Six Sigma Manufacturing: How to measure and chart
73
How to measure and chart links
- How to measure and chart in six sigma
This article further discusses the topic of measuring nd charting in six sigma. - Definition
This web site provides a brief definition of a process map and flowcharts. You are able to link to other pages for more information pertaining to the Six Sigma process and how it works in your company. - Box-Whisker Plots
This web site provides information about Box-Whisker Plots. It provides all the steps you need to plot your own Box-Whisker plot and provides an actual diagram to understand Box-Whisker plots.
Video
There are several different methods to measure and chart your data. Some of the common methods include:
- Flowcharts
- Process Maps
- SIPOC
- Box-Whisker Plots
- Cause and Effect Diagrams
- Check Sheets
- Interrelationship Diagram
- Stem and Leaf Plots
Flowcharts and Process Maps
Flowchart and process maps are a graphical representation of a process procedure that shows all the actions involved and the conclusion they reach. Flowcharts and process maps are used to show how a predefined input will affect a predefined output. Your process map should allow people who are unfamiliar with the process to understand the interaction of causes during the work-flow. A process map should also contain other information that pertains to your Six Sigma project such as input and output variables, time, cost, etc.
SIPOC Diagram
The SIPOC diagram is another tool used in the Six Sigma method. The SIPOC diagram is used to identify all the relevant process improvement projects before actual work starts. The SIPOC diagram helps to define a complex project that may be hard to understand. If you are implementing Six Sigma, the SIPOC diagram will be utilized during the Measure phase of the DMAIC methodology.
SIPOC stands for:
- S - Suppliers in your process.
- I - Inputs to the process.
- P - Process your team is improving.
- O - Outputs of the process.
- C - Customers that receive the process outputs.
Creating an SIPOC diagram is easy; here are the steps to create a SIPIC diagram:
First, create an area that will allow additional comments or additions to be added. Many companies use transparencies or templates. You may even use post-it notes to add to the diagram. Second, decide what your process is and map it into 4 or 5 high level steps. Third, identify the outputs of the process. Fourth, identify the customers who will receive the outputs of the process. Fifth, identify the inputs that are required in order to make the process function properly. Sixth, identify the suppliers of the inputs that are required by the process. Seventh, identify the customer requirements. Eighth, discuss the project with your stakeholders, sponsors, or Six Sigma Champion.
SIPOC diagrams are particularly useful when you have a complex project that is not well scoped. The SIPOC diagram is an excellent way to find all the relevant elements in your process.
How to measure and chart related links
- Six Sigma Methodology
This web site talks about the different processes in the Six Sigma methodology. It provides visual examples of flow charts, process maps, and SIPOC diagrams. You can also sign up for Six Sigma training courses. - Cause and Effect Diagram
This web site provides a great example of a Cause and Effect diagram. It walks you through all the different steps required to set up a proper Cause and Effect diagram and how to read it when you are done.
Video
Box-Whisker Plots
Box-Whisker plots are useful in interpreting and distributing data. Box-Whisker plots are based on medians. You will take all the collected data and rewrite the data in order form from smallest to largest. Next, you will find the median of all the numbers. Once you have the median, you will find the lower and upper median. The lower median is the middle of the lower numbers and the upper median is the middle of the upper numbers.
With your medians, you can construct the Box-Whisker plot. You will draw an ordinary number line that extends far enough in both directions to include all the numbers from your data collection. Begin by finding your main median and drawing a line above the number line, do the same for your lower and upper medians. From here you will draw a box to connect the upper and lower medians. The whiskers will extend from the lower median to the lowest number in your data set and also from the upper median to the largest number in your data set. Box-Whisker plots are used to give you a visual representation of range of the data. The Box-Whisker plot breaks your data into 4 equal parts, thus making it easier to interpret your data.
Cause and Effect Diagrams
The cause and effect diagrams and used to help you identify a problem and think through the problem. A cause and effect diagram is often called a fishbone diagram because it looks like the skeleton of a fish when it is completed.
