Dismiss
Surety is performing system maintenance this weekend. Electronic date stamps on new Prior Art Database disclosures may be delayed.
Browse Prior Art Database

# Method for Time Dependent Failure Rates Analysis

IP.com Disclosure Number: IPCOM000078832D
Original Publication Date: 1973-Mar-01
Included in the Prior Art Database: 2005-Feb-26
Document File: 2 page(s) / 66K

IBM

## Related People

Lewis, SH: AUTHOR [+2]

## Abstract

This is a method for predicting the reliability of electronic assemblies with components, whose failure rates vary with power-on usage and with component vintage. The method builds the failure rate of the assembly by combining the input data through a Monte Carlo simulation process.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 54% of the total text.

Page 1 of 2

Method for Time Dependent Failure Rates Analysis

This is a method for predicting the reliability of electronic assemblies with components, whose failure rates vary with power-on usage and with component vintage. The method builds the failure rate of the assembly by combining the input data through a Monte Carlo simulation process.

The input data consists of the component count on the assembly, expected use of the assembly, applicable costs and quality data,and the component failure rates. A curve fit routine, shown in the drawing, consists of extending and combining all the failure rates for the components on the assembly, to give a time-dependent failure rate step function for the assembly for each assembly vintage. Each step function is then transformed to a log scale and fitted to a straight line via the least squares technique. Transformation back to the rectangular scale gives an equation of the form: y = At/B/. This result can be expressed in the Weibull failure-rate function f(t) = proportional to Bt/B-1/, by letting proportional to B=A B-1=B.

The distribution of failure times,f(t) is obtained by integrating the failure-rate function Z(t), in the expression:

(Image Omitted)

These cells are initialized to zero prior to each iteration of simulation. An iteration is defined as a complete path through the entire inventory of the assembly under consideration.

The simulation process consists of installing the machine, simulating its use via generation of successive failures from the proper vintage fai...