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

Interpreting formulae and not just generating numbers … Abstract Chris and Fred discuss how formulae (or formulas!) can be very important … but it is way more important to understand what formulae represent, and how they work. Listen to this podcast if you want to learn more! Key Points Join Chris and Fred as they discuss how important it is to understand what formulae represent, including all those pesky parameters that are represented with Greek letters. Why is that? Topics include: What is the ‘mean’ or ‘average’ of something? Some of us might be able to quote a formula for the mean. But do you actually know what it is? Put simply … the mean is the ‘balance’ point of your data points if they sit on a pivoting plank. Think of the ‘balance point’ of the histogram of your data points. That’s it. What about the standard deviation? Well … once you get that histogram, find its moment of inertia (i.e. how hard it is to spin around) and then take its square root. Again … that’s it. But did you know that? OK … but why do I need to know how to interpret formulae? Take (for example) Weibull plotting. For those who don’t know what it is, it is just a special way of visualizing data that allows us to see failure characteristics if we know what to look for. If you know what the shape of these data points means, you can see where you need to service (i.e. when things start to wear out), or how many things are suffering from infant mortality. If you don’t know what equations mean, then all you can calculate are parameters, means, and other numbers that you won’t know how to help you make better decisions. Then there is error checking. If you know what your formulae and equations represent, then you can have a good guess at what the answer should be before you get the number. And this can help you make sure you didn’t make a simple error. If you are modeling the time to failure of something, and a computer tells you that the parameter ‘μ‘ of the normal distribution you use to model time to failure is negative, then you should immediately know you have a problem. Why? Because that parameter is the ‘mean’ or the balance point of data points that must be greater than zero. This approach has saved so many reliability engineers over the journey! 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 936 Interpreting Formulae appeared first on Accendo Reliability.

Change Management Success Abstract Greg and Fred discuss organizational change management and behavior change.  Greg’s change management is sell the sizzle.  Fred’s is give them the steak.  Discover why? Key Points Join Greg and Fred as they share change management stories including tips and tools for success.   Topics include: What is change management? How can you induce change management and behavior change? What are leadership, management, and scientific methods for change management? 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 935 Change Management Success appeared first on Accendo Reliability.

Contingency and Risk Planning Abstract Greg and Fred discuss various situations where contingency and risk planning may be required. Key Points Join Greg and Fred as they discuss contingency and risk ‘what if’ planning.  Topics include: When should a contingency and risk plan be developed. How to develop a contingency plan. What are the critical elements of a plan. 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 934 Contingency and Risk Planning appeared first on Accendo Reliability.

FTA vs RBD Abstract Chris and Fred discuss the differences, pros and cons of FTA (fault tree analysis) and RBD (reliability block diagrams). Need to learn more about how to model system reliability? Do you struggle to spell FTA or RBDs? Listen to this podcast! Key Points Join Chris and Fred as they discuss the differences between these things we call fault trees and reliability block diagrams. If you need to model system reliability, you really need to know what these are … and which one might work best for you. Topics include: System reliability modeling. This involves creating something that allows you to convert component reliabilities into system reliability. So if your system has seven components, and their reliabilities at 2 years are 75%, 89%, 45%, 92% and 56% … what is the reliability of your system? A system reliability model can use these figures to give you that number. What is a fault tree (FT)? It is a diagram that uses logic gates and circles (called basic events – representing component failure) that are connected with lines up to the ‘top event’ (usually representing system failure). The logic gates are often ‘AND’ or ‘OR’ gates, and they allow you to work out what combinations of components need to fail for your system to fail. What is a reliability block diagram (RBD)? It is a diagram where a ‘block’ represents each component, and lines are going from left to right through these blocks. If you can follow one of these lines from left to right passing only through working ‘blocks’ or components, then your system works. RBDs can often mimic the physical layout of your system where FTs do not. Why use one over the other? RBDs and FTs are usually just as useful at modeling system reliability as each other. While FTs can’t mimic the physical layout of a system, they are often much easier for us to analyze because they are laid out logically. FTs can also incorporate other things besides components failing (such as human error, environmental changes and so on). RBDs can’t do this. FTs are often useful for root cause analysis, so if you already have a FT for your system from RCA, perhaps it is easier to keep using them. So which one should you use? It’s up to you! Know the pros and cons of each one, and work out which one might be better for your problem. Also … perhaps your team is more fluent with FTs instead of RBDs, or vice versa. Take this into consideration. Work out what problem you are trying to solve, which decision you are trying to inform, and then make an informed decision! 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 933 FTA vs RBD appeared first on Accendo Reliability.

K out of N Systems Abstract Chris and Fred discuss about this thing called a ‘k out of n’ system … and how most people don’t really know about how they contribute to system reliability. Do you know what the reliability curve of a ‘k out of n’ system looks like when compared to series or parallel systems? Don’t know what these systems are? Listen to this podcast! Key Points Join Chris and Fred as they discuss what a ‘k out of n’ system and how many people know about how they work (sort of), but are not really sure of how they can help (or hinder) reliability. Topics include: What is a ‘k out of n’ system? Any system with a total of ‘n’ (usually identical) components, where the system only needs ‘k’ of these components to work for the system to work. What are ‘series’ and ‘parallel’ systems? A ‘series’ system is a system that might have ‘n’ components, where all of those components need to work for the system to work. Or from another perspective, if any component fails, the system fails. So a series system is an ‘n out of n’ system. A ‘parallel’ system is a system that might have ‘n’ components, but only one of those components needs to work for the system to work. So a parallel system is a ‘1 out of n’ system. So series sytems have more ways to fail, meaning their reliability curves are less than those for any of the components. Parallel systems have more ways to work, meaning their reliability curves are greater than those for any of the components. … and ‘k out of n’ systems? What do their reliability curves look like? For example, a ‘2 out of 3’ system mimics a parallel system early in its life, but it mimics a series system late in its life. But … as a rule – we are only interested in the early life of a component, so the only characteristic that matters is that it mimics a parallel system when young. So why use ‘k out of n’ systems? Lots of reasons. You need additional reliability performance (when young), it allows you to use smaller components (for example, a ‘2 out of 3’ pump system can share the load between two components or pumps, meaning they can be smaller), it allows degraded performance and lots of other reasons. Interested? Find a way to learn more! 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 932 K out of N Systems appeared first on Accendo Reliability.

