US-12619952-B2 - Method and system for providing recommendations for bill of materials revision

US12619952B2US 12619952 B2US12619952 B2US 12619952B2US-12619952-B2

Abstract

For providing recommendations for bill of materials revision, a database is storing bills of materials for a set of products. A pattern learning module processes the stored bills of materials to learn patterns. Embodiments for learning structural and temporal patterns are provided. A pattern application module applies the patterns to a current bill of materials for a product of interest and forecasts recommendations, with each recommendation indicating how the current bill of materials should be updated. A user interface outputs the recommendations along with the applied patterns and their confidence values. The method and system provide an automized framework that forecasts revision for products. That framework helps to boost product quality by avoiding late recognition of change needs that would most likely negatively impact product and cost performance. Automatically assessing the change needs reduces hours spent by domain experts on these tasks, which saves internal costs.

Inventors

YUSHAN LIU
Anna Himmelhuber

Assignees

SIEMENS AKTIENGESELLSCHAFT

Dates

Publication Date: 20260505
Application Date: 20230711
Priority Date: 20220727

Claims (7)

1 . A computer implemented method for providing recommendations for a bill of materials revision, wherein the following operations are performed by modules, and wherein the modules are hardware modules and/or software modules executed by one or more processors: storing, in a database and for a set of products, bills of materials, wherein: each bill of materials is represented by a hierarchy containing components and one of the products, and for each of the products several bills of materials are stored with different timestamps, wherein the bills of materials with different timestamps are integrated into a temporal knowledge graph with timestamps added on edges of the hierarchical graphs, processing, by a pattern learning module, the stored bills of materials to learn patterns, wherein the pattern learning module implements a graph-based rule learning algorithm that processes the hierarchical graphs by traversing nodes and edges to extract subgraphs, wherein at least some of the patterns are temporal patterns that are learned by for a number of iterations: randomly sampling a sequence of the subgraphs for one of the products from two bills of materials with subsequent timestamps, and calculating a confidence value indicating a probability of an occurrence of the sequence of the subgraphs in the bills of materials or in a subset of the bills of materials, wherein each temporal pattern contains the sequence of the subgraphs and the confidence value, and/or wherein at least some of the patterns are structural patterns that are learned by for a number of iterations: randomly sampling a subgraph from one of the bills of materials, and calculating a confidence value indicating a probability of an occurrence of the subgraph in the bills of materials or in a subset of the bills of materials, wherein each structural pattern contains one of the subgraphs and the confidence value, forecasting, by a pattern application module applying the patterns to a current bill of materials for a product of interest, recommendations, with each recommendation indicating how the current bill of materials should be updated by adding a component to the current bill of materials or replacing a component in the current bill of materials, and outputting, by a user interface, the recommendations along with the applied patterns and confidence values.
2 . The method of claim 1 , wherein patterns are discarded if the confidence value is below a threshold.
3 . The method according to claim 1 , wherein: each product has a product type, and during each iteration, the confidence value is calculated by: searching occurrences of the pattern in all other bills of materials for a same product type, and dividing the number of occurrences by a total number of bills of materials for the same product type.
4 . The method according to claim 1 , further comprising: identifying, by a clustering algorithm, clusters of bills of materials, wherein during each iteration, the confidence value is calculated by: searching occurrences of the pattern in all other bills of materials in the same cluster, and dividing the number of occurrences by the total number of bills of materials in the same cluster.
5 . A system for providing recommendations for a bill of materials revision, comprising: one or more processors; a memory coupled to the one or more processors; a database storing bills of materials for a set of products, wherein: each bill of materials is represented by a hierarchy containing components and one of the products, and for each of the products several bills of materials are stored with different timestamps, wherein the bills of materials with different timestamps are integrated into a temporal knowledge graph with timestamps added on edges of the hierarchical graphs, wherein the one or more processors are configured to: process the stored bills of materials to learn patterns, and to implement a graph-based rule learning algorithm that processes the hierarchical graphs by traversing nodes and edges to extract subgraphs, wherein at least some of the patterns are temporal patterns that are learned by for a number of iterations: randomly sampling a sequence of the subgraphs for one of the products from two bills of materials with subsequent timestamps, and calculating a confidence value indicating a probability of an occurrence of the sequence of the subgraphs in the bills of materials or in a subset of the bills of materials, wherein each temporal pattern contains the sequence of the subgraphs and the confidence value, and/or wherein at least some of the patterns are structural patterns that are learned by for a number of iterations: randomly sampling a subgraph from one of the bills of materials, and calculating a confidence value indicating a probability of an occurrence of the subgraph in the bills of materials or in a subset of the bills of materials, wherein each structural pattern contains one of the subgraphs and the confidence value, apply the patterns to a current bill of materials for a product of interest, recommendations, with each recommendation indicating how the current bill of materials should be updated by adding a component to the current bill of materials or replacing a component in the current bill of materials, and output the recommendations along with the applied patterns and confidence values.
