Quality Tools 


The Pareto Principle states that, for many events, roughly 80 % of the effects come from 20 % of the causes.The italian economist Vilfredo Pareto observed 1906 that 80 % of the land in italy was owned by 20 % of the people. (80 / 20 rule) Pareto developed the principle after he observed that 20 % of his pea pots in his garden contained 80 % of the peas.
The key to understanding the Pareto principle is the phenomenon that it deals with skewed distributions. The Pareto Principle is used to prioritize problems or to illustrate a phenomenon. The Pareto Chart separates the few vital from the many trivial. 


The Fishbone diagram, also known as Fishbone chart is one of the many problem solving tools Dr. Koaru Ishikawa (1915  1989) invented. The Fishbone diagram is used to explore and pinpoint the root causes of undesirabvle effects / problems. The effect (output) is the starting point (right side of the chart) for the creation of a cause and effect diagram, while the potential or real causes are the inputs that could lead to the problem. The Fishbone diagram technique is frequently used as a team brainstorming tool. For many fishbone diagrams the following clusters are frequently used: Methods, People, Machines, Materials, Environment, Measurement. 


Frank Gilbert (1868  1924) introduced in 1921 the "flow process chart". In his presentation to the American Society of Mechanical Engineering and he referred to the Flow Chart as "process charts first steps to find the one best way".
Flow charts are used to
 Identify problem areas in a process
 Identify relationships of various steps in a process
 Design of an ideal process
 Identify "bottlenecks"
 Planning of a project
 Documentation of a process (SOP)



Sir Francis Galton (1822  1911) created the statistical concept of correlation. Karl Pearson (1857  1936) a student and biographer of Francis Galton established the discipline of mathematical statistics and he developed the rigorous treatment of the mathemeatics of the Pearson Product Momentum Correlation (PPMC). A correlation is a single number that describes the degree of relationship of two or more variables. The correlation coefficient measures the degree of the linear relationship of two variables x / y. The correlation coefficient can have any value between 1.0 and +1.0. The sign of the correlation coefiicient (+,) defines the direction of the relationship, either positive or negative. 


A histogram is a graph that displays the frequency of occurences of all possible outcomes of repeatable events over a period of time. It displays the measurement of a variable and can unveil the pattern of an underlaying phenomenon. Frequently used applications for histograms include:
 Determination of averages and deviations of observations from the average from a data set
 Identification of the range of variations of processes
 Highlighting process outcomes which are outside the specifications (out layers)
 Setting standards for process improvement



Run charts were originally developed from control charts, which were first developed by Walter Shewhart. Run charts, also known as line graphs, display the process performance over time. In other words it displays the outcomes of a process over time. The xaxis represents the time period over which results are observed and the value of the output is plottet on the y axis. The run chart is very useful to identify trends, highlighting seasonality and cyclical patterns. Run charts are often used to monitor changes over time. Applications for run charts include:
 Tracking of current perfromance
 Setting goals to improve average performance
 Understanding variation in process performance
 To communicate how a process performed during a specific time period.
 Forecasting (Economy, stock market, sales of products)



Dr. Walter A. Shewhart while working for Western Electric laid the foundation for statistical process control and the invention of the control chart. The control chart is a graphical technique to determine whether a process is or is not in the state of statistical control. How to construct and interpret control charts:
 Collect data over time until you have suffiecient data points to perform statistical analysis.
 Determine mean, upper and lower limits and plot 3 lines (mean, upper and lower control limits)
 Determine if the observed process is in control.
 If a process is not in control, investigate and eliminate causes of the undesirable variation.
 Use problem solving methodology to make improvements.
 Continue collecting and plotting data, and test for out of control conditions.





