Industrial data service, data modeling, and data application platform

Patent metadata

Field	Value
Publication number	US-11410112-B2
Application number	US-201716621089-A
Country	US
Kind code	B2
Filing date	Oct 27, 2017
Priority date	Oct 27, 2017
Publication date	Aug 9, 2022
Grant date	Aug 9, 2022

Abstract

This disclosure relates to industrial data services, data modeling and applications for controlling an industrial operation. In one implementation, a platform is disclosed for allocating a data modeling request to a collaborative group of experts based on a two-dimensional data modeling flow data structure and a multilayer resource allocation graph to obtain a data model for controlling the industrial operation. The two-dimensional data modeling flow data structure and the multilayer resource allocation graph are established from an industrial graph knowledgebase using various data analytics and machine learning techniques.

First claim

What is claimed is:

1. A system, comprising:
a graph knowledgebase containing data elements linked to an industrial operation and relationships between the data elements;
a log for operational data of the industrial operation;
a memory for storing:
a predefined data modeling template comprising a data modeling stage dimension and a resource dimension representing predefined data modeling stages and predefined types of resources for data modeling tasks; and
a resource allocation graph structure having graph elements organized into a resource layer, a process layer, and a data modeling stage layer, where the graph elements in the data modeling stage layer correspond to the predefined data modeling stages; and

system circuitry configured to:
extract, from the graph knowledgebase, resource data elements classified into the predefined types of resources to obtain the graph elements for the resource layer of the resource allocation graph structure;
extract industrial process data elements from the graph knowledgebase to obtain the graph elements for the process layer of the resource allocation graph structure;
obtain, based on the graph knowledgebase, relationships between the graph elements in the resource layer and the process layer and between the graph elements in the process layer and the data modeling stage layer of the resource allocation graph structure;
upon receiving a request for a data modeling task:
determine a set of industrial processes associated with the data modeling task;
select a group of resource graph elements from the resource allocation graph structure based on the set of industrial processes and the relationships between the graph elements in the resource layer and the process layer; and
automatically allocate the data modeling task to the selected group of resource graph elements for modeling the operational data from the log at each of the predefined data modeling stages to obtain a data model according to the resource allocation graph structure; and
control the industrial operation based on the data model.

2. The system of claim 1 , where the graph elements of the resource layer comprise allocable human resource graph elements.

3. The system of claim 2 , where the allocable human resource graph elements of the resource layer are grouped into multiple predefined types of human resource graph elements and the group of graph elements selected form the resource layer comprise at least one graph element from each of the multiple predefined types of human resource graph elements.

4. The system of claim 3 , where the allocable human resource graph elements, when being extracted from the graph knowledgebase, are classified into the multiple predefined types of human resource graph elements using a classifier established from the graph knowledgebase based on a machine learning algorithm.

5. The system claim 3 , where the multiple predefined types of human resource graph elements comprise types of expertise associated with data modeling of industrial operational data.

6. The system of claim 3 , where the system circuitry is further configured to select, among a predefined collection of data modeling tasks, the data modeling task according to the data modeling request.

7. The system of claim 6 , the memory further comprising a data modeling flow data structure comprising a data modeling stage dimension and a resource dimension representing the predefined set of data modeling stages and the multiple predefined types of human resource graph elements, and where the data modeling flow data structure further specifies association between pairs of the predefined set of data modeling stages and the predefined types of human resource graph elements, and the data modeling task.

8. The system of claim 7 , where the system circuitry is configured to automatically allocate the data modeling request further according to the data modeling flow data structure.

9. The system of claim 1 , where inter-layer relationships between the graph elements are quantified based on the data elements and relationships in the graph knowledgebase and system circuitry is configured to automatically allocate the data modeling request further based on the quantified inter-layer relationships.

10. A method, comprising:
extracting, from a graph knowledgebase containing data elements linked to an industrial operation and relationships between the data elements, resource data elements classified into predefined types of resources to obtain graph elements for a resource layer of a resource allocation graph structure;
extracting industrial process data elements from the graph knowledgebase to obtain the graph elements for a process layer of the resource allocation graph structure;
obtaining, based on the graph knowledgebase, relationships between the graph elements in the resource layer and the process layer and between the graph elements in the process layer and a data modeling stage layer of the resource allocation graph structure;
in response to receiving a request for a data modeling task:
determining a set of industrial processes associated with the data modeling task;
selecting a group of resource graph elements from the resource allocation graph structure based on the set of industrial processes and the relationships between the graph elements in the resource layer and the process layer; and
automatically allocating the data modeling task to the selected group of resource graph elements for modeling the operational data from a log for operational data of the industrial operation at each of a set of predefined data modeling stages to obtain a data model according to the resource allocation graph structure; and

controlling the industrial operation based on the data model.

11. The method of claim 10 , where the graph elements of the resource layer comprise allocable human resource graph elements.

12. The method of claim 11 , where the allocable human resource graph elements of the resource layer are grouped into multiple predefined types of human resource graph elements and the group of graph elements selected form the resource layer comprise at least one graph element from each of the multiple predefined types of human resource graph elements.

13. The method of claim 12 , where the allocable human resource graph elements, when being extracted from the graph knowledgebase, are classified into the multiple predefined types of human resource graph elements using a classifier established from the graph knowledgebase based on a machine learning algorithm.

14. The method claim 12 , where the multiple predefined types of human resource graph elements comprise types of expertise associated with data modeling of industrial operational data.

15. The method of claim 12 , further comprising selecting, among a predefined collection of data modeling tasks, the data modeling task according to the data modeling request.

16. The method of claim 15 , further comprising maintaining a data modeling flow data structure including a data modeling stage dimension and a resource dimension representing the predefined set of data modeling stages and the multiple predefined types of human resource graph elements, and automatically allocating the data modeling request according to the data modeling flow data structure, where the data modeling flow data structure further specifies association between pairs of the predefined set of data modeling stages and the predefined types of human resource graph elements, and the data modeling task.

17. The method of claim 10 , where inter-layer relationships between the gr

Assignees

• Accenture Global Solutions Ltd

Inventors

• Hou Fang
• Wu Yikai
• Cheng Xiaopei
• Ma Guo
• Sun Guanyi
• Xie Bin
• Shen Hui

Classifications

• G06N7/01

  Probabilistic graphical models, e.g. probabilistic networks · CPC title

• G06Q10/067 Primary

  Enterprise or organisation modelling · CPC title

• G06Q10/10

  Office automation; Time management · CPC title

• G06Q10/103

  Workflow collaboration or project management · CPC title

• G05B19/41875

  characterised by quality surveillance of production · CPC title