Prediction of the cumulative number of failures for a repairable system based on past performance

853 Ogledov
649 Prenosov
Izvoz citacije: ABNT
VEBER, Boštjan ;NAGODE, Marko ;FAJDIGA, Matija .
Prediction of the cumulative number of failures for a repairable system based on past performance. 
Strojniški vestnik - Journal of Mechanical Engineering, [S.l.], v. 53, n.10, p. 621-634, august 2017. 
ISSN 0039-2480.
Available at: <https://www.sv-jme.eu/sl/article/prediction-of-the-cumulative-number-of-failures-for-a-repairable-system-based-on-past-performance/>. Date accessed: 27 oct. 2020. 
doi:http://dx.doi.org/.
Veber, B., Nagode, M., & Fajdiga, M.
(2007).
Prediction of the cumulative number of failures for a repairable system based on past performance.
Strojniški vestnik - Journal of Mechanical Engineering, 53(10), 621-634.
doi:http://dx.doi.org/
@article{.,
	author = {Boštjan  Veber and Marko  Nagode and Matija  Fajdiga},
	title = {Prediction of the cumulative number of failures for a repairable system based on past performance},
	journal = {Strojniški vestnik - Journal of Mechanical Engineering},
	volume = {53},
	number = {10},
	year = {2007},
	keywords = {failure prediction; repairable systems; numerical modeling; parameter estimations; },
	abstract = {The prediction of the cumulative number of failures for a repairable system is an important topic in reliability theory. A repairable system may end up in one of the three possible states after a repair: 'as good as new', 'as bad as old' and 'better than old but worse than new'. Current probabilistic models used in repairable system analysis account for the first two states, but they do not properly apply to the last one, which is, more common in practice. In this paper a robust solution to a probabilistic model that is applicable to all of the three after repair states, called generalized renewal process (GRP), is presented. This research demonstrates that the GRP based on an m-fold Weibull mixture offers a general approach to modeling complex repairable systems and discusses application of the EM algorithm to estimation of the GRP parameters. This paper also presents a review of the standard GRP based on two-parameter Weibull distribution. The GRP with m mixture components distributions is compared to the standard GRP by calculating the expected cumulative number of failure, and the error function. It is shown that the proposed GRP solution with a Weibull mixture accurately describes the failure data and it is suitable for predicting failures based on the past performance of the system, oven when a small amount of failure data is available.  },
	issn = {0039-2480},	pages = {621-634},	doi = {},
	url = {https://www.sv-jme.eu/sl/article/prediction-of-the-cumulative-number-of-failures-for-a-repairable-system-based-on-past-performance/}
}
Veber, B.,Nagode, M.,Fajdiga, M.
2007 August 53. Prediction of the cumulative number of failures for a repairable system based on past performance. Strojniški vestnik - Journal of Mechanical Engineering. [Online] 53:10
%A Veber, Boštjan 
%A Nagode, Marko 
%A Fajdiga, Matija 
%D 2007
%T Prediction of the cumulative number of failures for a repairable system based on past performance
%B 2007
%9 failure prediction; repairable systems; numerical modeling; parameter estimations; 
%! Prediction of the cumulative number of failures for a repairable system based on past performance
%K failure prediction; repairable systems; numerical modeling; parameter estimations; 
%X The prediction of the cumulative number of failures for a repairable system is an important topic in reliability theory. A repairable system may end up in one of the three possible states after a repair: 'as good as new', 'as bad as old' and 'better than old but worse than new'. Current probabilistic models used in repairable system analysis account for the first two states, but they do not properly apply to the last one, which is, more common in practice. In this paper a robust solution to a probabilistic model that is applicable to all of the three after repair states, called generalized renewal process (GRP), is presented. This research demonstrates that the GRP based on an m-fold Weibull mixture offers a general approach to modeling complex repairable systems and discusses application of the EM algorithm to estimation of the GRP parameters. This paper also presents a review of the standard GRP based on two-parameter Weibull distribution. The GRP with m mixture components distributions is compared to the standard GRP by calculating the expected cumulative number of failure, and the error function. It is shown that the proposed GRP solution with a Weibull mixture accurately describes the failure data and it is suitable for predicting failures based on the past performance of the system, oven when a small amount of failure data is available.  
%U https://www.sv-jme.eu/sl/article/prediction-of-the-cumulative-number-of-failures-for-a-repairable-system-based-on-past-performance/
%0 Journal Article
%R 
%& 621
%P 14
%J Strojniški vestnik - Journal of Mechanical Engineering
%V 53
%N 10
%@ 0039-2480
%8 2017-08-18
%7 2017-08-18
Veber, Boštjan, Marko  Nagode, & Matija  Fajdiga.
"Prediction of the cumulative number of failures for a repairable system based on past performance." Strojniški vestnik - Journal of Mechanical Engineering [Online], 53.10 (2007): 621-634. Web.  27 Oct. 2020
TY  - JOUR
AU  - Veber, Boštjan 
AU  - Nagode, Marko 
AU  - Fajdiga, Matija 
PY  - 2007
TI  - Prediction of the cumulative number of failures for a repairable system based on past performance
JF  - Strojniški vestnik - Journal of Mechanical Engineering
DO  - 
KW  - failure prediction; repairable systems; numerical modeling; parameter estimations; 
N2  - The prediction of the cumulative number of failures for a repairable system is an important topic in reliability theory. A repairable system may end up in one of the three possible states after a repair: 'as good as new', 'as bad as old' and 'better than old but worse than new'. Current probabilistic models used in repairable system analysis account for the first two states, but they do not properly apply to the last one, which is, more common in practice. In this paper a robust solution to a probabilistic model that is applicable to all of the three after repair states, called generalized renewal process (GRP), is presented. This research demonstrates that the GRP based on an m-fold Weibull mixture offers a general approach to modeling complex repairable systems and discusses application of the EM algorithm to estimation of the GRP parameters. This paper also presents a review of the standard GRP based on two-parameter Weibull distribution. The GRP with m mixture components distributions is compared to the standard GRP by calculating the expected cumulative number of failure, and the error function. It is shown that the proposed GRP solution with a Weibull mixture accurately describes the failure data and it is suitable for predicting failures based on the past performance of the system, oven when a small amount of failure data is available.  
UR  - https://www.sv-jme.eu/sl/article/prediction-of-the-cumulative-number-of-failures-for-a-repairable-system-based-on-past-performance/
@article{{}{.},
	author = {Veber, B., Nagode, M., Fajdiga, M.},
	title = {Prediction of the cumulative number of failures for a repairable system based on past performance},
	journal = {Strojniški vestnik - Journal of Mechanical Engineering},
	volume = {53},
	number = {10},
	year = {2007},
	doi = {},
	url = {https://www.sv-jme.eu/sl/article/prediction-of-the-cumulative-number-of-failures-for-a-repairable-system-based-on-past-performance/}
}
TY  - JOUR
AU  - Veber, Boštjan 
AU  - Nagode, Marko 
AU  - Fajdiga, Matija 
PY  - 2017/08/18
TI  - Prediction of the cumulative number of failures for a repairable system based on past performance
JF  - Strojniški vestnik - Journal of Mechanical Engineering; Vol 53, No 10 (2007): Strojniški vestnik - Journal of Mechanical Engineering
DO  - 
KW  - failure prediction, repairable systems, numerical modeling, parameter estimations, 
N2  - The prediction of the cumulative number of failures for a repairable system is an important topic in reliability theory. A repairable system may end up in one of the three possible states after a repair: 'as good as new', 'as bad as old' and 'better than old but worse than new'. Current probabilistic models used in repairable system analysis account for the first two states, but they do not properly apply to the last one, which is, more common in practice. In this paper a robust solution to a probabilistic model that is applicable to all of the three after repair states, called generalized renewal process (GRP), is presented. This research demonstrates that the GRP based on an m-fold Weibull mixture offers a general approach to modeling complex repairable systems and discusses application of the EM algorithm to estimation of the GRP parameters. This paper also presents a review of the standard GRP based on two-parameter Weibull distribution. The GRP with m mixture components distributions is compared to the standard GRP by calculating the expected cumulative number of failure, and the error function. It is shown that the proposed GRP solution with a Weibull mixture accurately describes the failure data and it is suitable for predicting failures based on the past performance of the system, oven when a small amount of failure data is available.  
UR  - https://www.sv-jme.eu/sl/article/prediction-of-the-cumulative-number-of-failures-for-a-repairable-system-based-on-past-performance/
Veber, Boštjan, Nagode, Marko, AND Fajdiga, Matija.
"Prediction of the cumulative number of failures for a repairable system based on past performance" Strojniški vestnik - Journal of Mechanical Engineering [Online], Volume 53 Number 10 (18 August 2017)

