TECHNOsoft

KNOWLEDGE SPACE

see more...

SEVEN TOOLS FOR QUALITY IMPROVEMENT

Written by Millenia A. SusantiThursday, 10 April 2008

Quite simply, quality is defined as giving the customer the right thing right in the first time. When quality is implemented, it yields increased market share and reduced scrap and rework. Thus many companies have adopted different methods to gain quality improvement
One of many methods to ease the quality improvement process was introduced by Kaoru Ishikawa. It was called Seven Tools of Quality. Ishikawa believed that 95% of a company’s problems can be solved using these seven tools. The tools are designed for simplicity so average person can have less complicated statistical analysis. The seven tools are:
1.Flow Charts
2.Cause and Effect Diagrams
3.Check Sheets
4.Histograms
5.Pareto Diagrams
6.Scatter Diagrams
7.Control Charts

Flow Charts
Flow charts describe sequence of activities graphically in accomplishing a task. It must reflect the actual process rather than what the process owner wants it to be. By generating flow charts, process owner can understand the process and working relationship between people and organization will be clarified. Furthermore, flow charts will show the duplicated effort and other non value added steps. So process owner can identify the target specific steps in order to make continuous improvement.

Cause and Effect Diagrams
Cause and effect diagrams are also called Ishikawa diagrams or fishbone diagrams. These diagrams show a relationship between the effect of the problem and the causes. Although cause and effect diagrams can be developed by individual, but it is better to be brainstormed by a team. To help in identifying the causes, people used to take 5 potential factors which are man, machine, material, method, and environment as the guidance.Cause and effect diagramsconsist of 2 sides. The right side lists the effect or the problem, while the left side lists the causes of the problem.

Check Sheets
Check sheets are the paper forms for collecting data in real time easily and concisely. According to the data on check sheets, frequency or pattern of the events, problems, defects, etc. can be observed. These data are then used as input data for other seven quality tools such as histograms and pareto diagrams. Furthermore, collected data on check sheets can be used as input to understand the real situation, analyze occurring problem, control the process, make the decision, and develop planning.

Histograms
Histograms are bar graphs that present the frequency distribution of data. The horizontal axis represents the class interval of data while the vertical axis represents the frequency of each class interval. Not only for displaying data, histogram can also be used as a tool for summarizing and analyzing data. The particular values of histogram are:
1.Histograms show the variation of data.
2.The shape of the histogram gives some clues for analyzing the process.
3.When there is a specification, histogram serves data outside specification. Based on these data,

process owner can determine the action of continuous improvement.
4.When the data is stratified based on the factors which are thought to cause variation, the cause

of variation become more easily detected.

Pareto Diagrams
Pareto diagrams are diagrams consist of bar graph and line graph that describe the proportion of each problem cause toward the overall cause. According to these quality tools, the major factor of the problem can be identified. Thus, it helps process owner in prioritizing the problem to be solved first. As Dr. J. M. Juran said, 80% of the problems are caused by 20% of the potential sources.
Not only to identify the major problem, pareto diagram can also be used to analyze the improvement after solving some problems. Process owner can evaluate the process before and after solving the problems.

Scatter Diagrams
Scatter diagrams describe the correlation between 2 variables, so it can be identified whether those 2 variables are related.
After developing cause and affect diagrams, sometimes people use scatter diagrams to determine objectively whether the cause and the effect are related.
By looking at a glance to the scatter diagrams, process owner can analyze the positive/negative correlation of 2 variables. When the trend line goes from the bottom left to the up right, it means those 2 variables have positive correlation. Otherwise, if the trend line goes from the up left to the bottom right, and then it means those 2 variables have negative correlation.
Not all 2 variables have particular relation. There are also 2 variables that have no correlation.
Besides analyzing the positive/negative correlation, process owner can also determine how well the correlation between 2 variables. It is demonstrated by the location of each point. The closer the points plotted to the line, the better the correlation.

Control Charts
Control charts were invented by Walter A. Shewhart in 1920s. These quality tools consist of line graphs completed with maximum limit and minimum limit which provide control area. The maximum limit often called upper control limit and the minimum limit often called lower control limit, while the central line represents an estimate of the process mean. Since the value is within control area, the process can be stated as a controllable process. The value outside control area indicates that the process is no longer stable because of some variation causes. Thus, this process needs proper corrective actions to eliminate the sources of variation.
Control charts also show the process change over time. If the change is good, its cause should be identified and possibly become the new process. Otherwise if the change is bad, its cause should be identified and then be eliminated.
Control charts include 4 types of chart:

1.xbarR ChartIt is used for controlling and analyzing a process using continuous values of

product quality within one subgroup.
2.x ChartIt is used to control and analyze data at a long interval or ineffective subgrouping.
3.pn Chart and p ChartThese charts are used when the quality of characteristic is represented

by the number of defective units or fraction defective.
4.c Chart and u ChartThese charts are used for controlling and analyzing a process by defects of

a product. A c chart is used for a product with constant size, while a u chart is used for a

product with varying size.

References
Kume, Hitoshi. 1992. Statistical Methods for Quality Improvement. 3A Corporation. TokyoOne

Adopted from http://www.midas-solution.com/