# US 8,275,575 B2 · Determining component failure rates using accelerated life data

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 8,275,575 B2
- Document ID: US8275575B2
- Status: Granted
- Kind: B2
- Application number: 12/408,401
- Filed: 2009-03-20
- Published: 2010-09-23
- Granted: 2012-09-25
- Earliest claimed: 2009-03-20
- Assignees: International Business Machines Corporation
- Inventors: Amanda E. Mikhail; Mark D. Plucinski
- Web page: https://patentgrep.com/patent/US8275575B2

## CPC

- G06Q 10/04: Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem" (market predictions or forecasting for commercial activities G06Q30/0202)
- G06Q: INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR

## Abstract

A method of statistical analysis is based on an accelerated degradation model for estimating the failure rate from a set of accelerated life data (e.g. stress, humidity, temperature, voltage, resistance, vibration, etc.). The method re-samples randomly distributed data and organizes it into subsets that may be ordered. The maximum is found for each subset. From the maximums a parametric LEV distribution is determined. Maximum Likelihood Estimation methods may be used to find the parameters employed in the determined LEV distribution. The failure rate is calculated using the determined LEV distribution of the sample maximums.

## Claims

### Claim 1

A method for selecting a particular component from a plurality of possible components, wherein the particular component is utilized in an electronic system, the method comprising:
performing, with a processor, accelerated life tests on the plurality of possible components resulting in accelerated life data for each possible component;
determining, with a processor, the type of statistical distribution of each accelerated life data set, and if a distribution is an unbounded distribution;
determining, with a processor, a failure rate by re-sampling the unbounded distributed accelerated life data; and
providing the particular component for use in the electronic system if the failure rate is below a threshold failure rate;
wherein re-sampling the unbounded distributed accelerated life data further comprises:
subdividing, with a processor, the accelerated life data into a plurality of sub datasets;
selecting, with a processor, the maximum accelerated life data point from each sub dataset, wherein the selected maximums form a maximum dataset; and
fitting, with a processor, the maximum dataset to a Largest Extreme Value distribution.

### Claim 2 (depends on 1)

The method of claim 1 wherein the Largest Extreme Value distribution as a function of “x” equals exp(−exp −(x−a)/b), and wherein “a” and “b” are parameters estimated using maximum likelihood estimates procedures.

### Claim 3 (depends on 1)

The method of claim 1 wherein re-sampling the accelerated life data corresponding to each component further comprises:
determining, with a processor, the probability that a sample from the maximum data set is greater than a predetermined critical value.

### Claim 4 (depends on 3)

The method of claim 3 wherein re-sampling the accelerated life data corresponding to each component further comprises:
adjusting, with a processor, the probability by dividing the probability by the quantity of further sub datasets;
calculating the failure rate with a processor;
adjusting for bias, with a processor; and
determining a confidence interval with a processor.

### Claim 5

A computer implemented method for determining failure rate of a component from accelerated life data, wherein a computer is configured for implementing the computer implemented method comprising:
determining, with a computer, the type of statistical distribution of an accelerated life data set, and if the distribution is an unbounded distribution; and
determining, with a computer, the failure rate of the component by re-sampling the accelerated life data of the component;
wherein re-sampling the accelerated life data of the component further comprises:
subdividing, with a computer, the accelerated life data into a plurality of sub datasets;
selecting, with a computer, the maximum accelerated life data point from each sub dataset, wherein the selected maximums form a maximum dataset; and
fitting, with a computer, the maximum dataset to a Largest Extreme Value distribution.

### Claim 6 (depends on 5)

The computer implemented method of claim 5 wherein the Largest Extreme Value distribution as a function of “x” equals exp(−exp −(x−a)/b), and wherein “a” and “b” are parameters estimated using maximum likelihood estimates procedures.

### Claim 7 (depends on 5)

The computer implemented method of claim 5 wherein re-sampling the accelerated life data of the component further comprises:
determining, with a computer, the probability that a sample from the maximum data set is greater than a predetermined critical value.

### Claim 8 (depends on 7)

The computer implemented method of claim 7 wherein re-sampling the accelerated life data of the component further comprises:
adjusting, with a computer, the probability by dividing the probability by the quantity of further sub datasets;
calculating the failure rate with a computer; and
adjusting for bias with a computer.

### Claim 9

A computer program product comprising a non-transitory computer usable medium having computer program logic recorded thereon for enabling a processor to determine the failure rate of a component, wherein the computer program logic when executed by the processor causes a computer to:
determine the type of statistical distribution of an accelerated life data set, and if the distribution is an unbounded distribution; and
determine the failure rate of the component by re-sampling the accelerated life data of the component;
wherein upon the computer program logic for re-sampling the accelerated life data of the component is executed by the processor further causes a computer to:
subdivide the accelerated life data into a plurality of sub datasets;
select the maximum accelerated life data point from each sub dataset, wherein the selected maximums form a maximum dataset; and
fit the maximum dataset to a Largest Extreme Value distribution.

### Claim 10 (depends on 9)

The computer program product of claim 9 wherein re-sampling the accelerated life data of the component further comprises:
determining the probability that a sample from the maximum data set is greater than a predetermined critical value.

### Claim 11 (depends on 10)

The computer program product of claim 10 wherein re-sampling the accelerated life data of the component further comprises:
adjusting the probability by dividing the probability by the quantity of further sub datasets;
calculating the failure rate; and
adjusting for bias.