Use of Associate Memory Learning Agent Technology to Identify Interchangeable Parts in Parts Catalogs

Abstract

The present invention provides a computer implemented method, apparatus, and computer usable program code for identifying interchangeable parts in parts catalogs. In one illustrative embodiment, a set of catalogs is searched for first attributes for a set of parts using an intelligent agent. The intelligent agent compares the first attributes for the set of parts to a set of second attributes for a selected part. A weight is assigned to each of the first attributes based on whether each of the first attributes is correlated to an attribute in the set of second attributes. Each part in the set of parts is ranked using the weight assigned to each attribute in the first attributes. A set of interchangeable parts is identified using the ranking for each part in the set of parts.

1 - 20 . (canceled) 
     
     
         21 . A computer implemented method for identifying interchangeable parts, the computer implemented method comprising:
 identifying a set of attributes for each part in a set of parts;   assigning a rank to each part in the set of parts based on a degree of correlation between the set of attributes for a part in the set of parts and a set of identified attributes to form a set of ranked parts; and   identifying a set of interchangeable parts from the set of parts using the set of ranked parts.   
     
     
         22 . The computer implemented method of  claim 21  wherein the set of identified attributes are attributes associated with a selected part and further comprising:
 parsing, by an intelligent agent, a text description from a parts catalog associated with the selected part to identify attributes of the selected part to form the identified attributes; and 
 parsing, by the intelligent agent, text descriptions from a set of parts catalogs associated with the set of parts to identify the set of attributes for each part. 
 
     
     
         23 . The computer implemented method of  claim 21  further comprising:
 comparing, by an intelligent agent, each attribute in a set of attributes for a given part in the set of parts to the identified attributes; and 
 assigning a weight to each attribute based on whether each attribute in the set of attributes is correlated to an identified attribute. 
 
     
     
         24 . The computer implemented method of  claim 23  further comprising:
 ranking the given part using the weight assigned to each attribute. 
 
     
     
         25 . The computer implemented method of  claim 23  further comprising:
 responsive to a determination that the weight assigned to a given attribute is less than or equal to a threshold weight, discarding the given attribute. 
 
     
     
         26 . The computer implemented method of  claim 23  further comprising:
 responsive to a determination that the weight assigned to a given attribute is greater than a threshold weight, selecting the given attribute to form a set of selected attributes, wherein the set of selected attributes is used to rank each part in the set of parts. 
 
     
     
         27 . The computer implemented method of  claim 23  wherein assigning a weight to each attribute further comprises:
 identifying a special expression in a text description for a part, wherein the special expression comprises a parameter defining an attribute specific to the part; and 
 identifying the attribute corresponding to the special expression, wherein the attribute is used to determine a similarity between different parts. 
 
     
     
         28 . The computer implemented method of  claim 23  wherein comparing each attribute in a set of attributes for a given part to the identified attributes further comprises:
 checking a configuration file to identify a value correlated to a given attribute using a set of name-value pairs associated with the given attribute, wherein the value is used to determine a similarity between different parts. 
 
     
     
         29 . The computer implemented method of  claim 21  further comprising:
 consolidating a purchase order of two or more interchangeable part numbers. 
 
     
     
         30 . An apparatus for identifying interchangeable parts, the apparatus comprising:
 a relational database system in communication with an associative memory, wherein the relational database system is configured to identify a set of attributes for each part in a set of parts, assign a rank to each part in the set of parts based on a degree of correlation between the set of attributes for a part in the set of parts and a set of identified attributes to form a set of ranked parts, and identify a set of interchangeable parts from the set of parts using the set of ranked parts.   
     
     
         31 . The apparatus of  claim 30  wherein the relational database system is further configured to parse a text description from a parts catalog associated with a selected part to identify attributes of the selected part to form the identified attributes, and configured to parse text descriptions from a set of parts catalogs associated with the set of parts to identify the set of attributes for each part, and wherein the associative memory is configured to store the parts catalog and the set of parts catalogs. 
     
     
         32 . The apparatus of  claim 30  wherein the relational database system is further configured to compare each attribute in a set of attributes for a given part in the set of parts to the identified attributes, and to assign a weight to each attribute based on whether each attribute in the set of attributes is correlated to an identified attribute. 
     
     
         33 . The apparatus of  claim 32  wherein the relational database system is further configured to rank the given part using the weight assigned to each attribute. 
     
     
         34 . The apparatus of  claim 32  wherein the relational database system is further configured to discard a given attribute in response to a determination that the weight assigned to the given attribute is less than or equal to a threshold weight. 
     
     
         35 . The apparatus of  claim 32  wherein the relational database system is further configured to select a given attribute to form a set of selected attributes in response to a determination that the weight assigned to a given attribute is greater than a threshold weight, and is further configured to use the set of selected attributes to rank each part in the set of parts. 
     
     
         36 . The apparatus of  claim 32  wherein the relational database system is further configured to identify a special expression in a text description for a part, wherein the special expression comprises a parameter defining an attribute specific to the part, and wherein the relational database system is further configured to identify an attribute corresponding to the special expression, and use the attribute to determine a similarity between different parts. 
     
     
         37 . The apparatus of  claim 32  wherein the relational database system is further configured to check a configuration file to identify a value correlated to a given attribute using a set of name-value pairs associated with the given attribute, and is further configured to use the value to determine a similarity between different parts. 
     
     
         38 . The apparatus of  claim 31  further comprising:
 a user interface in communication with the relational database system, the user interface configured to receive user input identifying the selected part.