0% found this document useful (0 votes)

11 views

Training - Data Science

The document discusses data science and the data mining tool Rapid Miner. It covers topics such as data preparation, types of data in Rapid Miner, potential uses of PJB company data, data mining roles and algorithms, the differences between classification and clustering, and demonstrates crawling Twitter data and creating a word cloud visualization in Rapid Miner.

Uploaded by

Afrizal Miqdad

Available Formats

Download as PPTX, PDF, TXT or read online on Scribd

0% found this document useful (0 votes)

11 views

Training - Data Science

Uploaded by

Afrizal Miqdad

Available Formats

Download as PPTX, PDF, TXT or read online on Scribd

You are on page 1/ 38

KSF After Training

“Data Analitik Menggunakan

Data Science”

Mengenal Data Science & Rapid Miner

MUHAMMAD SIDDIQ B.
Juli 2022

pendekatan ilmiah yang menerapkan

matematika, statistik dan alat komputer
untuk memproses BIG DATA

KEBIJAKAN

PENGETAHUAN

INFORMASI

DATA
The information contained in this document is strictly confidential. It is strictly forbidden to use, disclose, copy, modify, or
distribute this document without permission of PEMBANGKITAN JAWA-BALI, PT sumber : Data Science, Romi Satria Wahono
DATA

50% - 80%
waktu untuk DATA PREPARATION
Steve Lohr of The New York Times said: "Data scientists, according to
interviews and expert estimates, spend 50 percent to 80 percent of their
time mired in the mundane labor of collecting and preparing unruly
digital data, before it can be explored for useful nuggets."

1. Nominal (kategori)
- Binominal
contoh : true or false, male or female
- Polynominal
contoh : rare / medium / medium well / well

2. Numeric
- Integer
contoh : -1, 0 , 1
- Real
contoh : 0,5 atau -0,0005
- Date-Time
Contoh : 19/02/2020 00:15:00

3. dll.
The information contained in this document is strictly confidential. It is strictly forbidden to use, disclose, copy, modify, or
distribute this document without permission of PEMBANGKITAN JAWA-BALI, PT
Contoh Potensi Penggunaan Data PJB

Bidang Data Attribute Data Label / Target Metode Analisa

Pendidikan, penilaian
Decision Tree “engangement
SDM kinerja, daerah asal, grade, Engangement
mutasi karyawan”
status, dll
Harga rendah, target
SCM SC, PO, P3, Nilai,Unit, dll Market Basket Analysis
kontrak payung
Parameter gas buang
Lingkungan Parameter gas buang A-Y Forecasting
Z
Data IZAT (text mining), Top finding,
K3 Klastering, decision tree
Survey penggunaan izat optimalisasi
Parameter operasi &
sensor, EOH, temuan dan
Operation, Predictive maintenance,
tindak lanjut
Maintenance, Efisiensi, Kehandalan forecasting, estimasi, outlier
pemeliharaan, data
Engineering detection
kalibrasi dan pengujian
peralatan

CLASSIFICATION
- Decision Tree
- Bayes Classification
- Neural Network

CLUSTER ANALYSIS
- K-means

OUTLIER DETECTION

The information contained in this document is strictly confidential. It is strictly forbidden to use, disclose, copy, modify, or
distribute this document without permission of PEMBANGKITAN JAWA-BALI, PT
Monitoring Temp Bearing 1 & 2
data yang diperoleh dari OMS hanya sampai tgl 8
BEFORE pemasangan barrel pin dan re-adjust EGH Nov 2021 karena perlu penggantian hardisk

Trending Temperatur di Turbine Bearing dan Compressor Bearing

160 3000
Tipe C (24 Agt – 10 Oct 20)
150
Temp Bea 1 Max : 118,38 °C
2500

1.269 EOH
Temp Bea 2 Max : 117,41 °C
140

696 EOH Alarm 115 °C 2000

130
663 EOH
117.41
120
118.38112 EOH 117.12
1500
114.88 116.12
112.22 112.39 113.07
111.43 111.66 109.52 111.43
109.22 114 109.62
110 107.87 113.29
103.31 103.45 109.92 1000
107.24 108.13 107.28
105.23 104.54 105.96 105.05
100
101.44

