Bilowo, Gidon (2024) Optimalisasi orientasi pada 3D printer: analisis perbandingan NSGA-II dengan kombinasi NSGA-II dan QLEARNING. Masters thesis, Universitas Pelita Harapan.
![Title [thumbnail of Title]](http://repository.uph.edu/style/images/fileicons/text.png)
halaman Judul_Gidon Bilowo 1.pdf
Available under License Creative Commons Attribution Non-commercial Share Alike.
Download (259kB)
![Abstract [thumbnail of Abstract]](http://repository.uph.edu/style/images/fileicons/text.png)
halaman Abstrak_Gidon Bilowo 1.pdf
Available under License Creative Commons Attribution Non-commercial Share Alike.
Download (664kB)
![ToC [thumbnail of ToC]](http://repository.uph.edu/style/images/fileicons/text.png)
Halaman Daftar Isi_Gidon Bilowo 1_watermark.pdf
Available under License Creative Commons Attribution Non-commercial Share Alike.
Download (725kB)
![Chapter 1 [thumbnail of Chapter 1]](http://repository.uph.edu/style/images/fileicons/text.png)
BAB I_Gidon Bilowo 1.pdf
Available under License Creative Commons Attribution Non-commercial Share Alike.
Download (1MB)
![Chapter 2 [thumbnail of Chapter 2]](http://repository.uph.edu/style/images/fileicons/text.png)
BAB II_Gidon Bilowo 1.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.
Download (5MB)
![Chapter 3 [thumbnail of Chapter 3]](http://repository.uph.edu/style/images/fileicons/text.png)
BAB III_Gidon Bilowo 1.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.
Download (1MB)
![Chapter 4 [thumbnail of Chapter 4]](http://repository.uph.edu/style/images/fileicons/text.png)
BAB IV_Gidon Bilowo 1.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.
Download (4MB)
![Chapter 5 [thumbnail of Chapter 5]](http://repository.uph.edu/style/images/fileicons/text.png)
BAB V_Gidon Bilowo 1.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.
Download (418kB)
![Bibliography [thumbnail of Bibliography]](http://repository.uph.edu/style/images/fileicons/text.png)
Daftar Pustaka_Gidon Bilowo 1.pdf
Available under License Creative Commons Attribution Non-commercial Share Alike.
Download (1MB)
![Appendices [thumbnail of Appendices]](http://repository.uph.edu/style/images/fileicons/text.png)
Lampiran (Appendices)_Gidon Bilowo 1_watermark.pdf
Restricted to Repository staff only
Available under License Creative Commons Attribution Non-commercial Share Alike.
Download (3MB)
Abstract
This study aims to optimize the orientation of 3D printing to minimize support
material, reduce printing time, and improve surface quality. The Non-dominated Sorting
Genetic Algorithm II (NSGA-II) is employed to generate a Pareto front that represents the
trade-offs between conflicting objectives in multi-objective optimization. However,
NSGA-II has limitations in exploring the solution space and tends to stagnate in local
optima when parameters such as crossover and mutation probabilities are applied statically.
To address this issue, NSGA-II is integrated with Q-Learning to dynamically adjust
algorithm parameters based on rewards. In the initial stages, Q-Learning enhances
population diversity for broader exploration, while in the later stages, it focuses on
exploiting solutions close to the Pareto front.
The results show that the combination of NSGA-II and Q-Learning improves the
quality of the Pareto front compared to standalone NSGA-II. Experiments demonstrate
efficiency improvements across all objective functions: support material (2.1%), printing
time (3.8%), and surface roughness (1.9%). Validation using slicer software confirms that
the generated Pareto-optimal solutions are relevant and practically applicable in 3D
printing technology based on Fused Deposition Modeling (FDM).
This study contributes to the development of multi-objective optimization
methods combining NSGA-II and Q-Learning in the context of additive manufacturing. By
leveraging the principle of Pareto optimality, this approach provides a more effective
solution to balance material savings, time efficiency, and print quality. / Penelitian ini bertujuan untuk mengoptimalkan orientasi pencetakan 3D guna
meminimalkan material penopang, mengurangi waktu cetak, dan meningkatkan kualitas
permukaan. Algoritma Non-dominated Sorting Genetic Algorithm II (NSGA-II) digunakan
untuk menghasilkan Pareto front, yang mencerminkan trade-off antara tujuan yang saling
bertentangan dalam optimalisasi multiobjektif. Namun, NSGA-II memiliki keterbatasan
dalam eksplorasi ruang solusi dan cenderung stagnan pada solusi lokal ketika parameter
seperti probabilitas crossover dan mutasi diterapkan secara statis. Untuk mengatasi hal ini,
NSGA-II diintegrasikan dengan Q-Learning guna menyesuaikan parameter algoritma
secara dinamis berdasarkan reward. Pada tahap awal, Q-Learning meningkatkan diversitas
populasi untuk eksplorasi yang lebih luas, sedangkan pada tahap akhir, fokus diarahkan
pada eksploitasi solusi yang mendekati Pareto front.
Hasil penelitian menunjukkan bahwa kombinasi NSGA-II dan Q-Learning mampu
meningkatkan kualitas Pareto front dibandingkan NSGA-II murni. Uji coba menunjukkan
peningkatan efisiensi pada semua fungsi objektif: material penopang (2,1%), waktu
pencetakan (3,8%), dan kekasaran permukaan (1,9%). Validasi menggunakan perangkat
lunak slicer mengonfirmasi bahwa solusi Pareto-optimal yang dihasilkan relevan dan dapat
diterapkan dalam teknologi cetak 3D berbasis Fused Deposition Modeling (FDM).
Penelitian ini memberikan kontribusi pada pengembangan metode optimalisasi
multiobjektif berbasis kombinasi algoritma NSGA-II dan Q-learning dalam konteks
manufaktur aditif. Dengan memanfaatkan prinsip Pareto optimality, pendekatan ini
menawarkan solusi yang lebih efektif untuk menyeimbangkan penghematan material,
efisiensi waktu, dan kualitas hasil cetak.
Item Type: | Thesis (Masters) |
---|---|
Creators: | Creators NIM Email ORCID Bilowo, Gidon NIM01679220012 gidon.bilowo@gmail.com UNSPECIFIED |
Contributors: | Contribution Contributors NIDN/NIDK Email Thesis advisor Hardjono, Benny NIDN0404086401 UNSPECIFIED |
Uncontrolled Keywords: | NSGA-II ; Pareto Front ; Q-Learning ; FDM ; Optimalisasi Multiobjektif ; Orientasi 3D Printing |
Subjects: | T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7885-7895 Computer engineering. Computer hardware |
Divisions: | University Subject > Current > Faculty/School - UPH Karawaci > School of Information Science and Technology > Master of Informatics Current > Faculty/School - UPH Karawaci > School of Information Science and Technology > Master of Informatics |
Depositing User: | Phillips Iman Heri Wahyudi |
Date Deposited: | 22 Feb 2025 09:15 |
Last Modified: | 22 Feb 2025 09:15 |
URI: | http://repository.uph.edu/id/eprint/67188 |