Suppose in the manufacturing industry, we face a lot of issues regarding whether our production is attaining a specification target or not and in such scenarios it proves to be worthy enough. It can be valuable troubleshooting aids. It is used to study the density of data in any given distribution and understand the data that repeat more often. We can use to make comparisons of different machines, operators, vendors, etc.
Histogram was first introduced by Karl Pearson to show the probability distribution of continuous data.
What is a Histogram?
Histogram is a pictorial representation which displays the data value in bars to show the frequency of data items in successive numerical intervals of equal size. Interval sizes are shown on the X – axis and the frequencies on the Y-axis. The height of each bar represents the frequency of each interval size. It displays data in such a way that makes it easier to see the dispersion and central tendency in a process. We can use the histogram to examine the shape and spread of data. It is best used when the sample size is more than 50.
For example, in a steel manufacturing industry – a quality inspector wants to know the number of sheets produced within the thickness range. In another way, it can be used to identify whether the process is manufacturing the steel sheets within the required limit or not. And it can be either less or more according to the range.
Similarly, in a pharmaceutical company, a quality inspector wants to find out whether the caps of bottles are fastened correctly or not. It is important to check the bottles as it may spill out or difficult to open when it is fastened loosely or tightly. Here he can set a torque value require to open a cap say 20 and then he can select a sample of bottles. And plot in a histogram to see the visual representation.
Following are some of the usages of the histogram: –
Compare a process to requirement – We can use a histogram to compare with our requirement limits. Draw a specification line over a graph and then we can visualize whether the process is producing within the limit or off limit.
Comparison of two or more processes – We can plot the output of two or more processes in the histogram and then compare which one is giving the required results.
Outlook of histogram
→ Simple – It is a simple graph which displays the data of one variable.
→ With fit – It displays the data of one variable with a fitted distribution curve.
→ With groups – It displays the distribution of values which are divided into groups.
When to use histogram?
→ Data are numerical in nature.
→ To find the central tendency of data.
→ To examine the shape of data distribution.
→ Identification of an outlier and to check the variation in a process.
→ Most importantly to verify whether the output of a process is normal or not.
• The diagrammatic representation gives the outlook of data.
• It gives an easy understanding of a process line.
• Helps to make the decision-making process easy to deliver.
• Suitable in many fields like manufacturing, service sector, academics and so on.
A quality engineer wants to find out whether the process is producing the sheets of thickness 22 (target) or not. And can also check whether it is producing less or beyond the limits. So he creates a simple histogram and added a reference line of thickness 22.