Use the following steps to create a cause and effect diagram:
1. Identify the problem. Write down the problem you are facing in exact detail. If you are able, identify who is involved, what the problem is, when it occurred, and where it occurred. The problem needs to be written down in a "problem box" on the left hand side of a large sheet of paper. Now, draw a line across the paper horizontally from the box. This line will be the space where you develop ideas.
2. Work out the major factors involved. From the "spine" you will draw different lines for each factor that is contributing to the problem. Some factors can include people, systems, materials, equipment, external forces, etc. If you are able, draw as many factors that you can think of. If you are solving this problem with a group, this is a great time to brainstorm.
3. Identify all possible causes. For each of the factors, you need to brainstorm all the potential causes for the problem. These will be identified with smaller lines the come off the "bones" of the fish. If you have a larger problem that is complex, you can break this down into sub-categories.
4. Analyze your diagram. Your diagram should include all the possible causes of the problem. This stage will allow you to investigate what is causing the problems and set up further investigation. The investigations are simply designed to test whether your assessments are correct. Cause and Effect diagrams are a great way to provide a structured way to help you think through all the causes of a problem and carry a thorough analysis of the situation at hand.
Check Sheets
A Check Sheet is used to collect data and organize it in a way that can be easily understood, used, and analyzed. The Check Sheet will become a graphical representation of the data you have collected and can be done with pen and paper or inputted into a computer program.
Here are the main functions of a Check Sheet:
- To check the production process
- Defective items check - identify where the defects exist
- Defective location check - identify where the common defects are located
- Check-up confirmation checks - the last phase of inspecting the finished product
To set up a check sheet you will need a blank piece of paper and divide it into 2 columns. In the left column of the check sheet you will write down all the common mistakes and errors. Create a timeline across the top of the paper (hours, days, weeks, months, years) and place a check mark in the grid each time you notice an error or mistake. Check Sheets are a quick and easy way to spot patterns in the causes of problems.
Video
Links: How to measure and chart
- Stem and Leaf Plots
This web site provides information on stem and leaf plots. It provides tips on how to create a stem and leaf plot and also includes a visual description of a stem and leaf plot. - Software for Flow Charts
This web site provides software for creating a flow chart. You can use this software to identify potential problems in your manufacturing organization and decide on how to solve them.
Interrelationship Diagram
Interrelationship Diagrams are used to identify, analyze, and classify the cause and effect relationships that exist among key issues so the outcomes can become the focal point of an effective issue. Interrelationship Diagrams are useful to brainstorm an issue and identify the cause and effect of problems.
To create an interrelationship diagram, you will create one problem statement then use input from other tools from one of the above diagrams and create a goal for the diagram. Interrelationship Diagrams usually require a team of 4-6 people to brainstorm the problems and create the statement. After the statement is agreed upon, you will create a flipchart or create a circular pattern of notes and leave space for drawing arrows. Include a number or letter next to each idea for writing in at a later time.
You want to draw a cause/effect relationship between different ideas and draw an arrow from each one. You can choose any idea as the starting point, if you are using numbers or letters, work through them in order. The outgoing arrow from an idea will indicate that it is the stronger influence/cause.
You will ask each of the following questions for each point:
- Do you have a cause/influence relationship?
- If you have a cause/influence relationship, which direction is stronger?
- Create boxes and go through each box and ask the same questions.
- Draw one way arrows from each box in the direction of the stronger cause or influence.
Once you have completed this, tally the number of outgoing arrows and select key items for further discussion and planning. Find the item that has the highest number of outgoing arrows and highest incoming arrows, the item with the largest amount of outgoing numbers need to be addressed first. Next, tackle the item with the highest number of incoming arrows. The incoming arrows can be used for future planning.
Stem and Leaf Plots
Stem and Leaf Plots are another method to show the frequency of error with certain classes of values. To create the stem and leaf plot you will make a left hand column which contains your data then create "leaves" which are lists in the right-hand column. The horizontal leaves in the stem and leaf plot will correspond to the vertical bars and have lengths that equal the numbers in the frequency table.
It is generally easier to select numbers that are rounded to the nearest hundred like 100, 200, 300, etc. Your data will then be plotted in between each number.






