Accendo Reliability

Your Reliability Engineering Professional Development Site

  • About
    • Adam Bahret
    • Alex Williams
    • Andre Kleyner
    • Anne Meixner
    • Arthur Hart
    • Ash Norton
    • Carl Carlson
    • Chris Jackson
    • Chris Stapelmann
    • Dennis Craggs
    • Dev Raheja
    • Doug Lehr
    • Doug Plucknette
    • Fred Schenkelberg
    • George Williams
    • Gina Tabasso
    • Greg Hutchins
    • James Kovacevic
    • James Reyes-Picknell
    • Joe Anderson
    • John Paschkewitz
    • Katie Switzer
    • Kevin Stewart
    • Kirk Gray
    • Les Warrington
    • Mike Konrad
    • Mike Sondalini
    • Nancy Regan
    • Perry Parendo
    • Philip Sage
    • Ray Harkins
    • Rob Allen
    • Robert (Bob) J. Latino
    • Robert Kalwarowsky
    • Ryan Chan
    • Shane Turcott
    • Steven Wachs
    • Tim Rodgers
    • Usman Mustafa Syed
  • Reliability.fm
    • Dare to Know
    • Speaking Of Reliability
    • Rooted in Reliability: The Plant Performance Podcast
    • Maintenance Disrupted
    • Practical Reliability Podcast
    • Reliability Matters
    • Masterminds in Maintenance
    • Accendo Reliability Webinar Series
    • Asset Reliability @ Work
  • Articles
    • CRE Preparation Notes
    • on Leadership & Career
      • Advanced Engineering Culture
      • Engineering Leadership
      • Managing in the 2000s
      • Product Development and Process Improvement
    • on Maintenance Reliability
      • Aasan Asset Management
      • CMMS and Reliability
      • Conscious Asset
      • EAM & CMMS
      • Everyday RCM
      • Maintenance and Reliability
      • Plant Maintenance
      • ReliabilityXperience
      • RCM Blitz®
      • Rob’s Reliability Project
      • The Intelligent Transformer Blog
      • The RCA
    • on Product Reliability
      • Accelerated Reliability
      • Achieving the Benefits of Reliability
      • Apex Ridge
      • Musings on Reliability and Maintenance Topics
      • Reliability Engineering Insights
      • Reliability in Emerging Technology
    • on Risk & Safety
      • CERM® Risk Insights
      • Equipment Risk and Reliability in Downhole Applications
    • on Tools & Techniques
      • Big Data & Analytics
      • Experimental Design for NPD
      • Innovative Thinking in Reliability and Durability
      • Inside FMEA
      • Integral Concepts
      • Testing 1 2 3
      • The Manufacturing Academy
      • Reliability Reflections
  • eBooks
    • Reliability Engineering Management DRAFT
  • Resources
    • Accendo Authors
    • FMEA Resources
    • Feed Forward Publications
    • Openings
    • Books
    • Webinars
    • Journals
    • Higher Education
    • Podcasts
  • Groups
    • Reliability Integration
    • Mastermind
    • Study Groups
  • Courses
    • 14 Ways to Acquire Reliability Engineering Knowledge
    • Reliability Analysis Methods online course
    • SPC-Process Capability Course
    • Reliability Centered Maintenance (RCM) Online Course
    • Measurement System Assessment
    • Process Capability Analysis course
    • Root Cause Analysis and the 8D Corrective Action Process course
    • Return on Investment online course
    • 5-day Reliability Green Belt ® Live Course
    • 5-day Reliability Black Belt ® Live Course
    • CRE Preparation Online Course
  • Webinars
    • Upcoming Live Events
  • Calendar
    • Call for Papers Listing
    • Upcoming Webinars
    • Webinar Calendar
  • Login
    • Member Home
Don’t show this message again.

Cookies

This site uses cookies to give you a better experience, analyze site traffic, and gain insight to products or offers that may interest you. By continuing, you consent to the use of cookies. Learn how we use cookies, how they work, and how to set your browser preferences by reading our Cookies Policy.

by Fred Schenkelberg 5 Comments

The Four Functions

Four Functions

There are four functions related to life distributions of importance to reliability engineering.

Nearly all textbooks on reliability either introduce or use these functions. Likewise, nearly every calculation related to reliability statistics also uses at least one of functions.

Suppose we produced 100 units of a product and tracked them all till failure over time.

Eventually, all of them would fail. And, unlike the Parson’s One Hoss Shay the products and their components would not all fail at the same time.

One way to track the failures over time is to use a simple histogram.

By plotting the number of failures each month, for example, we would create a representation of the distribution of time to failures.

