Mode

The mode is the number which appears most frequently in a set of numbers. For a finite data set, as in a sample of measurements, the mode would be the number that appears the greatest number of times.

In a distribution curve, the mode is the number that corresponds to the highest point of the curve.

In continuous improvement, the mode can be used as an indicator of the central tendency of a process, machine, etc.

How can this be applied? Consider that most processes will display a skewed distribution for the cycle time. There is a typical time the process takes. Sometimes, an operator might beat the time, but rarely by very much. There is only so fast an operator can go. If there was a better process, it would be done all the time. On the other hand, when an operator misses cycle time, it can be by just a few seconds, or substantially longer when the problem is more serious. In this situation, the mode is a good indicator of repeatability.

Generally speaking, if you reduce the number of problems, the value of the mode will stay the same, but have a higher frequency. In effect, you are simply stripping the delays off the values from the tail. If you improve the speed of a process, the mode will shift.

Knowing the mode in this situation will help you with your planning. You can determine, for example, how to staff appropriately to handle the normal work and how many people (i.e. floaters) you might need to handle problems.

Multiple Modes

When dealing with a finite data set, it is common to run across ties in which there are two (or more) numbers occurring with the same frequency. In this case you would have more than one mode.

If, however, no number occurs multiple times, there is no mode. This is uncommon in large data sets.

Mode and Decimals

The mode is unique from the other two common measures of central tendency (mean and median) in one main respect. It doesn’t play as nicely with decimals.

The other two don’t change at all when you throw in numbers with many decimals in it. 23.4576 is just another number to the mean equation, and when you rank the numbers in a series, you can find the middle number no matter how many digits you use.

But when you have a data set that is very precise, meaning many digits, it can get a bit confusing.

Let’s say that in a data set, you have 12 values in the 16-17 range, but using two decimal digits, there are no multiple values. But by chance, you also have 3 data points in the 18-19 range, but two of them are both 18.37. That might be the only data point that repeats, so you have a mathematical dilemma.

The mode technically is 18.37, but in reality, the 16-17 range is the most common.

Make sure you understand what the data is telling you. If you are using whole numbers, the mode is rather straightforward. Nearly every example you can find on the internet also uses whole numbers. If you are using any sort of decimal, the value of using the mode can become a bit murky.

You may have to switch to looking at it in terms of ranges rather than the actual value.

Related Terms

Mode is generally used in conjunction with median (the middle value), and the mean (average). Those three terms, taken together, can give good insight into the data set, and are said to describe the “central tendency” of the data. Taken alone, each offers only a small portion of the information you need to make decisions.