Sensor network for optimized maintenance schedule
Embodiments determine an optimized maintenance schedule for a maintenance program that includes multiple levels, each level including at least one asset (i.e., asset type) and at least one of the levels including a plurality of assets. Embodiments receive historical failure data for each of the assets, the historical failure data generated at least in part by a sensor network. For each asset, embodiments generate a probability density function (“PDF”) using kernel density estimation (“KDE”). For each asset, based on a reliability rate threshold, embodiments determine a cumulative density function (“CDF”) using the PDF. For each asset, embodiments determine an optimized time to failure (“TTF”) using the CDF. Embodiments then create the schedule for each level that includes a minimum TTF for the assets at each level.
1. A method of determining an optimized maintenance schedule for a maintenance program comprising multiple levels, each level comprising at least one asset and at least one of the levels comprising a plurality of assets, the method comprising:
receiving historical failure data for each of the assets, the historical failure data generated at least in part by a sensor network;
for each asset, generating a probability density function (PDF) using kernel density estimation (KDE);
for each asset, based on a reliability rate threshold, determining a cumulative density function (CDF) using the PDF;
for each asset, determining an optimized time to failure (TTF) using the CDF; and
creating the schedule for each level that comprises a minimum TTF for all of the assets at each level.
2. The method of claim 1 , wherein the reliability rate threshold is based on a received minimum reliability rate and is determined using a mean of the PDF and a mean of a window around the mean of the PDF.
3. The method of claim 1 , wherein the maintenance program is for an automobile, and the assets are parts for the automobile.
4. The method of claim 3 , wherein the minimum TTF is based on a mileage of the automobile or on a number of days of age of the automobile.
5. The method of claim 1 , further comprising calculating a confidence level for the determined optimized TTF.
6. The method of claim 5 , wherein the calculating comprises using K-fold cross validation.
7. The method of claim 1 , wherein generating the PDF using KDE comprises:
f
^
h
(
x
)
=
1
n
∑
i
=
1
n
K
h
(
x
-
x
i
)
=
1
nh
∑
i
=
1
n
K
(
x
-
x
i
h
)
,
where K is a kernel comprising a non-negative function and h comprises a smoothing parameter and is greater than 0.
8. The method of claim 1 , further comprising using machine learning to cluster groups of the assets into sub-groups to improve an accuracy of the optimized TTF.
9. A non-transitory computer-readable medium storing instructions which, when executed by one or more processors, cause the processors to determine an optimized maintenance schedule for a maintenance program comprising multiple levels, each level comprising at least one asset and at least one of the levels comprising a plurality of assets, the determining comprising:
receiving historical failure data for each of the assets, the historical failure data generated at least in part by a sensor network;
for each asset, generating a probability density function (PDF) using kernel density estimation (KDE);
for each asset, based on a reliability rate threshold, determining a cumulative density function (CDF) using the PDF;
for each asset, determining an optimized time to failure (TTF) using the CDF; and
creating the schedule for each level that comprises a minimum TTF for all of the assets at each level.
10. The computer-readable medium of claim 9 , wherein the reliability rate threshold is based on a received minimum reliability rate and is determined using a mean of the PDF and a mean of a window around the mean of the PDF.
11. The computer-readable medium of claim 9 , wherein the maintenance program is for an automobile, and the assets are parts for the automobile.
12. The computer-readable medium of claim 11 , wherein the minimum TTF is based on a mileage of the automobile or on a number of days of age of the automobile.
13. The computer-readable medium of claim 9 , the determining further comprising calculating a confidence level for the determined optimized TTF.
14. The computer-readable medium of claim 13 , wherein the calculating comprises using K-fold cross validation.
15. The computer-readable medium of claim 9 , wherein generating a-the PDF using KDE comprises:
f
^
h
(
x
)
=
1
n
∑
i
=
1
n
K
h
(
x
-
x
i
)
=
1
nh
∑
i
=
1
n
K
(
x
-
x
i
h
)
,
where K is a kernel comprising a non-negative function and h comprises a smoothing parameter and is greater than 0.
16. The computer-readable medium of claim 9 , the determining further comprising using machine learning to cluster groups of the assets into sub-groups to improve an accuracy of the optimized TTF.
17. A cloud based sensor system for determining an optimized maintenance schedule for a maintenance program comprising multiple levels, each level comprising at least one asset and at least one of the levels comprising a plurality of assets, the system comprising:
a plurality of sensors coupled to assets and in communication over a network;
one or more processors that receives historical failure data for each of the assets from sensor readings of asset failures and, for each of the assets;
generating a probability density function (PDF) using kernel density estimation (KDE);
based on a reliability rate threshold, determining a cumulative density function (CDF) using the PDF;
determining an optimized time to failure (TTF) using the CDF;
the processors creating the schedule for each level that comprises a minimum TTF for all of the assets at each level.
18. The system of claim 17 , wherein the reliability rate threshold is based on a received minimum reliability rate and is determined using a mean of the PDF and a mean of a window around the mean of the PDF.
19. The system of claim 17 , wherein the maintenance program is for an automobile, and the assets are parts for the automobile.
20. The system of claim 19 , wherein the minimum TTF is based on a mileage of the automobile or on a number of days of age of the automobile.