Statistics and Reliability

Abstract

Chris and Fred discuss what came first … reliability or statistics? Reliability of course! There are plenty of organizations that focus on baking reliability into designs without the need for statistics. But statistics are a part of reliability engineering. So what is the relationship between the two? Listen to this podcast to learn more.

Key Points

Join Chris and Fred as they discuss the relationship between reliability and statistics. The reality is that engineers and designers have been designing reliable things for thousands of year – with statistics a relatively recent phenomenon. So reliability doesn’t need statistics. But if you read a textbook or complete a reliability course it feels like statistics is everything. So what is going on?

Topics include:

Reliability happens at the point of decision. A decision is not a statistical analysis. That statistical analysis might create information that may then go onto to help inform a decision that creates reliability. But it is not the statistical analysis that makes reliability happen.
Statistics is easy to teach … sometimes. Decision-making is a reflection of culture. Which means the extent to which reliability ‘happens’ is also a reflection of culture. And culture is a challenging topic to broach in courses. But probability and statistics has textbooks and standards that can form the basis of a course. So we often have lots of reliability ‘luminaries’ simply teaching probability and statistics because they don’t have the ability to teach culture.
… so there is now a ‘statistics in reliability’ industry. There are professors, reliability consultants and reliability engineers who are perceived to be valuable because they appear to have mastered probability and statistics. So when we suggest that they need to focus on good design, manufacturing and maintenance decisions instead of endless statistical analysis, those who only bring a knowledge of statistics to the table panic as they are no longer relevant.

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.

SOR 632 Statistics and ReliabilityChristopher Jackson

Download Audio RSS

Show Notes

About Christopher Jackson

Chris is a reliability engineering teacher ... which means that after working with many organizations to make lasting cultural changes, he is now focusing no developing online, avatar-based courses that will hopefully make the 'complex' art of reliability engineering into a simple, understandable activity that you feel confident of doing (and understanding what you are doing).

Comments

DonMacArthur7 says

March 8, 2021 at 8:54 AM

Thanks Chris and Fred for another enlightening discussion.

Don

Reply
- Christopher Jackson says
  
  March 8, 2021 at 8:55 AM
  
  Thanks Don! Feedback like this helps us work out if we are on track or not.
  
  Reply
- Richard Brodzinski says
  
  March 8, 2021 at 6:08 PM
  
  Good discussion. The key is to appreciate the statistical analysis in reliability is only a tool and not necessarily the most important one. In some industries use of statistics in reliability is mandatory. Good source for this topic are the papers by is Paul Barringer.
  
  Reply
  - Christopher Jackson says
    
    March 9, 2021 at 3:38 AM
    
    Thanks Richard. I believe you are referring to papers like ‘Where Is My Data For Making Reliability Improvements?’ when referring to Paul Barringer? He makes some really good points, like that data is all around us, and we need to use it to make better decisions. Thanks for the feedback and sharing!
    
    Chris
    
    Reply