95.5394.3895.5296.15 500
90

80 0
25 Jul 27 Jul 19 03 10 13 14 16 20 23 10 04 31 12 05 03 43,08 42,12 42,69 44,63 69,56 88,24 95,92 82,82 95,42 97,57 73,11 57,89 66,47 58,87 50,81 46,72
20 20 Aug Aug Oct Oct Nov Nov Dec Dec Jun Jun Aug Aug Nov Nov
20 20 20 20 20 20 20 Active
20 Power
21 21 21 Brg
21 T Turbine
21 21 Brg T Compressor Brg T Compressor2
Alarm Turbine Brg T Turbine2 29,9 27,88Speed
25,312/3
26,36 31,33 35,7 40,84 29,66 39,08 36,54 36,44 27,47 36,7 33,34 42,16 42,68
0,52 0,57 4,28 0,5 57,39 47,07 4,16 3,18 4,92 2,8 4,83 2,29 3,39 2,99 3,15 3
37,77 39,96 31,91 38,2 41,47 52,89 44,89 50,22 36,46 53,08 28,65 30,59 22,73 27,26 14,97 14,48
0,6 0,73 0,78 0,72 0,71 0,61 0,74 0,61 0,61 0,63 0,77 0,7 0,7 0,79 0,76 0,68
-1,13 -1,13 -1,04 -1,12 0,72 0,64 0,77 0,64 0,64 0,68 0,8 0,73 0,72 0,83 0,8 0,71

The information contained in this document is strictly confidential. It is strictly forbidden to use, disclose, copy, modify, or 19
distribute this document without permission of PEMBANGKITAN JAWA-BALI, PT
Monitoring Temp Bearing 1 & 2
AFTER pemasangan barrel pin dan re-adjust EGH 19 Maret 2022  31 Maret 2022 : 286 EOH
19 Maret 2022  5 April 2022 : 406 EOH (estimasi)

120
Trending Temperatur di Turbine Bearing dan Compressor Bearing
140

115
111.31 120
109.49 110 109.95
110
105.99 106.12 105.66 105.52 100
105.16 105.09 105.12 104.49
104.46 103.92 104.42
105 103.16 103.75
103.05 102.72
101.57 101.46 101.3 101.48 101.82 101.92
100.76 100.41 100.15 80
99.49 99.7 99.54 99.8 99.57 99.24
100 98.78 98.62
97.8
96.21 96.32 96.65
60
95 96.06
95.74 95.72 95.34
91.89 94.59 94.98 92.1
93.46 93.97
90.1
89.51 93.08 93.01 93.1 92.72 40
90 91.18
89.52

85 20

80.44
80 0
15:30 15:45 16:00 16:15 16:30 16:45 17:00 17:15 17:30 17:45 18:00 12:45 15:30 19:00 10:45

19 Maret 2022 (Start Up before Declare) 25 Mar 25 Mar 4 Apr 5 Apr

Active Power Brg T Turbine Brg T Turbine2 Brg T Compressor Alarm TempBearing
Brg T Compressor2

The information contained in this document is strictly confidential. It is strictly forbidden to use, disclose, copy, modify, or 20
distribute this document without permission of PEMBANGKITAN JAWA-BALI, PT
Step by Step

