Speaking Of Reliability: Friends Discussing Reliability Engineering Topics | Warranty | Plant Maintenance

Past Good and Bad Knowledge Abstract Chris and Fred discuss the so-called ‘bedrock documents and statistics’ that are used over and over again as if they are universally correct – even though they might have nothing to do with ‘your’ machines or systems. WHY? Key Points Join Chris and Fred as they discuss the role that documents that are quoted as if their figures and conclusions they contain are unambiguously correct, for all machines, for all time. Why? Topics include: ‘I want a quick answer and don’t want to think too much about it.’ If this sounds like you … we can’t help. You are the type of person who likes the documents we are talking about. Please seek therapy. If you want to be a designer or an engineer … be a designer or an engineer. Oh … Artificial Intelligence (AI) is a tool. Not a god. Give me an example. There is a report authored by Nowlan and Heap that multiple industries use to say things like ’89 % of components never wear out.’ The trouble is that Nowlan and Heap’s report contains reliability curves that don’t make sense (like a straight line going from 100 to 0 %, which can’t happen), based on data that is incomplete as components were being replaced early on in their lives (never allowing them to wear out and produce wear out data), and focused on United Airline’s fleet of aircraft circa 1978 (which isn’t shared, so we can’t interrogate it). Aircraft are around 10 times more reliable today as they were then, so even for today’s airlines these figures would be meaningless. But Nowlan and Heap’s report is frequently cited to this day as if it is just as authoritative on things like power-generating plants and manufacturing facilities. But we have been using these figures forever and everything is fine. Really? What does fine mean to you? Making the same mistakes over and over again and having this reality normalized? If you want to be competitive, take the time to understand your machines. This will help you make decisions that would almost certainly drastically improve reliability and availability through things like adjusting maintenance regimes and so on. Enjoy an episode of Speaking of Reliability. Where you can join friends as they discuss reliability topics. Join us as we discuss topics ranging from design for reliability techniques to field data analysis approaches. Download Audio RSS Show Notes The post SOR 992 Past Good and Bad Knowledge appeared first on Accendo Reliability.

What is MLE? Abstract Chris and Fred discuss the three-letter acronym ‘MLE’ stands for? Well, it stands for ‘maximum likelihood estimate.’ Ever heard of it? Do you know what it means? Key Points Join Chris and Fred as they discuss what the MLE or ‘maximum likelihood estimate’ means … usually when using software to conduct data analysis. Topics include: What does ‘likelihood’ mean? Most models (like the bell curve, lognormal distribution, and so on) are defined by parameters. For example, the two parameters that define the bell curve are (1) the mean and (2) standard deviation. So these two parameters entirely describe the shape of the bell curve. Now let’s say you have 20 data points from some random process. What is the ‘likelihood’ that a bell curve with a mean of 10 and a standard deviation of 2 ‘fits’ your data? It has been found that if we plot out all 20 data points under this bell curve we think might fit the data, and then draw lines up from each data point to the bell curve shape … and then multiply those heights – we get the ‘likelihood’ that the bell curve fits your data. Or perhaps more correctly, we get a number that represents the ‘likelihood’ that the bell curve models the random process that gave us those 2o data points. What does ‘maximum likelihood estimate’ mean? Let’s say we are not sure if the bell curve above, or another bell curve with a mean of 9.5 and standard deviation of 1.5 is a good fit. Well … we can go through the same process for the other bell cruve, and come up with another number that represents its ‘likelihood.’ Now we can try all possible combinations of means and standard deviations and come up with likelihoods for each. Computers are good at this, and they can (for example) work out that a bell curve with a mean of 9.76234… and standard deviation of 1.8986… will have the highest likelihood of all possible bell curves. This is the ‘maximum likelihood estimate.’ It’s just another way of coming up with our ‘best fit’ or ‘best guess.’ Other approaches involve ‘regression analysis’ or ‘root mean square analysis’ where we find a line that minimizes the squares of the ‘residual’ distances between candidate lines of best fit and the data points. So my software package allows me to choose from lots of different options … including MLE … which should we use? WHAT DECISION ARE YOU TRYING TO MAKE? There are no absolute guidelines that work for every scenario. The first question you should ask is … do you need the ‘best guess’ of something or the region within which you are ‘confident’ that something lies? For example, the ‘best guess’ at your item’s time to failure might be 5.4 years. But you might only be 95 % confident that the time to failure will exceed 1.9 years. If you are trying to understand the likelihood that your item will fail within a 2-year warranty period, you might not be interested in your best guess, but instead be interested in a 95 % confidence region. Enjoy an episode of Speaking of Reliability. Where you can join friends as they discuss reliability topics. Join us as we discuss topics ranging from design for reliability techniques to field data analysis approaches. Download Audio RSS Show Notes The post SOR 991 What is MLE? appeared first on Accendo Reliability.

