Fundamentals of Failure Analysis

podcast episode with speaker Fred Schenkelberg

Failure analysis is not just about finding the root cause of a failure; it’s also about preventing future failures. Yet, the process starts much like a crime scene. Understanding some basic steps and approaches helps you fully grasp the failure and its impact or implications.

Sometimes, just identifying a failure is the difficult part. Sometimes, it’s the collection of clues or evidence. Sometimes, it’s the skills and tools needed to fully understand the root cause. The intent of any failure analysis is to understand the problem well enough to craft and implement short and long-term solutions.

Let’s explore the basics of failure analysis, the basic steps, and some ways to identify and avoid ways a failure analysis may fail. If you like, grab your Sherlock Holmes cap and magnifying glass and prepare to improve your ability to learn from every failure.

[Read more…]

Parts Count Reliability Prediction (is awful)

podcast episode with speaker Chris Jackson

Isn’t reliability prediction a very important part of what reliability engineers should do? Shouldn’t we want to (for example) predict the improvement in warranty reliability a design change will make? Well … yes! But then there is ‘parts count reliability prediction.’ As is the case for lots of reliability engineering terms, there are phrases that are used to describe very specific types of concepts. And ‘parts count reliability prediction’ is awful. And in this webinar, we will explain why … using dogs! If this intrigues you … see you there!
[Read more…]

Fundamentals of Regression Analysis

podcast episode with speaker Fred Schenkelberg

Most of us are familiar with the data analysis method called regression analysis. Maybe not familiar enough in some cases. The analysis allows us to identify and model trends or relationships. For example, we might examine the relationship between the speed of a production line and the number of out-of-spec items.

While regression analysis is a common method that many of us use regularly, understanding some of the limitations and options can enhance our analysis insights and accuracy. Understanding the assumptions and how to test them is a fundamental skill worth mastering.

Let’s explore a few of the limitations, checks, and options when taking a look at some data. From checking assumptions to assessing the goodness of fit, and a quick exploration of various approaches beyond simple linear regression.

[Read more…]

Software Root Cause Analysis (with Fault Trees)

podcast episode with speaker Chris Jackson

Everyone knows how things fail … right? We know that we often have to deal with defects or design oversights that mean parts rub, crack, or erode quicker than we would like. And then something snaps, breaks or stops conducting, so our product stops working. But then there is software … which never fails. And that is because software only ‘fails’ when a coding error or bug makes it do something we didn’t want it to do. So it’s simple human error … which we can’t model, prevent, or create safeguards to prevent these errors … right? Wrong! And that is because software bugs are so common that we know how most of them happen. Which means we can work out what went wrong to allow that bug to happen. And that means we can quickly and efficiently improve software reliability!
[Read more…]

Fundamentals of Life Estimation

podcast episode with speaker Fred Schenkelberg

We often face questions, such as: “How long will this last?”, “Will this design survive for at least 5 years?”, and “When should we recommend servicing or replacement?” Being able to provide life estimates is a common skill we need to master.

Life testing is not the only method available to us. We can provide a guess (based on engineering judgement), to modeling and simulation, to field data analysis. And, there are more suitable methods depending on the particular situation.

Let’s explore the range of options available to estimate life for an item, plus the pros and cons of each method. You need to be aware of your options and thus be able to provide the best estimates when asked.

[Read more…]

Fundamentals of Environmental Testing

podcast episode with speaker Fred Schenkelberg

One of the elements of reliability concerns where and how the product will be used. This includes understanding the environmental stresses (such as wind, insects, and temperature) as well as the product’s handling, usage frequency, throughput, and other relevant factors.

In some cases, we need to evaluate our product’s ability to survive the expected environmental stresses. Conducted well, this testing will provide vital information to the design team about the product’s capabilities. It also may reveal design weaknesses.

Let’s explore when and how to identify which types of environmental testing should be done. The intent is to conduct specific environmental tests that will provide value to the organization and its customers. Doing any environmental test just because we can is not a valid reason to do so.

[Read more…]

What is Reliability?

podcast episode with speaker Chris Jackson

Everyone knows what reliability is. And that’s the problem. We have used the term ‘reliability’ in normal, everyday speech since we were children. So that means we have decades of preconceived ideas about what reliability is (and is not). But that means there are lots of different ideas about what reliability is. Some people think this thing called the ‘MTBF’ is a measure of reliability (it isn’t). And neither is the ‘failure rate.’ Neither are things called the ‘L10’ and ‘B10’ lives. So if you want to find out what reliability actually is (no equations) … this webinar is for you!
[Read more…]

Fundamentals of Success Testing

podcast episode with speaker Fred Schenkelberg

Product life testing is expensive. Thus, many seek ways to minimize the cost. Two common methods is to use the fewest samples that may lead to a meaningful result, and using acceleration methods to shorten test times. Designing a test based on the pass/fail (binomial distribution) approach provides a path to using very few samples, providing a clear pass/fail result, and is relatively risky.

Let’s examine the success testing approach, the assumptions, benefits and faults. Plus, what other options you may have available to get the information concerning expected product/component life. Knowing the details of success testing enables you to use it when it’s appropriate and when to avoid its use.

[Read more…]