Condition Based Maintenance Untuk Diagnostik Dan Prognostik Kegagalan Mesin Raw Mill Dengan Pendekatan Mahalanobis Taguchi System

Adiaksa, Tito Erlian (2022) Condition Based Maintenance Untuk Diagnostik Dan Prognostik Kegagalan Mesin Raw Mill Dengan Pendekatan Mahalanobis Taguchi System. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 02411840000155-Undergraduate_Thesis.pdf Restricted to Repository staff only Download (4MB) | Request a copy

Abstract

PT. XYZ memiliki permasalahan tingginya downtime dengan frekuensi kegagalan sebanyak 467 dengan total waktu sebanyak 2590 menit atau setara dengan 107 hari pada mesin raw mill. Mesin raw mill merupakan mesin yang terletak pada proses raw material grinding berfungsi untuk mencapur, menggiling, dan mengeringkan bahan baku utama. Penelitian ini menggunakan metode condition based maintenance untuk membuat skema keadaan diagnostik dan prognostik kegagalan mesin raw mill dengan memanfaatkan data sensor. Prognosis adalah proses yang dilakukan ketika suatu kegagalan belum terjadi. Dalam hal ini prognosis melakukan peramalan kegagalan untuk memutuskan kegagalan yang akan terjadi. Sedangkan diagnosis merupakan proses pendeteksian, pengisolasian, dan pengidentifikasian masalah. Dengan menggunakan pendekatan Mahalonobis Taguchi System (MTS) yang merupakan salah satu metode data mining menggunakan Mahalanobis Distance (MD) dan filosofi Taguchi Robust Engineering menjadikan MTS diusulkan sebagai metode diagnosis dan peramalan menggunakan data multivariate. Data multivariate dikelompokan menjadi data kondisi normal dan abnormal. Equipment yang digunakan dalam penelitian ini adalah sealing air fan (SAF), motor belt conveyor (MBC) dan bagian mill secara keseluruhan (VIB). Dengan menggunakan perhitungan mahalanobis distance didapatkan threshold untuk kondisi normal yaitu (-0.48 hingga 2.48), abnormal SAF (6.04 hingga 9.31), abnormal MBC (11.45 hingga 16.31), abnormal (59.15 hingga 163.97). Berdasarkan threshold tersebut menjadi acuan pembuatan skema diagnostik dan prognostik. Hasil penelitian ini menunjukkan bahwa skema yang dibuat telah terbukti untuk memprediksi kegagalan dan waktu kegaglan mesin raw mill
=============================================================================================================================
PT. XYZ has a problem with high downtime with a failure frequency of 467 with a total time of 2590 minutes or the equivalent of 107 days on a raw mill machine. The raw mill machine is a machine located in the raw material grinding process which functions to mix, grind, and dry the main raw materials. This study uses the condition based maintenance method to create a diagnostic and prognostic state scheme for raw mill machine failure by utilizing sensor data. Prognosis is a process that is carried out when a failure has not occurred. In this case the prognosis for forecasting failure to decide which failure will occur. Meanwhile, diagnosis is the process of detecting, isolating, and identifying problems. Using the Mahalanobis Taguchi System (MTS) approach, which is a data mining method using Mahalanobis Distance (MD) and the Taguchi Robust Engineering philosophy, MTS is proposed as a method of diagnosis and forecasting using multivariate data. Multivariate data were grouped into normal and abnormal condition data. The equipment used in this research is a sealing air fan (SAF), motor belt conveyor (MBC) and the whole mill (VIB). By using the calculation of the mahalanobis distance, the threshold for normal conditions is (-0.48 to 2.48), abnormal SAF (6.04 to 9.31), abnormal MBC (11.45 to 16.31), abnormal (59.15 to 163.97). Based on the threshold, it becomes a reference for making diagnostic and prognostic schemes. The results of this study indicate that the scheme made has been proven to predict the failure and failure time of the raw mill machine.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Breakdown, Condition Based Maintenance, Diagnosis, Downtime, Mahalanobis Taguchi System, Prognostik, Raw Mill, Semen, Breakdown, Cement, Condition Based Maintenance, Diagnostics, Downtime, Mahalanobis Taguchi System, Prognostics, Raw Mill.
Subjects:	T Technology > TH Building construction > TH3351 Maintenance and repair
Divisions:	Faculty of Industrial Technology and Systems Engineering (INDSYS) > Industrial Engineering > 26201-(S1) Undergraduate Thesis
Depositing User:	Mr. Marsudiyana -
Date Deposited:	12 Feb 2026 04:34
Last Modified:	12 Feb 2026 04:34
URI:	http://repository.its.ac.id/id/eprint/132401

Actions (login required)

View Item