Impact of IoT on Maintenance Abstract Kirk and Fred discuss the impact of the Internet of Things (IoT) on maintenance and the data feedback to the manufacturer on usage and failure data. Key Points Join Kirk and Fred as they discuss on the new capabilities of products to connect with the Internet and how they open up more detailed and widespread details on the actual field use, health prognostics, and failure modes. Topics include: Automobiles, and especially the new electric vehicles, now provide so much data on driving actions such as acceleration, braking, and other use conditions that they are much more useful for improving the design from the previous model year. Telemetry that is now built into capital manufacturing equipment and systems that can provide prognostics on system wear and degradation can prevent unscheduled and costly downtime. Many manufacturers are including IoT in their devices and are receiving massive amounts of data that must be parsed and analyzed to derive valuable prognostics for improving maintenance and design Enjoy an episode of Speaking of Reliability. Where you can join friends as they discuss reliability topics. Join us as we discuss topics ranging from design for reliability techniques to field data analysis approaches. Download Audio RSS Show Notes Please click on this link to access a relatively new analysis of traditional reliability prediction methods article from the US ARMY and CALCE titled  “Reliability Prediction – Continued Reliance on a Misleading Approach”. It is in the public domain, so please distribute freely. Trying to predict reliability for development is a misleading a costly approach. You can now purchase the most recent recording of Kirk Gray’s Hobbs Engineering 8 (two 4 hour sessions) hour Webinar “Rapid and Robust Reliability Development 2022 HALT & HASS Methodologies Online Seminar” from this link. For more information on the newest discovery testing methodology here is a link to the book “Next Generation HALT and HASS: Robust design of Electronics and Systems” written by Kirk Gray and John Paschkewitz. The post SOR 990 Impact of IoT on Maintenance appeared first on Accendo Reliability.

No Trouble Found Issue Abstract Kirk and Fred delve into the crucial topic of managing returned parts and products, even those that are functioning perfectly. Key Points Join Kirk and Fred as they discuss why parts or products returned to the manufacturer seem to have no failure on their test bench. Topics include: The cost of production loss can be significant for capital production equipment failures. A field service agent may replace many circuit boards or components, known as “shotgunning,” the system when only one of the replaced parts has caused the failure. Undoubtedly, there is a lot of noise when evaluating returned systems or parts identified as failed. Some failure mechanisms, such as fretting corrosion, are from poor contact plating or contamination in connectors, and re-seating or replacing the connector removes the cause of failure. Only through further analysis of the connector’s contact surfaces can it be diagnosed Companies that are dismissive and blame the customer for misuse, dismiss failures as expected “infant mortality,” and do not perform detailed failure analysis may perpetuate a costly hidden design or manufacturing defect. The value and concern companies have for the reliability of their products depend heavily on the industry’s expectations. Everyone expects high reliability in safety-related industries such as aircraft, where pinball machines’ unreliability has been historically accepted. Enjoy an episode of Speaking of Reliability. Where you can join friends as they discuss reliability topics. Join us as we discuss topics ranging from design for reliability techniques to field data analysis approaches. Download Audio RSS Show Notes Please click on this link to access a relatively new analysis of traditional reliability prediction methods article from the US ARMY and CALCE titled  “Reliability Prediction – Continued Reliance on a Misleading Approach”. It is in the public domain, so please distribute freely. Trying to predict reliability for development is a misleading a costly approach. You can now purchase the most recent recording of Kirk Gray’s Hobbs Engineering 8 (two 4 hour sessions) hour Webinar “Rapid and Robust Reliability Development 2022 HALT & HASS Methodologies Online Seminar” from this link. For more information on the newest discovery testing methodology here is a link to the book “Next Generation HALT and HASS: Robust design of Electronics and Systems” written by Kirk Gray and John Paschkewitz. The post SOR 989 No Trouble Found Issue appeared first on Accendo Reliability.

