# US 10,120,912 B2 · System and method for combination-based data analysis

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

- Number: US 10,120,912 B2
- Document ID: US10120912B2
- Status: Granted
- Kind: B2
- Application number: 13/718,912
- Filed: 2012-12-18
- Published: 2013-05-02
- Granted: 2018-11-06
- Earliest claimed: 2008-10-14
- Assignees: Siemens Product Lifecycle Management Software Inc.
- Applicants: Camstar Systems, Inc.
- Inventors: Amin Shah-Hosseini
- Web page: https://patentgrep.com/patent/US10120912B2

## CPC

- G06F 16/2465: {Query processing support for facilitating data mining operations in structured databases}
- G06Q 10/06: Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- 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 and system for supply chain data analysis. The method includes storing supply chain data including test data, genealogy data, repair data, some factors and some items, in one or more databases and selecting a portion of the factors from the stored data, and a time range for analysis. The method then selects one or more criterion for analysis; extracts a portion of the stored data; analyzes the extracted portion of the stored data to detect a plurality of faulty combination of factors and items that results in an unexpected change in a key performance index, according to said extracted portion of the plurality of combinations. The method then performs correlation analysis on said plurality of faulty combinations to determine a root cause for the detected combination of factors; and generates a subset of said plurality of faulty combinations, according to said root causes of said plurality of faulty combinations.

## Claims

### Claim 1

A computer implemented method for supply chain data analysis, the method comprising:
storing supply chain data including test data, genealogy data, repair data, a plurality of factors and a plurality of items, in one or more databases;
integrating, by one or more computers, the stored supply chain data in a plurality of combinations, each combination including one or more items and five or more factors, wherein each factor comprises one of the group consisting of product, product category, product version, test, subtest, measurement, station, station type, operator, assembly line, site, building, software version, hardware version, component, join, board version, fail category, error message, component parent, component supplier, component child, fail code, defect code, repair type, component, location of a component on a product, part number, lot name, lot size, customer, site, product, call reason, operator, defect code, severity, and failed component;
receiving a minimum failure rate and sample size;
extracting, by the one or more computers, a portion of the plurality of combinations according to received minimum failure rate and sample size, by analyzing factor properties from the stored data, integrating a portion of the factor properties to find different combinations of factors, retrieving subtest data and matching the subset data with the different combinations, retrieving measurement data and matching the measurement data with the combinations, and determining test structure and test limits;
analyzing said extracted portion of the plurality of combinations, by the one or more computers, to detect a plurality of faulty combinations of factors and items that results in an unexpected change in a key performance index, according to said extracted portion of the plurality of combinations;
performing correlation analysis on said plurality of faulty combinations, by the one or more computers, to determine a root cause for each faulty combination;
generating, by the one or more computers, a subset of said plurality of faulty combinations, according to said root causes of said plurality of faulty combinations;
generating a root cause chart, by the one or mare computers and according to the generated subset of said plurality of faulty combinations, that illustrates at least one parameter value that is a cause of the unexpected change in a key performance index; and
displaying or storing the root cause chart by the one or more computers.

### Claim 2 (depends on 1)

The method of claim 1, further comprising
performing trend analysis on said subset of said plurality of faulty combinations; and
making recommendations for promising combinations and range of test parameters to maximize a yield for selected factors.

### Claim 3 (depends on 1)

The method of claim 1, further comprising performing an entropy-based function that shows the variations between items for each of a plurality of selected factors.

### Claim 4 (depends on 1)

The method of claim 1, further comprising allocating resources including one or more of stations, fixtures, and operators to maximize resource utilization.

### Claim 5 (depends on 1)

The method of claim 1, wherein said detecting a root cause comprises breaking down a test measurement based on related factors and items in said plurality of faulty combinations.

### Claim 6 (depends on 1)

The method of claim 1, further comprising creating production plans based on a required number of units to be tested in a period of time, a test time, and a failure rate.

### Claim 7 (depends on 1)

The method of claim 1, further comprising extracting test data, return and maintenance data, and supplier data, from the stored data.

### Claim 8 (depends on 1)

The method of claim 1, further comprising extracting data insights from the stored data collected from test, return and repair, and supply chain processes.

### Claim 9 (depends on 1)

The method of claim 1, further comprising performing what-if analysis on a plurality of selected factors according to said subset of said plurality of faulty combinations.

### Claim 10 (depends on 1)

The method of claim 1, further comprising allocating resources including one or more of stations, fixtures, and operators according to one or more desired functions selected by the user including one or more of minimize cost, minimize scrap, and minimize time-in-process, and outputting best one or more combinations that meet the selected desired functions.

### Claim 11 (depends on 1)

The method of claim 1, further comprising accepting a threshold quantity and a threshold percentage for a group of data points that share same values for one or more of a plurality of selected factors, and displaying a bubble chart graph depicting a plurality of indicators with increasing sizes to illustrate more data in said group of data points.

### Claim 12 (depends on 1)

The method of claim 1, wherein said key performance index includes one or more of test failure, product return, part repair, return rate and genealogy factor.

### Claim 13

A system for supply chain data analysis comprising:
one or more databases for storing supply chain data including test data, genealogy data, repair data, a plurality of factors and a plurality of items;
an input device for selecting a portion of the plurality of factors from the stored data, and a time range for analysis, and selecting one or more criteria for analysis;
an extract & calculate module electrically coupled to the one or more databases for aggregating the stored supply chain data in a plurality of combinations, each combination including one or more items and five or more factors and extracting a portion of the plurality of combinations according to minimum failure rate and sample size by analyzing factor properties from the stored data, integrating a portion of the factor properties to find different combinations of factors, retrieving subtest data and matching the subset data with the different combinations, retrieving measurement data and matching the measurement data with the combinations, and determining test structure and test limits, wherein each factor comprises on of the group consisting of product, product category, product version, test, subtest measurement, station, station type, operator, assembly line, site, building, software version, hardware version, component, join, board version, fail category, error message, component parent, component supplier, component child, fail code, defect code, repair type, component, location of a component on a product, part number, lot name, lot size, customer, site product, call reason, operator, defect code, severity, and failed component; and
a multi-dimensional failure analysis module for analyzing said extracted portion of the plurality of combinations to detect a plurality of a faulty combinations of factors and items that results in an unexpected change in a key performance index, according to said extracted portion of the plurality of combinations, performing correlation analysis on said plurality of faulty combinations to determine a root cause for each faulty combination; and generating a subset of said plurality of faulty combinations, according to said root causes of said plurality of faulty combinations.

### Claim 14 (depends on 13)

The system of claim 13, wherein the multi-dimensional failure analysis module is configured to perform an entropy-based function that shows the variations between items for each selected factor.

### Claim 15 (depends on 13)

The system of claim 13, further comprising a test configuration analysis module for finding promising configurations and range of test parameters to maximize a yield.

### Claim 16 (depends on 15)

The system of claim 15, wherein the test configuration analysis module is configured to create production plans based on a required number of units to be tested in a period of time, test time, and failure rate.

### Claim 17 (depends on 13)

The system of claim 13, wherein the extract & calculate module is configured to extract data insights from the stored data collected from test, return and repair, and supply chain processes.

### Claim 18 (depends on 13)

The system of claim 13, further comprising a production planning module for allocating resources including one or more of stations, fixtures, and operators according to one or more desired functions selected by the user including one or more of minimize cost, minimize scrap, and minimize time-in-process, and outputting best one or more combinations that meet the selected desired functions.

### Claim 19 (depends on 13)

The system of claim 13, wherein said key performance index includes one or more of test failure, product return, part repair, return rate and genealogy factor.