Methods and systems for performing predictive maintenance on vehicle components to prevent cascading failure in a transportation system

Patent metadata

Field	Value
Publication number	US-11361237-B2
Application number	US-201816047452-A
Country	US
Kind code	B2
Filing date	Jul 27, 2018
Priority date	Jul 27, 2018
Publication date	Jun 14, 2022
Grant date	Jun 14, 2022

Abstract

Systems, methods, and processing nodes predicting and perform preventive maintenance in a transportation system. Predicting and performing preventive maintenance in a transportation system includes determining historical data for electronic devices in the transportation system. Predicting and performing preventive maintenance also includes determining dependencies of the electronic devices based on the historical data. Predicting and performing preventive maintenance includes determining a likelihood of a fault in the target electronic device during a time period based on the dependencies of the electronic devices and a mutual probability of failure of the target electronic device and parent electronic devices associated with the target electronic device. Predicting and performing preventive maintenance also includes initiating preemptive maintenance on the target electronic device based on the likelihood of the fault.

First claim

What is claimed is:

1. A method for predicting and performing preventive maintenance in a transportation system, the method comprising:
determining historical data for electronic devices in the transportation system,
wherein the electronic devices communicate in a network in the transportation system, and
wherein the historical data comprises historical device failures of the electronic devices;

determining dependencies of the electronic devices based on the historical data, wherein determining the dependencies comprises generating a graph of parent-child dependencies of the electronic devices;
selecting a target electronic device from the electronic devices to predict a fault during a time period;
determining a likelihood of a fault in the target electronic device during the time period based on the dependencies of the electronic devices and a mutual probability of failure of the target electronic device and parent electronic devices associated with the target electronic device; and
initiating preemptive maintenance on the target electronic device based on the likelihood of the fault.

2. The method of claim 1 , wherein the historical data for each of the electronic devices comprises a unique identifier and one or more dates of previous faults.

3. The method of claim 1 , wherein the electronic devices comprise electronic devices for collecting fares in the transportation system.

4. The method of claim 1 , wherein determining dependencies of the electronic devices comprises:
generating a model that describes mutual probabilities of links in the graph of parent-child dependencies of the electronic devices; and
training the model using a first set of the historical data.

5. The method of claim 4 , wherein determining the likelihood of the fault comprises:
applying a second set of the historical data to the model to determine the likelihood of the fault.

6. The method of claim 5 , wherein the model is a Bayes model.

7. The method of claim 1 , wherein initiating preemptive maintenance comprises:
performing the preemptive maintenance on the target electronic device.

8. A processing node for predicting and performing preventive maintenance in a transportation system, the processing node being configured to perform operations comprising:
determining historical data for electronic devices in the transportation system,
wherein the electronic devices communicate in a network in the transportation system, and
wherein the historical data comprises historical device failures of the electronic devices;

determining dependencies of the electronic devices based on the historical data, wherein determining the dependencies comprises generating a graph of parent-child dependencies of the electronic devices;
selecting a target electronic device from the electronic devices to predict a fault during a time period;
determining a likelihood of a fault in the target electronic device during the time period based on the dependencies of the electronic devices and a mutual probability of failure of the target electronic device and parent electronic devices associated with the target electronic device; and
initiating preemptive maintenance on the target electronic device based on the likelihood of the fault.

9. The processing node of claim 8 , wherein the historical data for each of the electronic devices comprises a unique identifier and one or more dates of previous faults.

10. The processing node of claim 8 , wherein the electronic devices comprise electronic devices for collecting fares in the transportation system.

11. The processing node of claim 8 , wherein determining dependencies of the electronic devices comprises:
generating a model that describes mutual probabilities of links in the graph of parent-child dependencies of the electronic devices; and
training the model using a first set of the historical data.

12. The processing node of claim 11 , wherein determining the likelihood of the fault comprises:
applying a second set of the historical data to the model to determine the likelihood of the fault.

13. The processing node of claim 12 , wherein the model is a Bayes model.

14. The processing node of claim 8 , wherein initiating preemptive maintenance comprises:
performing the preemptive maintenance on the target electronic device.

15. A non-transitory computer readable medium storing instructions for causing one or more processors to perform a method for predicting and performing preventive maintenance in a transportation system, the method comprising:
determining historical data for electronic devices in the transportation system,
wherein the electronic devices communicate in a network in the transportation system, and
wherein the historical data comprises historical device failures of the electronic devices;

determining dependencies of the electronic devices based on the historical data, wherein determining the dependencies comprises generating a graph of parent-child dependencies of the electronic devices;
selecting a target electronic device from the electronic devices to predict a fault during a time period;
determining a likelihood of a fault in the target electronic device during the time period based on the dependencies of the electronic devices and a mutual probability of failure of the target electronic device and parent electronic devices associated with the target electronic device; and
initiating preemptive maintenance on the target electronic device based on the likelihood of the fault.

16. The non-transitory computer readable medium of claim 15 , wherein the historical data for each of the electronic devices comprises a unique identifier and one or more dates of previous faults.

17. The non-transitory computer readable medium of claim 15 , wherein the electronic devices comprise electronic devices for collecting fares in the transportation system.

18. The non-transitory computer readable medium of claim 15 , wherein determining dependencies of the electronic devices comprises:
generating a model that describes mutual probabilities of links in the graph of parent-child dependencies of the electronic devices; and
training the model using a first set of the historical data.

19. The non-transitory computer readable medium of claim 18 , wherein determining the likelihood of the fault comprises:
applying a second set of the historical data to the model to determine the likelihood of the fault.

20. The non-transitory computer readable medium of claim 15 , wherein initiating preemptive maintenance comprises:
performing the preemptive maintenance on the target electronic device.

Assignees

• Conduent Business Services Llc

Inventors

• Agrawal Poorvi
• Solanki Bhupendra S
• Unny Narayanan

Classifications

• G06N7/01 Primary

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

• G06N20/00 Primary

  Machine learning · CPC title

• G07C5/006

  Indicating maintenance · CPC title

• G05B23/024

  Quantitative history assessment, e.g. mathematical relationships between available data; Functions therefor; Principal component analysis [PCA]; Partial least square [PLS]; Statistical classifiers, e.g. Bayesian networks, linear regression or correlation analysis; Neural networks · CPC title

• G05B23/0283

  Predictive maintenance, e.g. involving the monitoring of a system and, based on the monitoring results, taking decisions on the maintenance schedule of the monitored system; Estimating remaining useful life [RUL] (preventive maintenance, i.e. planning maintenance according to the available resources without monitoring the system G06Q10/06) · CPC title