Difference Between Quality and Reliability Statistics Abstract Philip and Fred discuss some of the basic differences and similarities between these two types of statistical toolsets. Key Points Join Philip and Fred as they discuss why it’s best to fully understand and use all the available statistial tools when faced with either quality or reliability related issues. Topics include: The simple differences due to the nature of the data. The common elements underlying both toolsets. A few examples how quality tools are good for reliability outcomes, and vice versa. Enjoy an episode of Speaking of Reliability. Where you can join friends as they discuss reliability topics. Join us as we discuss topics ranging from design for reliability techniques to field data analysis approaches. Download Audio RSS Show Notes   The post SOR 988 Difference Between Quality and Reliability Statistics appeared first on Accendo Reliability.

Dealing with Missing Data Abstract Philip and Fred discuss a few challenges and approaches to deal with missing data. Key Points Join Philip and Fred as they discuss what can be done when a dataset has missing data. Topics include: Data collection and storage issues Dealing with data points within the overall dataset Dealing with missing left censored data Enjoy an episode of Speaking of Reliability. Where you can join friends as they discuss reliability topics. Join us as we discuss topics ranging from design for reliability techniques to field data analysis approaches. Download Audio RSS Show Notes   The post SOR 987 Dealing with Missing Data appeared first on Accendo Reliability.

Pushing the Limits Abstract Greg and Fred discuss pushing the limits for personal and professional development.  Greg discusses wearing a pink tutu to Oregon Country Faire and Burning Man.  Fred discusses pushing the limits for product testing and product development. Key Points Join Greg and Fred as they discuss pushing personal, design, and testing limits. Topics include: What is a successful reliability test? What are the limits that we impose and design in product testing. How do we measure product test and development success? How do we ensure that our product testing matters? Enjoy an episode of Speaking of Reliability. Where you can join friends as they discuss reliability topics. Join us as we discuss topics ranging from design for reliability techniques to field data analysis approaches. Download Audio RSS Show Notes   The post SOR 986 Pushing the Limits appeared first on Accendo Reliability.

Pursuing Ribbons Abstract Greg and Fred discuss the importance of the pursuit of ribbons.  What are ribbons?  They are the  badges, degrees, brands, and certificates that we all strive for.   They confer knowledge, skills, and abilities to others.  They are the things of life that we do for personal improvement and enrichment.  Greg believes the pursuit of ribbons or recognition is what we all do from the day we are born to the day we pass on.  Fred believes knowledge by itself is sufficient.  What do you think? Key Points Join Greg and Fred as they discuss why pursuing ribbons like green, yellow, and black belts are so important to quality and reliability engineers.  Topics include: What are the types of ribbons we pursue? Why do ribbons matter? Why do you go after the ribbons you do? Enjoy an episode of Speaking of Reliability. Where you can join friends as they discuss reliability topics. Join us as we discuss topics ranging from design for reliability techniques to field data analysis approaches. Download Audio RSS Show Notes   The post SOR 985 Pursuing Ribbons appeared first on Accendo Reliability.

