the control chart a valuable diagnostic tool. The pattern might also reveal improvement in a process, indicated by decreased variation. Investigating this change to find the cause may provide deeper learning.
6) Histograms - While control charts allow seeing how a process performs over time, a histogram provides a graphical picture of the frequency distribution of the data. The histogram allows detection of distributions that do not demonstrate a typical bell-shaped curve, and shows how the process spread and central tendency relate to process specifications.
For a normal (bell-shaped) distribution, the most frequently appearing value (mode) is centered, with data appearing equally on either side. If data extend beyond the specification limits, the process or product is out of tolerance. Additionally, the histogram may also point out that there are actually two different distributions at work in the process (for example, differing impact of the same raw material purchased from two different suppliers).
7) Scatter Diagrams - A scatter diagram shows whether or not there is a correlation between two variables. Correlation does not necessarily mean a direct cause-and-effect relationship, however. If it appears that values for one of the variables can be predicted based on the value of another variable, then there is correlation.
If the slope of the plots is generally upward, it is said that a positive correlation exists between the variables. That is, as one increases, the other also increases. Other possible measures of correlation are negative, weak, or none. A negative correlation exists if the line slopes downward, meaning that as one variable increases in value, the second decreases. If no visible pattern appears to exist in the plotted points, the two variables are not correlated.
Quality Management BOK Reference
IV Quality Management Tools
IV.A Problem-Solving Tools
IV.A.1 The seven classic quality tools - Select, interpret, and evaluate output from these tools: Pareto charts, cause and effect diagrams, flowcharts, control charts, check sheets, scatter diagrams, and histograms.
Back to the Quality Management Tools CMC
Back to the Quality Management Body of Knowledge