METHODS AND APPARATUS TO PREDICT MACHINE FAILURES

• US 20180005132A1
• Filed: 08/31/2016
• Published: 01/04/2018
• Est. Priority Date: 07/01/2016
• Status: Active Grant

• Current Assignee: Deere & Company
• Original Assignee: Deere & Company
• Inventors: SINGH, Mahesh Pratap, POPHALE, Gauri Narayan, KUDALE, Mayuri, DAS, Hridoy
• Granted Patent: US 10,235,631 B2
• Time in Patent Office: Days
• CPC Class Codes: G05B 23/0283 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 Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks

Abstract

Methods and apparatus of modeling work vehicle data and predicting machine failures based on the same are disclosed. An example method includes accessing a first alert sequence from a work vehicle, the first alert sequence is associated with one or more of observed diagnostic code, observed vehicle sensor data, or observed vehicle status data; comparing the first alert sequence to a model including an association between reference alert sequences, reference machine failures, and probable parts used to repair the respective reference machine failures; identifying a substantial similarity between the first alert sequence and a first reference alert sequence of the reference alert sequences; and based on the model and the first reference alert sequence, identifying first probable parts of the probable parts used to repair a first reference machine failure of the reference machine failures, where the first reference machine failure is associated with the first reference alert sequence.

Claims

1. A method, comprising:

accessing a first alert sequence from a work vehicle, wherein the first alert sequence is associated with one or more of observed diagnostic code, observed vehicle sensor data, or observed vehicle status data;

comparing the first alert sequence to a model including an association between reference alert sequences, reference machine failures, and probable parts used to repair the respective reference machine failures;

identifying a substantial similarity between the first alert sequence and a first reference alert sequence of the reference alert sequences; and

based on the model and the first reference alert sequence, identifying first probable parts of the probable parts used to repair a first reference machine failure of the reference machine failures, where the first reference machine failure is associated with the first reference alert sequence.

2. The method of claim 1, further including updating the model based on the first alert sequence, wherein the model is a rule-based model associated with historic warranty and repair data for work vehicles and reference alert sequences.

3. The method of claim 1, further including, based on the model and the first alert sequence, determining a probability of the first reference machine failure occurring in the work vehicle.

4. The method of claim 3, further including generating a table including the association between the work vehicle, the probability of the first reference machine failure occurring in the work vehicle, and the first probable parts used to repair the first reference machine failure.

5. The method of claim 1, wherein the work vehicle is a first work vehicle, further including accessing second alert data from a second work vehicle and parsing the second alert data to identify a second alert sequence associated with the second work vehicle.

6. The method of claim 5, wherein parsing the second alert data includes identifying a sequence of alerts including at least one failure alert.

7. The method of claim 5, further including comparing the second alert sequence to the model;

identifying a substantial similarity between the second alert sequence and a second reference alert sequence of the reference alert sequences; and

based on the model and the second reference alert sequence, identifying second probable parts of the probable parts used to repair a second reference machine failure of the reference machine failures, where the second reference machine failure is associated with the second reference alert sequence.

8. The method of claim 7, further including updating the model by correlating a second machine failure and second probable parts used to repair the second reference machine failure;

text mining second alert data to identify the second machine failure within the second alert data;

identifying a second alert sequence associated with the second machine failure; and

updating the model by correlating the second alert sequence with second machine failure and the second probable parts used to repair the second reference machine failure.

9. A method, comprising:

generating a model including a correlation between a first machine failure and first probable parts used to repair the first machine failure;

text mining first alert data to identify the first machine failure within the first alert data;

identifying a first alert sequence associated with the first machine failure; and

updating the model by correlating the first alert sequence with the first machine failure and the first probable parts used to repair the first machine failure.

10. The method of claim 9, further including accessing second alert data from a second work vehicle and parsing the second alert data to identify a second alert sequence from the second work vehicle.

11. The method of claim 10, wherein parsing the second alert data includes identifying a sequence of alerts including at least one 1) an abnormal range alert associated with the color red;

2) a transition range alert associated with the color yellow; and

     3) a normal range alert associated with the color green.

12. The method of claim 10, further including comparing the second alert sequence to the first alert sequence;

identifying a substantial similarity between the first and second alert sequences; and

based on the model and the first alert sequence, identifying the first probable parts used to repair the first machine failure.

13. The method of claim 12, further including updating the model based on the identifying of the substantial similarity between the first and second alert sequences.

14. An article of manufacture comprising instructions, which, when executed, cause a machine to at least:

access a first alert sequence from a work vehicle;

compare the first alert sequence to a model, the model including an association between reference alert sequences, reference machine failures, and probable parts used to repair the respective reference machine failures;

identify a substantial similarity between the first alert sequence and a first reference alert sequence of the reference alert sequences; and

based on the model and the first reference alert sequence, identify first probable parts of the probable parts used to repair a first reference machine failure of the reference machine failures, where the first reference machine failure is associated with the first reference alert sequence.

15. The article of manufacture of claim 14, wherein the instructions, when executed, cause the machine to update the model based on the first alert sequence.

16. The article of manufacture of claim 14, wherein the instructions, when executed, cause the machine, based on the model and the first alert sequence, determine a probability of the first reference machine failure occurring in the work vehicle.

17. The article of manufacture of claim 16, wherein the instructions, when executed, cause the machine to generate a table including the association between the work vehicle, the probability of the first reference machine failure occurring in the work vehicle, and the first probable parts used to repair the first reference machine failure.

18. The article of manufacture of claim 14, wherein the instructions, when executed, cause the machine to access second alert data from a second work vehicle and parse the second alert data to identify a second alert sequence associated with the second work vehicle.

19. The article of manufacture of claim 18, wherein parsing the second alert data includes identifying a sequence of alerts including at least one failure alert.

20. The article of manufacture of claim 18, wherein the instructions, when executed, cause the machine to compare the second alert sequence to the model;

identify a substantial similarity between the second alert sequence and a second reference alert sequence of the reference alert sequences; and

based on the model and the second reference alert sequence, identify second probable parts of the probable parts used to repair a second reference machine failure of the reference machine failures, where the second reference machine failure is associated with the second reference alert sequence.