METHOD AND SYSTEM FOR OBTAINING A COMBINATION OF FAULTY PARTS FROM A DISPERSED PARTS TREE

• US 20090271354A1
• Filed: 06/22/2009
• Published: 10/29/2009
• Est. Priority Date: 10/05/2006
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Yuriyama, Madoka, Numao, Masayuki, Kitayama, Fumihiko, Matsuzawa, Hirofumi
• Granted Patent: US 7,769,706 B2
• Time in Patent Office: Days
• US Class Current: 706/47
• CPC Class Codes: G06N 5/025 Extracting rules from data
• CPC Class Codes: G06Q 10/06 Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
• CPC Class Codes: Y10S 706/902 Application using ai with detail of the ai system

Abstract

It is an object of the present invention to find out parts to be a highly possible cause of failure without searching all of part data of all of products. Dispersed parts data on a parts tree are sequentially accessed from a set of known failed products, and part attribute values each having a higher support in the faulty product are extracted. In this process, a subset of parts used in the faulty product is also obtained simultaneously. The part attribute values having higher supports and the subset of parts used in the faulty product are represented as a tree in which a parts type serves as a node. Next, an information gain of a rule that having the two part attribute values is a cause of failure is calculated on two part attribute values having higher supports on the tree of the parts type. This calculation is locally performed on a common parent part of two parts and parts having a certain information gain is outputted as a cause of failure. How to select these two part attributes is performed in such a way that part attributes located closer to each other on the tree are first evaluated, and first found part attributes are made a candidate of a cause of failure.

Claims

1. An information system of obtaining a combination of part attribute values as a candidate of a cause of failure from data of a bill of materials holding ID of parts and attribute values of parts by a computer, comprising:

a generation unit of the degree of support for searching the bill of materials by the ID of parts used for failure products;

acquiring the attribute values of parts of the searched parts;

comparing between the acquired attribute values of parts from each other; and

generating the degree of support to be counted in the case of the same values;

a part attribute value extracting unit for storing parts having attribute values of parts, whose generated degree of support is higher than a predetermined value and the attribute values thereof as frequently-appearing part attribute;

a support part combining unit for preparing a combination of parts having the stored frequently-appearing part attribute;

a common part unit for obtaining a common part where each part relating to the prepared combination is commonly set, regarding the bill of materials;

a failure entropy among common parts for calculating a failure entropy among common parts, which is an entropy of parts used for the failed products among the entire obtained common parts;

a failure entropy among support parts for calculating a failure entropy among support parts, which is an entropy of parts used for the failed parts in the common part using the parts having the frequently-appearing part attribute;

an information gain calculating unit for calculating an information gain, which is a difference between the failure entropy among the common product and the failure entropy among the support parts; and

a selecting unit for selecting a combination of parts having the frequently-appearing part attribute in the case that the calculated information gain is greater than a predetermined threshold value.

2. The information system according to claim 1, wherein

the part attribute value extracting unit combines a plurality of attribute values, and regards as one value and stores as the frequently appearing part attribute.

3. The information system according to claim 1, wherein

in the common part unit,

a common part where the prepared combination is commonly set is obtained from a bill of materials, where a bill of materials having a tree structure of data structure where products and parts comprising said products are hierarchically linked are dispersed and stored based upon network address per node.

4. The information system according to claim 3, wherein

in the common part unit,

one part relating to the prepared combination situated at one child node in the part database and data of parent part situated at a common parent node with another part relating to the prepared combination, situated at another child node are obtained based upon the network address.

5. The information system according to claim 1, wherein

in the selecting unit, the threshold value is increased little by little and the prepared combination is narrowed down.

6. A computer program product of obtaining a combination of part attribute values as a candidate of a cause of failure from data of a bill of materials holding ID of parts and attribute values of parts by a computer, causing a computer to execute the steps of:

searching the bill of materials by the ID of parts used for failure products;

acquiring the attribute values of parts of the searched parts;

comparing between the acquired attribute values of parts from each other; and

generating the degree of support to be counted in the case of the same values;

storing parts having attribute values of parts, whose generated degree of support is higher than a predetermined value and the attribute values thereof as frequently-appearing part attribute;

preparing a combination of parts having the stored frequently-appearing part attribute;

common parts for obtaining a common part where each part relating to the prepared combination is commonly set, regarding the bill of materials;

calculating a failure entropy among common parts, which is an entropy of parts used for the failed products among the entire obtained common parts;

calculating a failure entropy among support parts, which is an entropy of parts used for the failed parts in the common part using the parts having the frequently-appearing part attribute;

calculating an information gain, which is a difference between the failure entropy among the common product and the failure entropy among the support parts; and

selecting a combination of parts having the frequently-appearing part attribute in the case that the calculated information gain is greater than a predetermined threshold value.

7. The computer program product according to claim 6, wherein

for the part attribute values, a plurality of attribute values of parts are combined and regarded as one value.

8. The computer program product according to claim 6, wherein

in the step of the common parts,

a common part where the prepared combination is commonly set is obtained from a bill of materials, where a bill of materials having a tree structure of data structure where products and parts comprising said products are hierarchically linked are dispersed and stored based upon network address per node.

9. The computer program product according to claim 8, wherein

in the step of the common parts,

one part relating to the prepared combination situated at one child node in the part database and data of parent part situated at a common parent node with another part relating to the prepared combination, situated at another child node are obtained based upon the network address.

10. The computer program product according to claim 6, wherein

in the step of selecting, the threshold value is increased little by little and the prepared combination is narrowed down.