When it comes to seeing how processes change over time, the ability to graph data can be invaluable, especially for continuous improvement practitioners. They can even help emphasize different attributes of the same data.
But with so many graphs out there, how do you know if you’re using the best Lean or Six Sigma chart for your particular data set?
This series of videos will answer that question for you.
What Do You Need to Know for Our Six Sigma Graphs Course?
Before we dive into the top Lean and Six Sigma graphs for business, consider this hypothetical scenario.
Your manager wants to gain a better idea of how a plastic molding process is performing and asks for your help. She’s particularly interested in how consistent the weights of these plastic parts are.
To help you get started, she provides you with weight data of 200 parts that were randomly sampled and weighed over the last few days of production. With so much raw data to sift through, it can be difficult to come to any accurate conclusion. In this case, you decide graphing the data is the best way to make sense of it.
And though there are a number of graphs to choose from, you settle on creating a standard histogram. From this graph, you notice that the data seems to form a bell shape, which means the sample mean is roughly four pounds.
From this one graph alone, you are in a much better position to explain how this particular process is performing to your manager.
What You’ll Learn with Graphs
In this video, you’ll learn about the most popular types of graphs used by Lean and Six Sigma practitioners, as well as how and when to use them.
Keep in mind — many graphs are designed to work with variable data while others are designed to work with attribute data. And some of these graphs work better with larger sample sizes, while others perform well with smaller sample sizes.
A Pareto chart is a type of bar graph that has been sorted in order of decreasing frequency, and often depicts which situations are more significant. This chart gets its name from Italian economist and socialist, Vilfredo Pareto, who observed in 1906 that approximately 80% of the land in Italy was owned by 20% of the population. Business consultants such as Joseph Juran later applied this Pareto Principle to quality control by discovering that 80% of all effects come from 20% of the causes.
Generally, Pareto charts are used to identify improvement opportunities, particularly when any kind of defect information is encountered.
One of the most popular graphs on this list, pie charts are intended to display the relative frequency of count data, or whole numbers or integers. In other words, pie charts are best for comparing the relationship of parts to a whole.
Whenever you hear “percent of…” or “part of… ‚” that’s an indication that a pie chart could meet your needs.
If you want to better understand the shape and spread of variable data, or data that is acquired through measurements, histograms are the way to go. The horizontal, or X axis of a histogram, represents the sale of a data. This scale is then divided into equal intervals called bins. These bins act as buckets that hold data within a specific range.
Whenever data is normally distributed, histograms will generally have a bell shape to them, like in the example mentioned earlier. The bell shape conveys that the data is fairly symmetric since each side is essentially a mirror image of the other side.
While it’s possible to create histograms with smaller sample sizes, they may not be a good representation of the true process. A good best practice is having 50 or more data points to ensure accuracy.
As its name implies, dot plots are graphical displays of data using dots. This type of graph is recommended when dealing with small sample sizes of variable data.
While the dot plot may not help us understand the overall shape of the data, we can discover whether our data seems to be clustered in a specific region or whether we have outliers or unusual data that warrants further investigation.
Individual Value Plot
Instead of summarizing the results into boxes and whiskers like dot plots, individual value plots represent each value of data individually as a dot.
With these types of graphs, our vertical, or Y axis, represents the scale of the data. Each data point is then plotted accordingly.
Often used in exploratory data analysis, box plots help us see larger patterns in the distribution of variable data.
Although box plots may seem antiquated in comparison to a histogram, they have the advantage of taking up less space, which is useful when comparing distributions between multiple groups or datasets.
Time Series Plot
A time series plot allows you to display data collected in a time sequence from any process. This can help you track how your data is trending over time and if the data points are exhibiting any patterns.
Control charts, also known as Shewhart charts or process behavior charts, are used to identify whether a manufacturing or business process is in a state of control.
In this case, upper and lower control limits are added in order to help you understand the variation in the process.