Automated data-driven closed-loop-feedback method for adaptive and comparative quality control in a discrete data aggregation environment

• US 9,286,585 B2
• Filed: 10/26/2012
• Issued: 03/15/2016
• Est. Priority Date: 10/26/2012
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: DiMarco, Robert A, Xue, Feng, Giancaspro, Vincent J, Whitney, Michael J, Yoon, Jung H
• Primary Examiner(s): Rodriguez, Carlos Ortiz
• Assistant Examiner(s): Norton, Jennifer L
• Application Number:
  US13/661,400
• Publication Number:
  US 20140121805A1
• Time in Patent Office: 1,236 Days
• Field of Search: 700/109, 700/110, 705/7.41
• US Class Current: 1/1
• CPC Class Codes: G06Q 10/00 Administration; Management
• CPC Class Codes: G06Q 10/06 Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
• CPC Class Codes: G06Q 10/06395 Quality analysis or management
• CPC Class Codes: G06Q 30/00 Commerce

Abstract

A method for determining quality includes receiving supplier quality data from a supplier. The supplier quality data includes discrete quality data. The supplier quality data is for a quantity of a component supplied to a company or for a service provided on behalf of the company over a supplier measurement period. The method includes aggregating quality data from two or more suppliers. The quality data is for the supplied component or provided service over a baseline time period. The method includes determining a quality benchmark using the aggregated quality data and setting one or more alarm levels. The alarm levels are based on the quality benchmark of the aggregated quality data. The method includes determining supplier quality level for the supplier for the supplier measurement time period. The supplier quality is based on the supplier quality data. The method includes determining if the supplier quality data exceeds an alarm level.

Claims

1. A method comprising:

receiving supplier quality data, the supplier quality data comprising discrete quality data, the supplier quality data for one of

a quantity of a component supplied to a company over a supplier measurement period; and

a service provided on behalf of the company over a supplier measurement period, wherein the service is provided to one or more of

the company; and

a customer;

wherein the supplier quality data comprises information regarding quality of the component supplied to the company or the service provided on behalf of the company;

aggregating quality data received from two or more suppliers, the quality data for the component supplied to the company or service provided for the company over a baseline time period, the quality data comprising discrete quality data, wherein aggregating the quality data comprises one of

summing component failures of two or more suppliers over a baseline time period and summing a quantity of components tested by the two or more suppliers over the baseline time period, and

summing averages scores for a service provided by the two or more suppliers over the baseline time period and dividing the sum by the number of suppliers;

determining a quality benchmark using the aggregated quality data, wherein the quality benchmark comprises one of

a ratio of a quantity of component failures to a quantity of components tested within the baseline time period for the aggregated quality data; and

an average score of the service within the baseline time period, the average score based on the aggregate quality data;

setting one or more alarm levels, by multiplying the quality benchmark of the aggregated quality data by one or more alarm multipliers, wherein each alarm multiplier is associated with an alarm level, wherein each alarm level is associated with a different level of severity of a quality issue, wherein the alarm multiplier is based on the level of severity of the quality issue, the level of severity being based on how close the alarm level is to the quality benchmark;

determining a supplier quality level for the supplier measurement time period, the supplier quality level based on the supplier quality data;

determining if the supplier quality level exceeds an alarm level; and

initiating an alarm response in response to determining that the supplier quality level exceeds an alarm level, each alarm response appropriate for a level of severity of the alarm level exceeded by the supplier quality level.

2. The method of claim 1, wherein the aggregated quality data and the supplier quality data comprise data for a component and comprise a number of components tested and a number of components within the tested components failing a quality test, the quality data correlated to a time period relevant to creation of the components.

3. The method of claim 1, wherein the aggregated quality data and the supplier quality data comprise data for a service provided by a supplier and comprise a quality score for the provided service, wherein the quality score for service provided by each supplier is based on a same scoring system.

4. The method of claim 1, wherein the supplier quality data and the aggregated quality data comprise data for a group, the group comprising a subclass of a class comprising the component or service, and wherein the quality benchmark, the one or more alarm levels, and the one or more supplier alarm levels are based on quality data for the group.

5. The method of claim 1, wherein determining a supplier quality level comprises determining a ratio of a quantity of component failures to a quantity of components tested, the quantity of component failures and the quantity of components tested within in the supplier quality data for the supplier measurement period.

6. The method of claim 1, wherein the one or more supplier alarms comprise two or more alarms set at differing levels from the quality benchmark and wherein the alarm response module initiates alarm responses of differing severity for each of the two or more alarms, wherein an alarm response of higher severity is associated with an alarm level further away from the quality benchmark and an alarm response of lower severity is associated with an alarm level closer to the quality benchmark than the alarm level with a higher severity alarm response.

7. The method of claim 1, further comprising:

adding supplier quality data to the aggregated quality data to get updated aggregated quality data;

determining an updated quality benchmark using the updated aggregated quality data; and

setting one or more alarm levels based on the updated quality benchmark.

