Statistical Process Control Vibrasi Bearing untuk Identifikasi Degradasi

Riyani Desriawati, Sutawanir Darwis, Nusar Hajarisman, Suliadi Suliadi, Achmad Widodo


Statistical Process Control (SPC) is usually applied to  the production process of goods, with the aim of detecting the quality of a production item that is within or beyond the specified specifications. In this study, SPC was applied to the bearing vibration signal to detect the first observable defect on a machine that functions as part of a prognostic tool for maintenance decision making. The detection of damage and prognostic are two important aspects in machine maintenance based on current conditions or better known as Condition (data) Based Maintenance (CBM). This paper discusses the shewhart average level chart and adaptive shewhart average level chart to detect the first observable defect. The  shewhart chart is built with two assumptions, i.e. that the data must vary randomly around an established mean and follows a normal distribution. However, the adaptive Shewhart  chart there is no need for normal assumption. The exploration of our data shows that the assumption of normality is not fulfilled, so that the Shewhart average level chart is not implemented. The adaptive Shewhart  chart shows that the warning line for bearing 1 amounted to 5.547 and 3.631, for bearing 2 amounted to 5.491 and 3.635, for bearing 3 amounted to 5.762 and 3, 573, for bearing 4 of 5.604 and 33.615. The action line for bearing 1 is 6.026 and 3.152, for bearing 2 is 5.955 and 3.171, for bearing 3 is 6.309 and 3.026, for bearing 4 is 6.101 and 3.118. The first observable defect was t = 81 for bearing 1,  t = 146 for bearing 2,  t = 40 for bearing 3 and  t = 61 for bearing 4.  The adaptive Shewart chart can be used as a toll to estimate the initiation of transition state from normal to degenerate.


: defect first observable, bearing vibration, adaptive Shewhart

Full Text:



Coble, J. B. (2010). Merging Data Sources to Predict Remaining Useful Life-an Automated Method to Identify Prognostic Parameters. Disertasi. Knoxville: Doctor of Philosophy Degree, The University of Tennessee.

Jardine, A.K.S., Daming, L & Dragan B. (2005). A Review on Machinery Diagnostics and Prognostics Implementing Condition-Based-Maintenance. 20(2006). 1483-1510.

López, L, Juan F, Jhouben J, Cuesta , R & Hector Fa. (2007). Gaussian Processes in Ball Bearing Prognostics. 84(203). 214-219.

Nasa Bearing Data Set, (Online). ( diakses pada 23 Juli 2019).

Negara, G A. (2018). Deteksi Cacat Multi Jenis pada Bantalan Tipe Double Row Menggunakan Sinyal Vibrasi. Skripsi. Yogyakarta: Program Studi Teknik Mesin, Fakultas Teknik, Universitas Muhammadiyah Yogyakarta.

Wan, W & Wenjuan Z. Early defect: Application of Statistical Process Control Methods. 14(3). 225-236.



  • There are currently no refbacks.

Copyright Notice

Creative Commons License
STATISTIKA is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License