Melani, Nanda Rizki (2026) Analisis Keandalan High Pressure Boiler Feed Pump Berbasis Weibull-RCM Dengan Peningkatan Akurasi Estimasi Parameter Melalui Metode Bootstrap. Other thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 5002221051-Undergraduate_Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
5002221051-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (3MB) | Request a copy |
Abstract
High Pressure Boiler Feed Pump (HP BFP) merupakan salah satu komponen kritis pada sistem Pembangkit Listrik Tenaga Gas dan Uap (PLTGU) karena berfungsi menyuplai air umpan bertekanan tinggi menuju Heat Recovery Steam Generator (HRSG). Kegagalan pada komponen ini dapat menyebabkan gangguan operasi, peningkatan downtime, dan penurunan keandalan sistem pembangkit sehingga diperlukan analisis keandalan sebagai dasar penentuan kebijakan perawatan. Namun, estimasi parameter Weibull pada data Time Between Failure (TBF) dengan ukuran sampel relatif kecil dapat menghasilkan ketidakpastian estimasi yang tinggi. Penelitian ini bertujuan menganalisis keandalan HP BFP menggunakan distribusi Weibull dan menentukan kebijakan perawatan melalui pendekatan Reliability Centered Maintenance (RCM) dengan peningkatan kualitas estimasi parameter menggunakan metode bootstrap. Data yang digunakan merupakan data TBF HP BFP Blok I PT X periode Januari 2015–Desember 2024 yang diklasifikasikan ke dalam enam failure mode. Parameter Weibull diestimasi menggunakan Maximum Likelihood Estimation (MLE). Untuk mengurangi variasi estimator pada sampel relatif kecil, diterapkan bootstrap parametrik median sebanyak 10.000 replikasi. Hasil analisis menunjukkan bahwa failure mode A, C, D, dan F memiliki β > 1 yang menunjukkan pola kegagalan wear-out, sedangkan failure mode B dan E memiliki β < 1 yang menunjukkan pola early failure. Metode bootstrap menghasilkan standard error (SE) yang lebih kecil dan interval kepercayaan dengan batas bawah positif dibandingkan MLE. Berdasarkan hasil analisis keandalan, failure mode A (Vibration and Mechanical Failure) menjadi prioritas utama perawatan karena memiliki MTBF terkecil sebesar 111,03 hari dan nilai keandalan terendah pada waktu operasi 365 hari sebesar 0,0016.
======================================================================================================================================
The High Pressure Boiler Feed Pump (HP BFP) is one of the critical components in a Combined Cycle Power Plant (CCPP) as it supplies high-pressure feedwater to the Heat Recovery Steam Generator (HRSG). Failure of this component may cause operational disruptions, increased downtime, and reduced system reliability, making reliability analysis essential for determining appropriate maintenance policies. However, estimating Weibull parameters from Time Between Failure (TBF) data with relatively small sample sizes may result in high estimation uncertainty. This study aims to analyze the reliability of the HP BFP using the Weibull distribution and determine maintenance policies through the Reliability Centered Maintenance (RCM) approach by improving parameter estimation using the bootstrap method. The data used consist of HP BFP TBF records from Block I of PT X covering the period from January 2015 to December 2024, which were classified into six failure modes. The Weibull parameters were estimated using the Maximum Likelihood Estimation (MLE) method. To reduce estimator variability in relatively small samples, a parametric median bootstrap with 10,000 replications was employed. The results show that failure modes A, C, D, and F have shape parameter values of β > 1, indicating a wear-out failure pattern, whereas failure modes B and E have β < 1, indicating an early failure pattern. The bootstrap method produced smaller standard errors (SE) and confidence intervals with positive lower bounds compared with the MLE method. Based on the reliability analysis, failure mode A (Vibration and Mechanical Failure) was identified as the highest maintenance priority because it has the shortest Mean Time Between Failure (MTBF) of 111.03 days and the lowest reliability value of 0.0016 at an operating time of 365 days.
| Item Type: | Thesis (Other) |
|---|---|
| Uncontrolled Keywords: | Bootstrap, Distribusi Weibull, High Pressure Boiler Feed Pump, Keandalan, Reliability Centered Maintenance, Bootstrap, High Pressure Boiler Feed Pump, Reliability, Reliability Centered Maintenance, Weibull Distribution. |
| Subjects: | Q Science > QA Mathematics > QA273.6 Weibull distribution. Logistic distribution. Q Science > QA Mathematics > QA276 Mathematical statistics. Time-series analysis. Failure time data analysis. Survival analysis (Biometry) |
| Divisions: | Faculty of Science and Data Analytics (SCIENTICS) > Mathematics > 44201-(S1) Undergraduate Thesis |
| Depositing User: | Nanda Rizki Melani |
| Date Deposited: | 31 Jul 2026 03:21 |
| Last Modified: | 31 Jul 2026 03:21 |
| URI: | http://repository.its.ac.id/id/eprint/140584 |
Actions (login required)
 |
View Item |