Optimalisasi penyeimbangan lini perakitan menggunakan metode genetic algorithm dibandingkan metode heuristik = Assembly line balancing optimization using genetic algorithm method compared to heuristic method

Frederick, Nathaniel (2024) Optimalisasi penyeimbangan lini perakitan menggunakan metode genetic algorithm dibandingkan metode heuristik = Assembly line balancing optimization using genetic algorithm method compared to heuristic method. Bachelor thesis, Universitas Pelita Harapan.

	Text (Title) Title.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (87kB)
	Text (Abstract) Abstract.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (221kB)
	Text (ToC) ToC.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (357kB)
	Text (Chapter1) Chapter1.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (529kB)
	Text (Chapter2) Chapter2.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (4MB)
	Text (Chapter3) Chapter3.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (717kB)
	Text (Chapter4) Chapter4.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (5MB)
	Text (Chapter5) Chapter5.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (419kB)
	Text (Bibliography) Bibliography.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (569kB)
	Text (Appendices) Appendices.pdf Restricted to Repository staff only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (4MB)

Abstract

Perkembangan industri dari setiap periode memiliki karakteristik spesifik dalam mengubah perilaku dan pendekatan sektor industri. Pada Industri 4.0, karakteristik tersebut ditandai dengan kemunculan Internet of Things and Services serta Cyber-Physical System (CPS). Salah satu penerapannya dalam industri mencakup penyeimbangan lini perakitan. Tujuan penelitian ini adalah melakukan praktik lapangan dan mengamati kinerja metode penyeimbangan lini perakitan, termasuk genetic algorithm dan heuristik, seperti shortest processing time, largest candidate rule, dan ranked positional weight. Metode yang digunakan adalah observasi perakitan rancangan awal. