When a process is said to be in control, statistically speaking, that means that all the variation can be attributed to common causes. All of the observed variation is just a function of the natural randomness built into a system or process. In a nutshell, an in control process is Read more…
A control, or control group is a tool used to confirm whether changes are actually having an effect. The control group is exposed to the same conditions as the test group with the exception of the variable that is being examined. For example, you may be experimenting whether increasing tire Read more…
The central limit theorem, in layman’s terms, says that regardless of the shape of the distribution of a population, in most cases, the mean of random samples taken from that population will approximate a normal distribution. The larger the sample, or the larger the number of samples, the closer the Read more…
The “10-Foot, 3-Second” rule is a rule of thumb regarding visual controls. From 10 feet away, you should be able to assess the status of an operation within 3 seconds. This does not mean that a novice needs to be able to do it, but someone who is familiar with Read more…
The Analytic Hierarchy Process, or AHP, is a decision-making tool developed in the 1970’s by Thomas L. Saaty. Its key characteristics are that it breaks big decisions into smaller ones, and relies on direct, one-on-one comparisons to make judgments. Essentially, the analytic hierarchy process breaks criteria into progressively smaller criteria, Read more…
The alternative hypothesis is the assumption that there is a statistically significant difference between two sets of data. This is essentially the opposite of the null hypothesis. The alternative hypothesis is accepted if the null hypothesis is rejected. What this might mean in practice is that there is a statistical Read more…
“Null hypothesis” is a statistical term that basically means that there is an assumption that there is no statistical difference between observations. For example, the null hypothesis would say that any differences between a sample and a population would be due only to random chance. Statistical testing then confirms or Read more…
The “5Ps of Effective Meetings” is a simple memory device to help you, not surprisingly, run more effective meetings. Because of the rising complexity of business operations, problems increasingly require a collaborative solution. Those collaborative solutions tend to require meetings. Unfortunately, this increased demand for meetings also tends to increase Read more…
Continuous data can have any value within a given range. Compare this to discrete data which is limited in the values it takes. For example, the number of dots on a pair of dice or the number of wheels on a car limit you to a finite set of values. Read more…
PDSA stands for Plan-Do-Study-Act, or less commonly, Plan-Do-Study-Adjust. It is a structured, iterative problem-solving approach popularized by W. Edwards Deming, who originally was mentored on the process by Walter Shewhart. With that origin in mind, it should come as no surprise that this method is also known as the Deming Read more…
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 Read more…
The median is the point at which there are the same number of values above it as there are below it. This can apply to a sample, a full population, or distribution curve. When the data consists of a finite set (rather than a distribution curve), if there is an Read more…