Method and system for query answering over probabilistic supply chain information

Patent metadata

Field	Value
Publication number	US-11295264-B2
Application number	US-201414478123-A
Country	US
Kind code	B2
Filing date	Sep 5, 2014
Priority date	Sep 5, 2013
Publication date	Apr 5, 2022
Grant date	Apr 5, 2022

Abstract

A method predicts a derivable event in a logistic network. The method includes generating a Bayesian network describing a structure of at least a part of a logistic network. A query is received for the derivable event that depends on a combination of base events in the logistic network. The Bayesian network is instantiated for a plurality of points in time. A prediction of the derivable event is deduced from the instantiated Bayesian network by use of complex event processing.

First claim

The invention claimed is:

1. A method for predicting a derivable event in a logistic network, which comprises the steps of:
generating a Bayesian network describing a structure of at least a part of the logistic network;
receiving a query for the derivable event that depends on a combination of base events in the logistic network;
instantiating the Bayesian network for a plurality of points in time in response to the query in order to obtain a plurality of different copies of the Bayesian network; and
deducing a prediction of the derivable event by using complex event processing to combine information from the plurality of different copies of the Bayesian network, wherein the derivable event is an arrival of an item in a supply chain, and wherein the item is selected from the group consisting of objects from which goods will be produced at a factory.

2. The method according to claim 1 , wherein the Bayesian network contains nodes and edges, wherein each of the nodes of the Bayesian network have a probabilistic function representing a probabilistic distribution for one of the base events, and wherein the edges represent conditional dependencies between the base events.

3. The method according to claim 2 , wherein at least a part of the nodes of the Bayesian network are dependent on other nodes of the Bayesian network, representing events that are dependent on other base events in the logistic network, and wherein a deduction of the prediction for the derivable event is at least partially based on dependent nodes in the Bayesian network.

4. The method according to claim 2 , wherein states of the nodes of the instantiated Bayesian networks are set based on data representing the base events in the logistic network.

5. The method according to claim 4 , wherein the data are expressed in a resource description framework (RDF) format.

6. The method according to claim 1 , which further comprises performing the instantiating of the Bayesian network by creating adapted copies of the Bayesian network for every point in time relevant for a received query wherein the adapted copies are adapted to the base events that were registered in the logistic network, and/or wherein the complex event processing determines how information from different adapted copies of the Bayesian network is combined to deduce the prediction of the derivable event.

7. The method according to claim 1 , which further comprises performing the instantiating of the Bayesian network by construction of one Bayesian network per point in time in the logistic network.

8. The method according to claim 1 , wherein the query and/or the derivable event contains a condition, wherein the condition is a temporal condition and/or an availability of an object.

9. The method according to claim 8 , wherein the condition is a time based availability of the object.

10. The method according to claim 1 , wherein the query is expressed using a query language.

11. The method according to claim 1 , wherein the prediction for the derivable event is repeated based on updated information in the Bayesian network when a base event occurs in the logistic network arrives.

12. The method according to claim 1 , wherein the query is expressed using SPARQL and evaluated over the instantiated Bayesian network.

13. The method according to claim 1 , the logistic network is a supply chain.

14. A device, comprising:
an apparatus selected from the group consisting of a processor and a hard-wired circuit, said apparatus configured to perform a method for predicting a derivable event in a logistic network, which comprises the steps of:
generating a Bayesian network describing a structure of at least a part of the logistic network;
receiving a query for the derivable event that depends on a combination of base events in the logistic network;
instantiating the Bayesian network for a plurality of points in time in response to the query in order to obtain a plurality of different copies of the Bayesian network; and
deducing a prediction of the derivable event by using complex event processing to combine information from the plurality of different copies of the Bayesian network, wherein the derivable event is an arrival of an item in a supply chain, and wherein the item is selected from the group consisting of objects from which goods will be produced at a factory.

15. A non-transitory computer program product, directly loadable into a memory of a digital computer, containing software code portions for performing a method for predicting a derivable event in a logistic network, which comprises the steps of:
generating a Bayesian network describing a structure of at least a part of the logistic network;
receiving a query for the derivable event that depends on a combination of base events in the logistic network;
instantiating the Bayesian network for a plurality of points in time in response to the query in order to obtain a plurality of different copies of the Bayesian network; and
deducing a prediction of the derivable event by using complex event processing to combine information from the plurality of different copies of the Bayesian network, wherein the derivable event is an arrival of an item in a supply chain, and wherein the item is selected from the group consisting of objects from which goods will be produced at a factory.

16. A non-transitory computer readable medium, having computer-executable instructions adapted to cause a computer system to perform a method for predicting a derivable event in a logistic network, which comprises the steps of:
generating a Bayesian network describing a structure of at least a part of the logistic network;
receiving a query for the derivable event that depends on a combination of base events in the logistic network;
instantiating the Bayesian network for a plurality of points in time in response to the query in order to obtain a plurality of different copies of the Bayesian network, wherein the Bayesian network includes a plurality of nodes, each one of the plurality of nodes has a probabilistic function representing a probabilistic distribution for a respective event; and
deducing a prediction of the derivable event by using complex event processing to combine information from the plurality of different copies of the Bayesian network, wherein the derivable event is an arrival of an item in a supply chain, and wherein the item is selected from the group consisting of objects from which goods will be produced at a factory.

Assignees

• Siemens Ag

Inventors

• Hansen Ralf
• Hubauer Thomas
• Lamparter Steffen
• Lepratti Raffaello

Classifications

• G06N7/01 Primary

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

• G06Q10/08 Primary

  Logistics, e.g. warehousing, loading or distribution; Inventory or stock management · CPC title

• G06N7/005

  Physics · mapped topic