A control chart is a graphical representation of measurement data for an important product or process parameter. In manufacturing, the diameter of bearing balls can be used as an example. In service industries, measurements could be missing information on insurance claims forms. According to statistical sampling procedures, measurements are taken at different times. The control chart shows how these measurements change over time and is designed based on the normal distribution, or the classic bell curve. It's easy to see if actual measurements fall within the statistical limits of this distribution. The upper limit is called the "upper control limit" and the lower limit is called the "lower control limit." If the measurements on the chart exceed the upper control limit or fall below the lower control limit, it indicates that the process is out of control. This requires a thorough investigation to identify the problem and find components that have changed in a non-random manner. Was the steel rod used to make the bearing balls too hard? Too soft? Or was the cutting adjustment value set incorrectly on the steel rod cutter?
A Pareto chart is a simple chart tool used to count and display the number of various types of defects or problems over a certain period of time. Its functionality is represented by bars of different lengths on the chart. This concept is based on the research of 19th-century Italian economist Vilfredo Pareto, which suggests that 20% of all possible causes lead to about 80% of the problems, while the remaining 80% of the causes only result in 20% of the problems and defects.
To make improvement methods most effective, it’s essential to first address the few critical causes responsible for the majority of quality issues. A Pareto chart helps determine the small number of critical causes behind the majority of the problems. It can also be used to identify where in the production process certain defects are most likely to occur.
A fishbone diagram, also known as a cause-and-effect diagram or Ishikawa diagram (named after Kaoru Ishikawa, who first proposed this tool), resembles a fishbone. The problem or defect (i.e., the effect) is marked at the "fish head." Fishbones extend from the main line, listing possible causes of the production problem according to their likelihood of occurrence. The fishbone diagram helps explain how these causes interact with each other. It also illustrates how these possible causes appear sequentially over time. This aids in beginning to solve the problem.
A trend chart, sometimes called a run chart, is used to show the measurement results obtained over a specific time interval (e.g., a day, week, or month). The measured quantities are plotted on the vertical axis and time on the horizontal axis.
The trend chart acts like a constantly changing scoreboard. Its primary purpose is to identify significant time patterns in various types of problems, allowing for investigation into their causes. For instance, plotting the distribution of defective products by hour or day may reveal that problems only occur when using materials from a particular supplier, suggesting that the supplier's materials might be the issue. Alternatively, it may show that a specific machine consistently produces a certain type of defect, indicating that the problem might lie with that machine.
A histogram, also known as a bar chart, represents the number of products in the first category (corresponding to a series of mutually independent measurement values) using the length of bars. Categories are labeled and arranged either horizontally or vertically. A histogram can indicate which categories represent the majority of the measurements. It also shows the relative size of the first category. The histogram provides a factual distribution of the measurement results. The graph can show whether the distribution is normal, i.e., if its shape is approximately bell-shaped.
A scatter diagram provides a standard way to indicate how one variable relates to another. For example, to understand the relationship between the tensile strength of a metal wire and its diameter, typically the wire is stretched until it breaks, and the exact force required to break it is recorded. Plotting the results with diameter on the horizontal axis and force on the vertical axis allows you to see the relationship between tensile strength and wire diameter. This information is useful for product design.
A flowchart, sometimes called an input-output diagram, visually describes the detailed steps of a work process. Flowcharts are extremely helpful in accurately understanding how work is carried out and deciding how to improve the process. This method can be applied across an entire enterprise to track and diagram the company's operational methods visually. Flowcharts use some standard symbols to represent certain types of actions, such as using a diamond box for decision-making and a rectangle for specific activities. More important than these symbols is clearly describing the sequence of the work process. Flowcharts can also be used to design improved work processes by first drawing how the work should be done and then comparing it with the actual situation.