Implementasi machine learning dengan algoritma regresi untuk memprediksi kualitas produk = Implementation of machine learning with regression algorithm to predict quality of product

Dastariana, Anggi (2021) Implementasi machine learning dengan algoritma regresi untuk memprediksi kualitas produk = Implementation of machine learning with regression algorithm to predict quality of product. Bachelor thesis, Universitas Pelita Harapan.

[img] Text (Title)
Title (28).pdf
Restricted to Repository staff only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (732kB)
[img]
Preview
Text (Abstract)
Abstract.pdf
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (59kB) | Preview
[img]
Preview
Text (ToC)
ToC.pdf
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (407kB) | Preview
[img]
Preview
Text (Chapter1)
Chapter1.pdf
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (185kB) | Preview
[img] Text (Chapter2)
Chapter2.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (715kB)
[img] Text (Chapter3)
Chapter3.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (286kB)
[img] Text (Chapter4)
Chapter4.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (628kB)
[img] Text (Chapter5)
Chapter5.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (597kB)
[img] Text (Chapter6)
Chapter6.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (56kB)
[img]
Preview
Text (Bibliography)
Bibliography.pdf
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (285kB) | Preview
[img] Text (Appendices)
Appendices.pdf
Restricted to Repository staff only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (2MB)

Abstract

Aktivitas industri dan manufaktur berkembang pesat seiring meningkatnya permintaan konsumen terhadap suatu barang. Untuk memenuhi permintaan tersebut, produsen harus menjaga kualitas setiap barang sesuai dengan standar kebutuhan ataupun keinginan dari konsumen, sehingga dibutuhkan pengujian pada setiap produk yang dihasilkan. Produk dengan kondisi yang tidak baik akan dieliminasi dan dibuang sehingga dapat menimbulkan kerugian dan dibutuhkan tindakan pencegahan agar produk dengan status reject dapat dikurangi ataupun dihindari. Dalam penelitian ini tindakan pencegahan tersebut dilakukan dengan memanfaatkan dan menganalisis data proses dan data produk yang berkaitan dengan pengujian laboratorium dengan menggunakan analisis regresi linear dan regresi logistik dengan implementasi Machine Learning. Proses analisis dilakukan dengan mencari korelasi (regresi) dari data pengujian proses (sub-material) dengan parameter elongation at break dan tensile strength dan data pengujian akhir (barang jadi water hose) yang terdiri dari parameter bursting pressure dan crack resistance. Performa model dievaluasi dengan menggunakan R2 score dan p-value. Dari model simple linear regression didapatkan model dengan populasi yang paling mendekati garis regresi adalah variabel elongation at break dengan R2 score 81.8% dan tensile break dengan R2 score 73.5%. Sedangkan untuk model multiple linear regression menghasilkan R2 score yang lebih tinggi yaitu 86.8%. Model regresi logistik dievaluasi dengan confussion matrix yang menghasilkan parameter accuracy, precision, dan recall sebesar 91% dan F1-score sebesar 90%. Tindak lanjut yang dilakukan adalah standarisasi batas minimum pengujian rubber compound untuk parameter tensile strength dan elongation at break. Dengan cara ini data pengujian water hose yang tidak memenuhi standar mengalami penurunan sebesar 60% untuk parameter bursting pressure dan 44% untuk parameter crack resistance. / Industrial and manufacturing activities are growing rapidly with increasing consumer demand for goods. To fulfill the demand of such consumers, goods supplier must maintain the quality of each goods made in accordance with the standards of needs or wishes of the consumer, so conduct tests is needed on each goods production. Products with not good condition will be eliminated and disposed so it can cause disadvantage and the company must take preventive action in order to reduce or avoid products with reject status. In this study, preventive action will be possible by utilizing and analyzing the testing data of each product with a linear regression and logistic regression analysis in the development of the Machine Learning model. The analysis process is carried out by looking for a correlation (regression) from the process test data (sub-material) with elongation at break and tensile strength parameters and the final test data (water hose finished goods) consisting of bursting pressure and crack resistance parameters. The performance of the developed model is evaluated using R2 score and P-value. From simple linear regression, obtained models with the population closest to the regression line is elongation at break variable with R2 score of 81.8% and tensile break with R2 score 73.5%. As for the multiple linear models regression produce a higher R2 score of 86.8%. Logistic regression evaluated with confussion matrix which result in an accuracy, precision and recall parameter of 91% and F1-score of 90%. The follow-up after the development of the model was the standardization of the minimum limit for rubber compound testing at tensile strength and elongation at break parameters. After the follow-up, the water hose test data that not eligeble with the standards decreased by 60% for the bursting pressure parameter and 44% for the crack resistance parameter.

Item Type: Thesis (Bachelor)
Creators:
CreatorsNIMEmail
Dastariana, AnggiNIM01035180005anggidast@gmail.com
Contributors:
ContributionContributorsNIDN/NIDKEmail
Thesis advisorMartoyo, IhanNIDN0318057301UNSPECIFIED
Uncontrolled Keywords: analisis regresi; machine learning; kualitas produk
Subjects: T Technology > TK Electrical engineering. Electronics Nuclear engineering
Divisions: University Subject > Current > Faculty/School - UPH Karawaci > Faculty of Science and Technology > Electrical Engineering
Current > Faculty/School - UPH Karawaci > Faculty of Science and Technology > Electrical Engineering
Depositing User: Users 16928 not found.
Date Deposited: 08 Mar 2021 03:29
Last Modified: 08 Mar 2021 03:29
URI: http://repository.uph.edu/id/eprint/27953

Actions (login required)

View Item View Item