The histogram represents a Probability Density Function (PDF), f(t). Where the function is always positive and has a total area of one.

Mathematically for continuous distributions

And, for discrete distributions, we have a similar relationship using summation.

In this short note, we will only discuss the continuous distribution case.

Based on probability theory ( the subject of another discussion), we know that at the initial point none of the 100 units have failed. And, we know that eventually all of the units will fail.

Thus over any particular interval of interest, time, in this case, we can determine the probability of failure

The mean and variance terms can also be determined at this point.

Where E is the expected value and T is a random variable (more on this in another posting) and knowing the first moment of the random variable about the origin is defined as the distributions mean, μ. For a continuous distribution

And, the variance, σ, is the second moment of the expected value. For continuous distributions

Of course, the positive square root of the variance is the standard deviation.

f4A common question we face is how many will fail at some point in time. If we produce 100 units of product and ship them to the customer, how many will fail by the end of the year?

The Cumulative Distribution Function (CDF) F(t) for a continuous distribution describes the fraction failing by age t.

The CDF describes the probability that an event, a failure, occurs at some point in time.

F(t) ranges from zero to one and is monotonic, meaning as time, t, increases the probability steadily increases.

And, as you may suspect, the CDF is related to the PDF, as the CDF quantifies the area under the curve (the PDF) over time.

Note the upper limit is the end point of the period of interest, and the start is, well, the start, or when everything is considering functioning or working.

Using the histogram above we can step through each instance and calculate the probability of failure to produce the CDF plot.

And, from the plot, we can estimate the probability of failure at one year as about 60% failed by 8760 hours.

The CDF looks at the probability of failures, yet sometimes we are interested in how many will still be operating after some period of time.

Considering the overall probability ranges from zero to one, and a little probability theory, we can define the Reliability function as the complement of the CDF.

As you can see, the R(t) is similar to F(t), only R(t) starts with all the units working descends to none working.

The fourth function, the Hazard Rate, h(t), or Failure Rate, is the rate units, which have survived to a point in time, will fail.

Think conditional probability. If 90 units are still working at time, t, then what is the expected probability of failure for the units now?

The failure rate can and often does change over time. The familiar ‘bathtub curve’ shows the hazard rate decreasing initially in the ‘infant mortality’ phase, then steady, and finally increasing during the wear-out phase.

In practice, it is rare that the failure rate remains constant for a significant period of time.

The hazard rate is the limit of the instantaneous failure rate given no failures up to time t, or

Therefore, if you know one function one can determine the other three functions.

Most of the time we work with these functions in relation to a particular distribution.

Over the next couple of posts, we will explore the commonly used distributions and their related functions.

In the meantime, work out how to derive all functions when starting with one in particular.

The refresher on algebra and calculus will be good for you.


Related:

Reliability Function (article)

Using The Exponential Distribution Reliability Function (article)

Exponential Reliability (article)

 


Also published on Medium.

Filed Under: CRE Preparation Notes, Probability and Statistics for Reliability Tagged With: Discrete and continuous probability distributions

« Basic Statistics
Reliability Function »

Comments

  1. Giovanni Vallone says

    December 14, 2012 at 1:13 AM

    Fred, maybe due to a mistyping, the formula f4 shall be corrected adding the dt after the function t^2 f(t) .

    Reply
    • Fred Schenkelberg says

      December 15, 2012 at 2:35 PM

      Thanks for the note – I’m working on editing the formula and updating.

      cheers,

      Fred

      Reply

Leave a Reply Cancel reply

Your email address will not be published. Required fields are marked *

CRE Preparation Notes

Article by Fred Schenkelberg

Join Accendo

Join our members-only community for full access to exclusive eBooks, webinars, training, and more. It’s free and only takes a minute.

Get Full Site Access

Not ready to join?
Stay current on new articles, podcasts, webinars, courses and more added to the Accendo Reliability website each week.
No membership required to subscribe.

Get Weekly Email Updates


Subscribe to Weekly Updates

Get updates on the latest content added to the site, including: articles, podcasts, webinars, live events and assorted other reliability engineering professional development materials.






We care about your privacy and will not share, leak, loan or sell your personal information. View our privacy policy.


  • CRE Preparation Notes
  • CRE Prep
  • Reliability Management
  • Probability and Statistics for Reliability
  • Reliability in Design and Development
  • Reliability Modeling and Predictions
  • Reliability Testing
  • Maintainability and Availability
  • Data Collection and Use

© 2021 FMS Reliability · Privacy Policy · Terms of Service · Cookies Policy