6 . A computer program product, comprising a computer readable hardware storage device having computer readable program code stored therein, said program code executable by a processor of a computer system to implement a method according to claim 1 .
7 . A provision device for the computer program product according to claim 6 , wherein the provision device stores and/or provides the computer program product.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS This application claims priority to EP Application No. 22187192.4, having a filing date of Jul. 27, 2022, the entire contents of which are hereby incorporated by reference. FIELD OF TECHNOLOGY The following refers to changing/updating product-component structures over time. In the quality management domain, the product-component structure is called bill of materials (BoM), which includes all sub-assemblies and components needed to manufacture an end product. BACKGROUND It is hard to assess the suitability of the large mass of frequently introduced new components for certain products (e.g., resistors). Recognizing whether certain product structures have to be revised and how they have to be revised is a key need. Domain experts are responsible for making these changes/revisions in the product engineering domain. An example of a bill of materials revision process is a desktop PC with subcomponent structure (hard drive, processor, main board, etc.) where the processor is replaced by a successor version. Improving this desktop PC (e.g., to increase its end-of-life date) is a manual revision task arranged by a domain expert. For different products, the bill of materials structure can also be very deep and/or very wide. J. C. Hernández Matías, H. Pérez García, J. Pérez García, A. Vizán Idoipe: Automatic generation of a bill of materials based on attribute patterns with variant specifications in a customer-oriented environment, Journal of Materials Processing Technology, Volume 199, Issues 1-3, 1 Apr. 2008, Pages 431-436, disclose a method that automatically generates bill of materials. SUMMARY According to embodiments of the method for providing recommendations for bill of materials revision, the following operations are performed by modules, wherein the modules are hardware modules and/or software modules executed by one or more processors: storing, in a database and for a set of products, bills of materials, wherein each bill of materials is represented by a hierarchical graph containing components and one of the products, andin particular, for each of the products several bills of materials are stored with different timestamps, processing, by a pattern learning module, the stored bills of materials to learn patterns, wherein at least some of the patterns are temporal patterns that are learned by for a number of iterations randomly sampling a sequence of subgraphs for one of the products from two bills of materials with subsequent timestamps, andcalculating a confidence value, with the confidence value indicating a probability of the occurrence of the sequence of subgraphs in the bills of materials or in a subset of the bills of materials,wherein each temporal pattern contains a sequence of subgraphs and its confidence value, and/orwherein at least some of the patterns are structural patterns that are learned by for a number of iterations randomly sampling a subgraph from one of the bills of materials, andcalculating a confidence value, with the confidence value indicating a probability of the occurrence of the subgraph in the bills of materials or in a subset of the bills of materials,wherein each structural pattern contains one of the subgraphs and its confidence value, forecasting, by a pattern application module applying the patterns to a current bill of materials for a product of interest, recommendations, with each recommendation indicating how the current bill of materials should be updated, in particular by adding a component to the current bill of materials or replacing a component in the current bill of materials, andoutputting, by a user interface, the recommendations along with the applied patterns and their confidence values. The system for providing recommendations for bill of materials revision comprises the following modules, wherein the modules are hardware modules and/or software modules executed by one or more processors: a database, storing bills of materials for a set of products, wherein each bill of materials is represented by a hierarchical graph containing components and one of the products, andin particular, for each of the products several bills of materials are stored with different timestamps, a pattern learning module, configured for processing the stored bills of materials to learn patterns, wherein at least some of the patterns are temporal patterns that are learned by for a number of iterations randomly sampling a sequence of subgraphs for one of the products from two bills of materials with subsequent timestamps, andcalculating a confidence value, with the confidence value indicating a probability of the occurrence of the sequence of subgraphs in the bills of materials or in a subset of the bills of materials,wherein each temporal pattern contains a sequence of subgraphs and its confidence value, and/orwherein a