Authors :
Fajar Sodik; Kusrini; Kusnawi
Volume/Issue :
Volume 8 - 2023, Issue 10 - October
Google Scholar :
https://tinyurl.com/yjb5sser
Scribd :
https://tinyurl.com/28s2tdba
DOI :
https://doi.org/10.5281/zenodo.10049761
Abstract :
Maintenance is something that must be done
on equipment to maintain its reliability. It is necessary to
determine the correct maintenance period to make it
more effective and efficient so that reliability is
maintained while being efficient in terms of costs
incurred. This research aims to determine the best
algorithm between polynomial regression and Nadaraya
Watson kernel regression to determine the maintenance
period for train detection equipment and determine the
variables that influence the determination of the
maintenance period, which has an impact on equipment
reliability. Testing the polynomial regression model
produces a mean absolute error of 8.05, a mean squared
error of 568.74, and a determination coefficient of 0.999,
while the Nadaraya Watson regression model produces a
mean absolute error of 3.14, a mean squared error of
19.43, and a determination coefficient of 0.938. Thus, it
can be concluded that the Nadaraya Watson Kernel
Regression model can be used well to determine the
maintenance period for train detection equipment.
Keywords :
Polynomial Regression; Nadaraya Watson Kernel Regression; Train Detector.
Maintenance is something that must be done
on equipment to maintain its reliability. It is necessary to
determine the correct maintenance period to make it
more effective and efficient so that reliability is
maintained while being efficient in terms of costs
incurred. This research aims to determine the best
algorithm between polynomial regression and Nadaraya
Watson kernel regression to determine the maintenance
period for train detection equipment and determine the
variables that influence the determination of the
maintenance period, which has an impact on equipment
reliability. Testing the polynomial regression model
produces a mean absolute error of 8.05, a mean squared
error of 568.74, and a determination coefficient of 0.999,
while the Nadaraya Watson regression model produces a
mean absolute error of 3.14, a mean squared error of
19.43, and a determination coefficient of 0.938. Thus, it
can be concluded that the Nadaraya Watson Kernel
Regression model can be used well to determine the
maintenance period for train detection equipment.
Keywords :
Polynomial Regression; Nadaraya Watson Kernel Regression; Train Detector.
