Reliability and Tolerance Analysis
A tolerance specification is how we communicate the allowance for part variation.
Variation happens, and when it is within what we expect, great. Otherwise, quality and reliability issues ensue. Let’s explore how to set tolerances to improve reliability properly.
Let’s understand the process data, stability, and design constraints in an attempt to find a balance to create meaningful tolerances.
[NOTE: Check out the ebook Statistical Tolerance Analysis]
Worst case isn’t always feasible, yet it is very common.
Consider using the root mean squared or the Monte Carlo approach. The worst-case approach relies on little data, just a range, whereas RMS assumes a normal distribution of the variation.
The Monte Carlo approach may require the most information, and it provides the best results.
A few quick overviews and examples, plus plenty of time to discuss the role of the reliability professional in the tolerance seeing process.
Join the conversation during the webinar.
This Accendo Reliability webinar originally broadcast live on 16 June 2015.
To view the recorded webinar and slides, visit the webinar page.
Related Content
Confidence and Tolerance Intervals episode
Design Tolerances Based on Economics (Using the Taguchi Loss Function) episode
Estimating the Normal Distribution Parameters and Tolerance Limits article
Inputs to Consider When Setting Tolerances article
Please login with your site registration to view the video recording and download the ebook.
If you haven’t registered, it’s free and takes only a moment.
This quick introduction to three statistical analysis methods lets you determine or assess part tolerances quickly.
You will also learn why tolerances are critical to achieving a reliable product or system.
The ebook includes guidelines for data analysis, starting with histograms and building probability density functions for use in two of the three methods.
Leave a Reply