Nafisa, Lia Zahrotun (2019) Penerapan Metode Bootstrap pada Zero-Failure Data dalam Penentuan Nilai Keandalan (Studi Kasus: Mesin PT. X). Undergraduate thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
06111440000086_Undergraduate_Theses.pdf Download (2MB) | Preview |
Abstract
Keandalan suatu mesin menjadi hal yang penting dalam suatu perusahaan. Untuk itu diperlukan uji keandalan pada suatu mesin agar dapat mengetahui seberapa tahan waktu mesin tersebut bekerja dengan baik. Akan tetapi terdapat permasalahan ketika data mesin yang didapat berupa zero-failure data yang berukuran kecil sehingga data tersebut sulit untuk dianalisa bahkan tidak bisa diketahui distribusinya. Untuk mengatasi masalah tersebut, diterapan metode bootstrap pada zero-failure data yang ada. Bootstrap adalah metode resampel yang dapat bekerja tanpa asumsi distribusi pada suatu data. Selanjutnya, digunakan estimator densitas kernel Gaussian untuk mendapatkan nilai keandalannya. Studi kasus dilakukan pengambilan data pada komponen mesin di PT. X, yang mana mesin tersebut harus bekerja tanpa henti untuk memenuhi jumlah produksi diperusahaan. Didapatkan nilai keandalan komponen bearing mesin Double Planer pada awal waktu pemasangan sampai pada waktu 28 minggu cukup tinggi yaitu 0.8694013. Untuk kerusakan waktu 29 minggu menurun dikarenakan beberapa faktor yaitu faktor kurangnya pelumasan pada bearing, pemasangan bearing yang kurang pas, dan tekstur kayu yang dibentuk.
================================================================================================
Reliability of a machine becomes important in a company. For this reason, a reliability test is needed on a machine so that it can determine how long the machine is working properly. But there is a problem when the machine data obtained is in the form of zero-failure data that is small in size so that the data is difficult to analyze and even the distribution cannot be known. To overcome this problem, the bootstrap method applied to existing zero-failure data. Bootstrap is a resampel method that can work without distribution assumptions on a data. Next, a Gaussian kernel density estimator is used to get the reliability value. Case studies carried out data retrieval on machine components at PT. X, where the machine must work nonstop to meet the amount of production in the company. The value of reliability of double planer engine bearings at the beginning of installation time up to 28 weeks is quite high at 0.8694013. For 29 weeks of declining time due to several factors, namely the lack of lubrication in the bearing, the installation of bearings that are less suitable, and the texture of the wood formed.
| Item Type: | Thesis (Undergraduate) |
|---|---|
| Additional Information: | RSMa 620.004 52 Naf p-1 2019 |
| Uncontrolled Keywords: | Reliability, Zero-failure data, Bootstrap method, Gaussian kernel density estimator |
| Subjects: | H Social Sciences > HA Statistics > HA29 Theory and method of social science statistics Q Science Q Science > QA Mathematics Q Science > QA Mathematics > QA275 Theory of errors. Least squares. Including statistical inference. Error analysis (Mathematics) T Technology > TS Manufactures > TS174 Maintainability (Engineering) . Reliability (Engineering) |
| Divisions: | Faculty of Science and Data Analytics (SCIENTICS) > Mathematics > 44201-(S1) Undergraduate Thesis |
| Depositing User: | Lia Zahrotun Nafisa |
| Date Deposited: | 25 May 2022 05:57 |
| Last Modified: | 25 May 2022 05:57 |
| URI: | http://repository.its.ac.id/id/eprint/66152 |
Actions (login required)
 |
View Item |