A few quality tools often used in Six Sigma:
| Flow charts and process maps | are often the first step to understanding our processes. |
| Checksheets | are quick, easy, flexible tool for collecting data. |
| Histograms | are often the first representation of our data beyond the numbers. |
| Pareto charts | are used to prioritize problems or root causes. |
| Cause and effect diagrams | are used to brainstorm possible root causes of a problem. |
| Scatter diagrams | are the first step to understanding the relationship between two variables. |
Pareto Charts
Pareto charts are useful for prioritizing your efforts, root causes and problems. From this idea, came the famous 80/20 rule. This concept was adapted for quality by Joseph Juran: 80% of defects come from 20% of the causes. He said a Pareto chart helps us separate the vital few from trivial many.
Pareto diagrams use discrete data that is counts or frequencies. The data is arranged from the most frequent to the least frequent, whose scale is on the left; the line represents a cumulative percent and its scale is the right. Visually it is easy to see where our efforts need to go for maximum effect.
Cause and Effect Diagrams
It is a formal brainstorming tool for identifying possible root causes. The purpose is to gather information and ideas from as many people as possible to explore all possible causes of a problem. The diagram looks like the skeleton of a fish, and our analysis is intended to identify major factors (big bones), that contribute to the problem. And then further break these down into more elementary causal factors (little bones), that cause these factors to create or prevent problems.
The number of major bones is not critical but it is usually 4 to 6. You can make up your own schemes whatever fits your department or your application best. The important thing is to consider lots of different types of causes. A few examples:
| 6 M’s | Measurement, Material, Methods, Management, Machinery, Manpower |
| 4 P’s | Policies, Procedures, People, Plant |
| 4 S’s | Surroundings, Suppliers, Systems, Skills |
This is a brainstorming tool, it generates a large quantity of ideas about possible causes. It does not identify root cause. The team needs to use other means to identify probable causes.
Flow Charts and Process Mapping
A process is an activity usually a number of steps that receive an input and convert that input to an output. A process map is often the starting point for improvement efforts. There are 5 major types:
| Basic | outline major steps |
| Detailed | used to improve a process |
| Top down | major steps, next level of sub steps |
| Deployment | detailed process with people |
| Opportunities | highlights opportunities for improvement |
The purpose of creating a process map is to gain an understanding of a process. Process maps are useful for documentation, training and certification. For the high level business processes that cross functional boundaries, there often is no one person who knows the whole process. Each business process often contain multiple work processes which can be mapped separately and will provide much more detail. One of the biggest challenges for a process mapping team is determining the detail and using it consistently.
Checksheets
Checksheets are simple and flexible tools for collecting data. There is not single form used in all applications. There are 3 main types of check sheets:
- Recording checksheets
- Checklists
- Measles charts
Before you collect data, you should make a plan – what data you want, how; who is going to collect it; what you will do with it.
Scatter diagrams
This is used to investigate the possible relationship of 2 variables that relate to the same event. It gives you a visual representation of relationships that can be confirmed through correlation regression. analysis. Each point on the plot represents the pair of value for x (dependent variable, you try to fix) and y (independent variable, might be the root cause).
We can tell how strong the relationship is, but we don’t know if x causes y, or y causes x, or something else causes change in both of x and y. If we want to know more, we need to do more analysis.
- Correlation, whose value is between -1 (strong negative) and 1 (strong positive), tells you the direction and strength of the relationship. Correlation of zero means no significant relationship.
- Regression tells you the nature of the relationship between the dependent variable and independent variable. It also allows you to create and test a predictive equation.
Histograms
Histogram is a graphic representation of what frequency distribution looks like. It is typically a bar graphs: vertical axis represents the frequency; horizontal axis represents the various events and things that happen. This requires discrete data; or continuous data converted to discrete data. You need to determine how many groups (usually square root of number of data points).
The first thing that we look for is whether or not the data is normally distributed. Where is approximately the center of data? What is the spread? Any outliers? These questions can help you decide what type of analysis is to do next.
Six Sigma Metrics
These are 4 common Six Sigma measures:
| defects per unit | DPU = number of defects / number units produced |
| defects per million opportunities | DPMO = number of defects / opportunities * 1000000 or = DPU / opportunities per unit * 1000000 |
| cycle time | Time to complete one cycle of an operation. From the beginning to the end of a process including wait time and other delays. Don’t be confused with: 1. Process Time is the sum of all processing times for all of the steps in a process. It does not include waiting time and other delays. It is not unusual for cycle tome to be 10-20 times is much as process time. 2. Lead Time is the time between when a customer places an order and when that order is delivered. It includes cycle time and other delays. |
| roll throughput yield | the probability that a process with more than 1 step will produce an error / defect free unit. First Pass Yield = 1 – DPU Roll Throughput Yield = multiplication of all FPY |
Quality Cost
Quality cost represent the difference between the actual cost of a product or service, and what the reduced cost would be, if there is no possibility of the substantial failure or defects. Traditionally quality cost is divided as:
Cost of poor quality
| external failure cost | the most expensive. They occur after the product has reached the customer or incorrect service has been delivered. Clearly customer will be dissatisfied. |
| internal failure cost | better at least bad product or service did not reach the customer. |
Cost of good quality
| Appraisal cost | in our effort to detect errors or defects. |
| Prevention cost | are often one time instead of ongoing. For example use lean tools to prevent errors and defects. |
My Certificate
For more on Six Sigma: Quality Tools and Metrics, please refer to the wonderful course here https://www.coursera.org/learn/six-sigma-principles
Related Quick Recap
I am Kesler Zhu, thank you for visiting my website. Check out more course reviews at https://KZHU.ai
Don't forget to sign up newsletter, don't miss any chance to learn.
Or share what you've learned with friends!
Tweet