Dataset Attribute Correlation Cross Training

& Label Matrix Validation & Testing

Siapkan data yang

ingin dianalisa

114 rows, 28.897 coloumns

Dataset Attribute Correlation Cross Training

& Label Matrix Validation & Testing

Tentukan attribute,
label, dll

sumber : Data Science, Romi Satria Wahono

Dataset Attribute Correlation Cross Training

& Label Matrix Validation & Testing

Mencari attribute yang

paling berkorelasi

Attribute Correlation Cross Training

Dataset
& Label Matrix Validation & Testing

Mencari
methode/algoritma
yang optimal

Testing Fold Training Fold

Neural Network

Linear Regression

Deep Learning

Attribute Correlation Cross Training

Dataset
& Label Matrix Validation & Testing

Lakukan pemodelan,
training, dan testing

The information contained in this document is strictly confidential. It is strictly forbidden to use, disclose, copy, modify, or
distribute this document without permission of PEMBANGKITAN JAWA-BALI, PT
Keep moving forward
Data Training Data Testing
Rev Data Training Data Testing Attribute "UNIT" Start Finish Start Finish Exclude RMS
2.5 attribute 02/19/2020 08/20/2020 08/10/2020 08/20/2020 0,536
2.5 attribute 02/19/2020 08/10/2020 08/10/2020 08/20/2020 0,753
2.5 id 02/19/2020 08/10/2020 08/10/2020 08/20/2020 0,560
2.5 id 02/19/2020 08/10/2020 08/10/2020 08/20/2020 Active Power 1,147
2.5 id 02/19/2020 08/10/2020 08/10/2020 08/20/2020 Tem Bea Comp & Turb 0,652
2.5 id 02/19/2020 08/10/2020 08/10/2020 08/20/2020 All Temp Bearing 0,995
2.5 id 02/19/2020 08/10/2020 08/10/2020 08/20/2020 All Temp Bearing & Temp Thrust 1,076
3.0 Sebelum C 180 hari 180 hari attribute 02/19/2020 08/20/2020 03/19/2022 06/11/2022 3,150
3.0 Sebelum C 180 hari 180 hari id 02/19/2020 08/20/2020 03/19/2022 06/11/2022 8,292
3.1 Setelah C 180 hari 180 hari attribute 02/19/2020 08/20/2020 12/15/2020 06/13/2021 3,722
3.1 Setelah C 180 hari 180 hari id 02/19/2020 08/20/2020 12/15/2020 06/13/2021 3,059
3.2 Setelah A 180 hari 180 hari id 02/19/2020 08/20/2020 03/19/2022 06/11/2022 8,292
4.1 Setelah C 2304 kol 2304 kol id 02/19/2020 08/20/2020 12/14/2020 06/01/2021 3,588
4.2 Setelah A 2304 kol 2304 kol id 02/19/2020 08/20/2020 05/15/2022 06/11/2022 8,311
5.1 Setelah C 2304 kol 1056 kol id 02/19/2020 08/20/2020 12/14/2020 12/26/2020 3,991
5.2 Setelah A 2304 kol 1056 kol id 02/19/2020 08/20/2020 05/28/2022 06/11/2022 6,267
6.2 Setelah A 2304 kol 479kol id 02/19/2020 08/20/2020 06/11/2022 06/11/2022 7,666
7.0 Sebelum C id 02/19/2020 08/20/2020 01/11/2019 01/11/2019 4,857
8.0 Setelah C id 10/10/2020 09/06/2021 09/07/2021 11/07/2021 1,631

The information contained in this document is strictly confidential. It is strictly forbidden to use, disclose, copy, modify, or
distribute this document without permission of PEMBANGKITAN JAWA-BALI, PT
Note
1. Pemodelan yang dilakukan belum bisa mendeteksi abnormal pada parameter
operasi, perlu dilakukan analisa lebih lanjut dengan mempertimbangkan hal
berikut:
• Karakteristik parameter operasi akan berubah setelah inspeksi major dan
dilakukan adjust di area bearing 1
• Perlu analisa Multi-Label
• Data tidak konsisten (saat OMS error, tidak ter-record)
2. Untuk analisa data yang tidak memiliki siklus teratur, jangan lakukan forecasting.
Data “date_time” dijadikan “id”, bukan “attribute”

PJB Data Science Club MTW Data Science Club

https://bit.ly/pjbdatascienceclubteknik

The information contained in this document is strictly confidential. It is strictly forbidden to use, disclose, copy, modify, or
distribute this document without permission of PEMBANGKITAN JAWA-BALI, PT
The information contained in this document is strictly confidential. It is strictly forbidden to use, disclose, copy, modify, or distribute this document without
permission of PEMBANGKITAN JAWA-BALI, PT