Method for calculating error rate of alarm

Patent metadata

Field	Value
Publication number	US-9704382-B2
Application number	US-201615091261-A
Country	US
Kind code	B2
Filing date	Apr 5, 2016
Priority date	Sep 9, 2015
Publication date	Jul 11, 2017
Grant date	Jul 11, 2017

Abstract

Disclosed is a method for calculating an error rate of alarm that reports a failure of facility, calculating a true probability and a false probability of alarms generated for each time zone by a past failure record data and calculating the error rate of the alarm at a rate of the false probability to the true probability of the alarms generated for each time zone.

First claim

What is claimed is:

1. A method for calculating an error rate of an alarm that reports a failure of a facility, the method comprising:
calculating a true probability of alarms generated within a period of time occurring during a day and a false probability of the alarms generated within the period of time occurring during the day based on a past failure record data of the facility; and
calculating the error rate of the alarm as a ratio of the false probability of the alarms to the true probability of the alarms generated within the period of time occurring during the day.

2. The method of claim 1 , wherein calculating the true probability and the false probability of the alarms comprises:
calculating a cumulative distribution function (CDF) of a service life of the facility based on the past failure record data of the facility; and
calculating the cumulative distribution function of a lead time based on the past failure record data of the facility.

3. The method of claim 2 , wherein calculating the cumulative distribution function of the service life of the facility comprises:
sampling a time interval in which a facility inspection is performed, based on service life of the facility, as a plurality of sampled service life data of the facility; and
calculating the cumulative distribution function of the service life of the facility based on the plurality of sampled service life data of the facility.

4. The method of claim 3 , wherein calculating the cumulative distribution function of the service life of the facility comprises:
configuring a plurality of data groups including the plurality of the service life data, by randomly extracting any number of data among the plurality of sampled facility service life data;
calculating a probability that a certain facility service life data is included for each data group; and
calculating the cumulative distribution function of the certain facility service life data in the entire plurality of sampled service life data by calculating an average value of the probability.

5. The method of claim 2 , wherein calculating the cumulative distribution function of the lead time comprises:
sampling a difference between a time of occurrence of failure in the past and a time of generation of the alarm; and
calculating the cumulative distribution function of the lead time by a difference between the sampled time of occurrence of failure time in the past and the time of generation of the alarm.

6. The method of claim 1 , further comprising:
storing a failure record data used to calculate the error rate of the alarm and storing the error rate of the alarm.

7. The method of claim 6 , wherein storing the failure record data comprises:
storing a time at which a facility inspection is performed on the facility, an alarm generation time, a failure occurrence time, and a lead time of the alarm generation time.

8. The method of claim 1 , further comprising:
when a new alarm is generated, estimating a new error rate calculated by new failure record data corresponding to the failure record data.

9. The method of claim 8 , further comprising:
providing the new alarm to a user if the new error rate of the new alarm is less than a preset threshold value.

10. An apparatus for calculating an error rate of an alarm that reports an occurrence of failure of a facility, the apparatus comprising:
a memory configured to store a past failure record data of the facility;
a bootstrap estimation unit configured to calculate a true probability of alarms generated within a period of time occurring during a day and a false probability of the alarms generated within the period of time occurring during a day based on the past failure record data of the facility; and
an error rate calculation unit configured to calculate the error rate of the alarm as a ratio of the false probability of the alarms to the true probability of the alarms generated within the period of time occurring during the day.

11. The apparatus of claim 10 , wherein the bootstrap estimation unit is further configured to calculate a cumulative distribution function (CDF) of a service life of the facility, and a cumulative distribution function of a lead time, based on the past failure record data of the facility.

12. The apparatus of claim 11 , wherein the bootstrap estimation unit is further configured to sample a time interval of a facility inspection based on the service life of the facility, as a plurality of sampled service life data of the facility, and calculate the cumulative distribution function of the facility service life based on the plurality of sampled service life data of the facility.

13. The apparatus of claim 12 , wherein the bootstrap estimation unit is further configured to configure a plurality of data groups including the plurality of the service life data, by randomly extracting any number of data among the plurality of sampled facility service life data, calculate a probability that a certain facility service life data is included for each data group, and calculate the cumulative distribution function of the certain facility service life data in the entire plurality of sampled service life data by calculating an average value of the probability.

14. The apparatus of claim 11 , wherein the bootstrap estimation unit is further configured to sample a difference between a time of occurrence of failure in the past and a time of generation of the alarm, and calculate the cumulative distribution function of the lead time by a difference between the sampled time of failure occurrence of failure time in the past and the time of generation of the alarm.

15. The apparatus of claim 10 , wherein the memory is further configured to store:
a failure record data used to calculate the error rate of the alarm; and
the error rate of the alarm.

16. The apparatus of claim 15 , wherein the failure record data is at least one of a time of performance of facility inspection on the facility, an alarm generation time, a failure occurrence time, and a lead time of the alarm generation time.

17. The apparatus of claim 10 , further comprising:
a selection unit configured to estimate a new error rate calculated by new failure record data corresponding to the failure record data, in response to generation of the new alarm.

18. The apparatus of claim 17 , further comprising:
an alarm generation unit configured to provide the new alarm to a user if the new error rate of the new alarm is less than a preset threshold value.

19. An apparatus for calculating an error rate of an alarm that reports a failure of a facility, the apparatus comprising:
at least one processor;
a memory configured to load a computer program executed by the processors; and
a storage configured to store a computer program that manages a license, wherein the computer program comprises a set of instructions which are executable by the at least one processor to:
calculate a true probability of alarms generated and a false probability of alarms generated within a period of time occurring during a day and a false probability of the alarms generated within the period of time occurring during a day based on a past failure record data; and
calculate the error rate of the alarm as a ratio of the false probability to the true probability of the alarms generated within the period of time occurring during the day.

20. A non-transitory computer-readable recording medium having embodied thereon a computer program, which when executed by a computer causes the computer to

Assignees

• Samsung Sds Co Ltd

Inventors

• Kwon Soon-Mok
• Kang Ji-Hoon
• Yoo Dong-Ho

Classifications

• G06Q10/20 Primary

  Administration of product repair or maintenance · CPC title

• G08B29/185 Primary

  Signal analysis techniques for reducing or preventing false alarms or for enhancing the reliability of the system · CPC title

• G08B21/187

  Machine fault alarms · CPC title

• G06Q50/26

  Government or public services (business processes related to the transportation industry G06Q50/40) · CPC title

• G06Q10/06

  Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling · CPC title