FMEA Detection – Useful? Abstract Carl and Fred discuss a reader question about assessing risk related to detection, as part of FMEA procedure, specifically its usefulness and application. Key Points Join Carl and Fred as they discuss FMEA detection in FMEA. Topics include: What is FMEA detection? How is FMEA detection assessed in FMEA? What is the value of FMEA detection? What is the consequence if FMEA detection is not used as part of FMEA procedure? What is difference between detection during product development and detection in service? Why does high likelihood of detection get a low score in detection risk assessment? FMEA Detection rating guides both design improvement and testing improvement Supplemental FMEA for Monitoring and System Response What is Criticality in FMEA? RPN flaw and how is it adddressed What if all tests are passed yet there is substantial warranty problems? What does that say about detection risk? Cultural issues can impede solving detection problems Severity and occurrence can be more important then detection; consider Action Priority 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 931 FMEA Detection – Useful? appeared first on Accendo Reliability.

Reliability Goals and Requirements Abstract Carl and Fred discuss the importance of well-written reliability requirements and whether they should be goals or actual requirements. Key Points Join Carl and Fred as they discuss reliability goals and requirements. Topics include: Project goals or metrics are not sufficient What to do if reliability is difficult to measure Functional tests do not necessarily measure reliability Reliability requirements should include all four aspects of the definition of reliability, measurable during product development, and incorporated into technical specifications Reliability requirements in technical specifications should be clear and not disjointed Which components need reliability requirements? How to determine critical parts? Relation between warranty and reliability Reliability is a requirement just like other requirements Difference between goals and requirements What happens when testing systems are overloaded? Using models to verify reliability 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 930 Reliability Goals and Requirements appeared first on Accendo Reliability.

Redundancy Risk Balance Abstract Greg and Fred discuss what is redundancy and why it’s a critical business tool. Key Points Join Greg and Fred as they discuss process and product redundancy in a world that wants to reduce friction and be lean. Topics include: What is redundancy? Why redundancy is important in a lean world? When and how should redundancy be used? 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 929 Redundancy Risk Balance appeared first on Accendo Reliability.

Risk and Reliability Lenses Abstract Greg and Fred discuss risk and reliability lenses for solving today’s dynamic problems. Key Points Join Greg and Fred as they discuss the intersection of risk and reliability lenses under uncertainty.  Fred says: “Reliability is the critical lens to understanding  and addressing today’s engineering problems”.  Greg, a one-trick pony, says: “Risk lens is the critical lens for looking at all of today’s uncertain problems.”   So, who’s right?  Why does this matter to engineers? Topics include: What’s the intersection of quality and risk? Why are risk, quality, and reliability lenses important? Why are today’s problems dynamic and uncertain? How can we or should we as engineers approach uncertain systems and dynamic problems? 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 928 Risk and Reliability Lenses appeared first on Accendo Reliability.

Reliability from WWII Abstract Chris and Fred discuss some of the things we have learned in the world of reliability directly due to World War II. Not many people know that around one third of today’s medical vaccines were developed out of necessity from World War II … and something similar happened when it came to making things reliable. Key Points Join Chris and Fred as they discuss what role World War II played in the field of reliability engineering that we can still be thankful for today. It is not often that we go through a ‘history’ lesson in the world of reliability engineering, but in this case there is a lot to be learned about how to do things right (or wrong). Topics include: The ‘Japanese Economic Miracle’ … which refers to the inexplicable rise of the Japanese economy after World War II. What happened? Japan turned into a manufacturing powerhouse that somehow managed to build products that were high quality, very reliable, and cheaper than ‘Western’ products. How did they do this? The allied forces essentially took control from the mid-1940s, and essentially allowed a technocratic domestic government to form. They poured money into the rebuilding of Japanese industry (in part due to fear of the ‘communist’ threat that a renewed Japan could perhaps buffer against). While the country received a lot of resources, this is no doubt a ‘Japanese-owned’ good news story. And most countries have been trying to catch up since. Let’s not forget electronics. Electronics (like the capacitors, vacuum tubes and resistors embedded within things like radars and radios) are infuriatingly non-intuitive. We can look at a broken transmission for a tank, and usually see what failed. We can also look at how dirty lubricants are, and have some idea about how long it will take before we need to replace it. We can’t do this with electronic components (at least easily). Nor could we easily understand how vibration might rupture solder joints. So we needed to create a more scientific approach that uses statistics. They also tended to put the right people in charge. Starting with the generals. And all the way down to the technical domains. Do you think a scientist like Oppenheimer would ever be allowed to be in charge of a project as important as the nuclear bomb in today’s military society? (the answer is no … we usually have the same ‘warfighting’ general, pilot or admiral in charge of everything). Funny how military projects are always late and over budget … 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 927 Reliability from WWII appeared first on Accendo Reliability.