8. An apparatus comprising:

a receiver module that receives supplier quality data, the supplier quality data comprising discrete quality data, the supplier quality data for one of

a quantity of a component supplied to a company over a supplier measurement period; and

a service provided on behalf of the company over a supplier measurement period, wherein the service is provided to one or more of

the company; and

a customer;

wherein the supplier quality data comprises information regarding quality of the component supplied to the company or the service provided on behalf of the company;

an aggregation module that aggregates quality data received from two or more suppliers, the quality data for the component supplied to the company or service provided for the company over a baseline time period, the quality data comprising discrete quality data, wherein aggregating the quality data comprises one of

summing component failures of two or more suppliers over a baseline time period and summing a quantity of components tested by the two or more suppliers over the baseline time period, and

summing averages scores for a service provided by the two or more suppliers over the baseline time period and dividing the sum by the number of suppliers;

a baseline module that calculates a quality benchmark using the aggregated quality data from the aggregation module, wherein the quality benchmark comprises one of

a ratio of a quantity of component failures to a quantity of components tested within the baseline time period for the aggregated quality data; and

an average score of service provided by the suppliers within the baseline time period, the average score based on the aggregate quality data;

an alarm level module that sets one or more alarm levels by multiplying the quality benchmark of the aggregated quality data by one or more alarm multipliers, wherein each alarm multiplier is associated with an alarm level, wherein each alarm level is associated with a different level of severity of a quality issue, wherein the alarm multiplier is based on the level of severity of the quality issue, the level of severity being based on how close the alarm level is to the quality benchmark;

a supplier quality module that determines a supplier quality level for the supplier measurement time period, the supplier quality level based on the supplier quality data;

an alarm module that determines if the supplier quality level exceeds an alarm level; and

an alarm response module that initiates an alarm response in response to the alarm module determining that the supplier quality level exceeds an alarm level, each alarm response appropriate for a level of severity of the alarm level exceeded by the supplier quality level.

9. The apparatus of claim 8, wherein the aggregated quality data and the supplier quality data comprise one of

data for a component and comprise a number of components tested and a number of components within the tested components failing a quality test, the quality data correlated to a time period;

and

data for a service provided by a supplier and comprise a quality score for the provided service, wherein the quality score for service provided by each supplier is based on a same scoring system.

10. The apparatus of claim 8, wherein the alarm response module initiates alarm responses of differing severity for each of two or more alarms wherein an alarm response of higher severity is associated with an alarm level further away from the quality benchmark and an alarm response of lower severity is associated with an alarm level closer to the quality benchmark than the alarm level with a higher severity alarm response.

11. A computer program product for determining supplier quality, the computer program product comprising a non-transitory computer readable storage medium having computer readable program code embodied therein, the computer readable program code configured to:

receive supplier quality data, the supplier quality data comprising discrete quality data, the supplier quality data for one of

a quantity of a component supplied to a company over a supplier measurement period; and

a service provided on behalf of the company over a supplier measurement period, wherein the service is provided to one or more of

the company; and

a customer;

wherein the supplier quality data comprises information regarding quality of the component supplied to the company or the service provided on behalf of the company;

aggregate quality data received from two or more suppliers, the quality data for the component supplied to the company or service provided for the company over a baseline time period, the quality data comprising discrete quality data, wherein aggregating the quality data comprises one of

summing component failures of two or more suppliers over a baseline time period and summing a quantity of components tested by the two or more suppliers over the baseline time period, and

summing averages scores for a service provided by the two or more suppliers over the baseline time period and dividing the sum by the number of suppliers;

calculate a quality benchmark using the aggregated quality data, wherein the quality benchmark comprises one of

a ratio of a quantity of component failures to a quantity of components tested within the baseline time period for the aggregated quality data; and

an average score of service provided by the suppliers within the baseline time period, the average score based on the aggregate quality data;

set one or more alarm levels, the alarm levels by multiplying the quality benchmark of the aggregated quality data by one or more alarm multipliers, wherein each alarm multiplier is associated with an alarm level, wherein each alarm level is associated with a different level of severity of a quality issue, wherein the alarm multiplier is based on the level of severity of the quality issue, the level of severity being based on how close the alarm level is to the quality benchmark;

determine a supplier quality level for the supplier measurement time period, the supplier quality level based on the supplier quality data;

determine if the supplier quality level exceeds an alarm level; and

initiate an alarm response in response to determining that the supplier quality level exceeds an alarm level, each alarm response appropriate for a level of severity of the alarm level exceeded by the supplier quality level.

12. A method comprising:

receiving supplier quality data, the supplier quality data comprising discrete quality data, the supplier quality data for a quantity of a component supplied to a company over a supplier measurement period, wherein the service is provided to one or more of

the company; and

a customer

wherein the supplier quality data comprises information regarding the quality of the component supplied to the company or the service provided on behalf of the company;

aggregating quality data received from two or more suppliers by summing component failures of two or more suppliers over a baseline time period and summing a quantity of components tested by the two or more suppliers over the baseline time period, the quality data for the component supplied to the company, the quality data comprising discrete quality data;

determining a quality benchmark using the aggregated quality data, the quality benchmark comprising a ratio of a quantity of component failures to a quantity of components tested within the baseline time period for the aggregated quality data;

setting one or more alarm levels by multiplying the quality benchmark by one or more alarm multipliers, wherein each alarm multiplier is associated with an alarm level, wherein each alarm level is associated with a different level of severity of a quality issue, wherein the alarm multiplier is based on the level of severity of the quality issue, the level of severity being based on how close the alarm level is to the quality benchmark;

determining a supplier quality level for the supplier measurement time period by determining a ratio of a quantity of component failures to a quantity of components tested, the quantity of component failures and the quantity of components tested within in the supplier quality data for the supplier measurement period;

determining if the supplier quality level exceeds an alarm level; and

initiating an alarm response in response to determining that the supplier quality level exceeds an alarm level, each alarm response appropriate for a level of severity of the alarm level exceeded by the supplier quality level.