Your Reliability Engineering Professional Development Site
The chi-square (Χ2) test provides the basis for the second case of hypothesis tests for variances. In this case, we want to compare observed and expected frequencies, or counts, of outcomes when there is no defined variance. In other words, we are working with attribute data. [Read more…]
Hypothesis testing of data may include two populations that have un-equal standard deviations. The t-test for differences considered in a previous post used the assumption of equal variances to pool the variance value. In this test, we want to consider if one population is different in some way than the other and we use the samples from each population directly even if the population have difference variances. [Read more…]
Hypothesis testing of paired data may include two populations that have the equal standard deviations. The t-test for differences considered in a previous post used the standard deviation of the differences. In this test, we want to consider if one population is different in some way than the other and we use the samples from each population directly. [Read more…]
Hypothesis testing previously discussed (link to past posts) generally considered samples from two populations. Maybe the experiments explored design changes, different component vendors, or two groups of customers. Occasionally you may find data that has some relationship between the samples, or where the samples are from the same population. Paired (or matched) data involves samples that are related in some meaningful way. [Read more…]
Statistics is the language of variation. Everything varies, and we use variance (σ2) to describe the spread of the data. For any experimental work aimed at making improvements, whether in the design, manufacturing process or field performance, there are two ways to make improvements. Move the center of the distribution, or reduce the spread of the data. [Read more…]
This is also called the “p test”
When comparing proportions that are from a population with a fixed number of independent trials and each trial has a constant probability of one or another outcome (Bernoulli experiments) then we can use a p test. p is the probability of success, and 1-p is the probability of failure. Caution: stay consistent once you define success otherwise, like me, you’ll have a bit of confusion. n is the number of trials. [Read more…]
A continuous distribution is useful for modeling time to failure data. For reliability practitioners, the Weibull distribution is a versatile and powerful tool. I often fit a Weibull when first confronted with a life dataset, as it provides a reasonable fit given the flexibility provided by the distributions parameters. [Read more…]
In another post, I started the discussion about variability and interquartile range. This is part 2 of that discussion and will focus on variance.
With rare exception, most distributions or groups of data require more than one parameter (or statistic) to fully describe both the location and spread (scale and shape) of the group of data. Is the data clumped tightly about some value, or spread out over a wide range. [Read more…]
This week I received a question from the ASQ Librarian concerning a person’s question about one of the CRE Question bank questions. It was a nice two-part question concerning a hypothesis test of a sample means value and degradation.
Here’s the question as sent over for consideration. [Read more…]
Last week I posted about the Binomial probability density function, and it is useful when calculating the probability of exactly x successes out of n trials given p probability of success for each trial.
Well, what happens if you want to know the probability of 2 or more successes for example? [Read more…]
There are many cases where the results of an experiment (or trial) are either it works or it doesn’t, pass/fail, success/failure. Only two possible outcomes one of which we define as success the other outcome as failure. The binomial distribution is suitable if the random variable (the set of experimental or trial outcomes) when
Running through a couple of practice CRE exams recently (yeah, I know I should get out more…) found a few formulas kept coming up in the questions. While it is not a complete list of equation you’ll need for the exam, the following five will help in many of the questions. They seem popular maybe because the relate to key concepts in the body of knowledge, or they are easy to use in question creation. I do not know why. [Read more…]
Received a sample problem from someone preparing for the CRE exam saying it was a tricky one.
The hazard rate function for a device is given by
0.001 if t ≤ 10 hours and 0.01 if t > 10 hours
What is the reliability of this device at 12 hours?
I first draw the hazard function [Read more…]
Down to the last week of preparation for the exam on March 2nd. Good luck to all those signed up for that exam date. Time to focus on preparing your notes, organizing your references and doing a final run through of practice exams. [Read more…]
A skill needed for the CRE exam is the ability to look up probabilities given a z-value using a standard normal table. It’s old school, I know, yet without software, you most likely will have to find a few values in this manner.
A z-value is the number of standard deviations from the mean at least that’s how I think of it. The area under the curve to the right or left of the z-value is then the probability of interest. Of course, I’m talking about the normal distribution probability density function or what we commonly all the ‘bell-shaped’ curve. [Read more…]