Instead of Reliability Prediction Abstract Chris and Fred discuss reliability prediction and how it can relate to the ‘design’ phase when there is no data. How do you ‘predict reliability?’ Key Points Join Chris and Fred as they discuss what reliability prediction is. There are quite a few answers to this question … and some of them can be helpful! Topics include: What is ‘reliability prediction’ #1? Depends on who you ask. Reliability prediction is often seen as looking up a book or standard of ‘failure rates’ for specific ‘types’ of components. These ‘prediction tables’ include generic descriptions of things like ball valves, small two stroke engines and so on. The many problems with this is that these tables are not compiled with a lot of rigour, neglects the fact that the ‘same’ component from different suppliers can much more or less reliable than any other … and once you have a ‘number’ for that component reliability is never thought of again. What is ‘reliability prediction’ #2? What it should be is a more analytical, considered approach to looking at your design and analyzing what it’s reliability is. And this doesn’t need ‘data.’ There is nothing wrong with engineering judgment, and then understanding the (lack of) confidence in the results. This is OK. But it allows you to work out which ‘vital few’ components drive reliability performance. And those are the ones you study. What is ‘reliability prediction’ #3? Not the MTBF. That is the only number you get from those tables. But the MTBF is not reliability. … and why aren’t we trying to IMPROVE reliability before MEASURING it? Good question! There are many companies out there that make really reliable stuff without ever measuring how reliable it is. That’s right. So if you haven’t finished designing your product (presumably because there are more design changes to be made to improve the design), then why would you even try to come up with a number beforehand? Enjoy an episode of Speaking of Reliability. Where you can join friends as they discuss reliability topics. Join us as we discuss topics ranging from design for reliability techniques to field data analysis approaches. Download Audio RSS Show Notes The post SOR 984 Instead of Reliability Prediction appeared first on Accendo Reliability.

Good Reliability Testing Abstract Chris and Fred discuss the difference between good and poor reliability testing. Key Points Join Chris and Fred as they discuss what makes a meaningful and valuable reliability test. And more importantly, how to use test results to make valuable decisions. And to make a decision – you need to read the reports! Topics include: Leadership needs to be bought in. There are plenty of really good tests where for whatever reason, the leadership team won’t read (or believe) the reports. NASA did this when they disregarded reports that suggested that detaching foam insulation couldn’t damage the wings of the space shuttle (which is precisely what happened in the 2003 Columbia disaster). And then there are leadership groups that give reliability engineers two data points and then want to demonstrate that the item will have a 99 % chance of lasting 20 years. And it matters ‘when’ you do testing. Most militaries focus on receiving ‘production ready’ tanks, planes, and weapon platforms after years and millions of dollars of development to be subjected to ‘acceptance testing.’ This ‘acceptance testing’ can often be exhaustive, involve lots of data points … but because it is done after the production process is complete, if the item fails the test, there is no feasible way of walking away from buying that item. There is no time or money left to do anything to improve the design, so the test results get ‘doctored’ or data points are explained away with minor corrective actions so that the fundamentally unchanged item is now deemed reliable enough. Pointless. There needs to be a strategy … tied to a decision. If at test doesn’t influence a decision, then don’t do it. If you are going to accept an item regardless of how it performs on the test … don’t do it. If you are going to conduct HALT testing, but the boss will never read the report … don’t do it. Work out what decisions you need to inform, work out if there is a test that can help, and only then implement that test. And when it comes to reliability … nothing beats actually understanding ‘how’ your thing fails. If you are a reliability engineer who only gains confidence based on data points in a spreadsheet, then you are not a ‘good’ reliability engineer! Enjoy an episode of Speaking of Reliability. Where you can join friends as they discuss reliability topics. Join us as we discuss topics ranging from design for reliability techniques to field data analysis approaches. Download Audio RSS Show Notes The post SOR 983 Good Reliability Testing appeared first on Accendo Reliability.