Avtorji

Inštitucije

  • University of Ljubljana, Faculty of Mechanical Engineering, Slovenia
  • University of Ljubljana, Faculty of Mechanical Engineering, Slovenia
  • University of Ljubljana, Faculty of Mechanical Engineering, Slovenia

Informacije o papirju

Strojniški vestnik - Journal of Mechanical Engineering 53(2007)10, 621-634

The prediction of the cumulative number of failures for a repairable system is an important topic in reliability theory. A repairable system may end up in one of the three possible states after a repair: 'as good as new', 'as bad as old' and 'better than old but worse than new'. Current probabilistic models used in repairable system analysis account for the first two states, but they do not properly apply to the last one, which is, more common in practice. In this paper a robust solution to a probabilistic model that is applicable to all of the three after repair states, called generalized renewal process (GRP), is presented. This research demonstrates that the GRP based on an m-fold Weibull mixture offers a general approach to modeling complex repairable systems and discusses application of the EM algorithm to estimation of the GRP parameters. This paper also presents a review of the standard GRP based on two-parameter Weibull distribution. The GRP with m mixture components distributions is compared to the standard GRP by calculating the expected cumulative number of failure, and the error function. It is shown that the proposed GRP solution with a Weibull mixture accurately describes the failure data and it is suitable for predicting failures based on the past performance of the system, oven when a small amount of failure data is available.  

failure prediction; repairable systems; numerical modeling; parameter estimations;