Go offline with the Player FM app!
Speaking Of Reliability: Friends Discussing Reliability Engineering Topics | Warranty | Plant Maintenance
«
»
Forecasting Returns
Fetch error
Hmmm there seems to be a problem fetching this series right now. Last successful fetch was on September 02, 2024 16:33 ()
What now? This series will be checked again in the next day. If you believe it should be working, please verify the publisher's feed link below is valid and includes actual episode links. You can contact support to request the feed be immediately fetched.
Manage episode 407557479 series 2359263
Forecasting Returns
Abstract
Chris and Fred discuss how you go about forecasting returns … and understanding things like warranty reliability for products that are ‘shipped all over the place? … at different times? Sound familiar?
Key Points
Join Chris and Fred as they respond to a question from one of our listeners where he was struggling to understand how to forecast returns based on really simple data … essentially the number of things sold in a given month versus the number of returns for the same month. It was proposed that we simply use the exponential distribution, find the lambda and so on … but this really won’t work.
Topics include:
- But we don’t have the data? Really? So you have absolutely no way of checking when your customers purchased your product so you can check if the returns are in the warranty period? You are happy to ‘assume’ that any customer returning a product will only do so if they have checked that the product is in the warranty period? I don’t think so. So the data is almost always there in your organization … it’s just that stove pipes, fiefdoms, and other human-made barriers exist to get access to it.
- It usually comes down to how your organization manages data. If you don’t believe me … then who (outside of your organization) is to blame for you ‘not having’ that data? Hackers? Spies? Aliens? You need to have sales data so you can complete your tax returns. It is entirely up to your organization to take managing data seriously.
- … and you can’t assume a constant hazard rate. This is tempting, as it deals with failure rates. And an easy way to calculate a number (that looks like a failure rate) is to divide the numbers of things coming in by the numbers going out. But of course, the easiest way to drive down this number (make it look good) is increase production for a month, which increases the number of things going out which makes the ‘number’ smaller. If you shut down for Christmas (for example) and have one or two products leaving your facility, then your ‘number’ goes through the roof. None of these numbers have anything to do with actual failure rates …
- So if your organization sucks at giving you the data? Fight for it. Manually chase it down. Ring the right people. Explain what’s going on. Perhaps get people in power to help. But don’t give up and just find a number that doesn’t mean anything.
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.
Show Notes
The post SOR 949 Forecasting Returns appeared first on Accendo Reliability.
650 episodes
Fetch error
Hmmm there seems to be a problem fetching this series right now. Last successful fetch was on September 02, 2024 16:33 ()
What now? This series will be checked again in the next day. If you believe it should be working, please verify the publisher's feed link below is valid and includes actual episode links. You can contact support to request the feed be immediately fetched.
Manage episode 407557479 series 2359263
Forecasting Returns
Abstract
Chris and Fred discuss how you go about forecasting returns … and understanding things like warranty reliability for products that are ‘shipped all over the place? … at different times? Sound familiar?
Key Points
Join Chris and Fred as they respond to a question from one of our listeners where he was struggling to understand how to forecast returns based on really simple data … essentially the number of things sold in a given month versus the number of returns for the same month. It was proposed that we simply use the exponential distribution, find the lambda and so on … but this really won’t work.
Topics include:
- But we don’t have the data? Really? So you have absolutely no way of checking when your customers purchased your product so you can check if the returns are in the warranty period? You are happy to ‘assume’ that any customer returning a product will only do so if they have checked that the product is in the warranty period? I don’t think so. So the data is almost always there in your organization … it’s just that stove pipes, fiefdoms, and other human-made barriers exist to get access to it.
- It usually comes down to how your organization manages data. If you don’t believe me … then who (outside of your organization) is to blame for you ‘not having’ that data? Hackers? Spies? Aliens? You need to have sales data so you can complete your tax returns. It is entirely up to your organization to take managing data seriously.
- … and you can’t assume a constant hazard rate. This is tempting, as it deals with failure rates. And an easy way to calculate a number (that looks like a failure rate) is to divide the numbers of things coming in by the numbers going out. But of course, the easiest way to drive down this number (make it look good) is increase production for a month, which increases the number of things going out which makes the ‘number’ smaller. If you shut down for Christmas (for example) and have one or two products leaving your facility, then your ‘number’ goes through the roof. None of these numbers have anything to do with actual failure rates …
- So if your organization sucks at giving you the data? Fight for it. Manually chase it down. Ring the right people. Explain what’s going on. Perhaps get people in power to help. But don’t give up and just find a number that doesn’t mean anything.
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.
Show Notes
The post SOR 949 Forecasting Returns appeared first on Accendo Reliability.
650 episodes
所有剧集
×Welcome to Player FM!
Player FM is scanning the web for high-quality podcasts for you to enjoy right now. It's the best podcast app and works on Android, iPhone, and the web. Signup to sync subscriptions across devices.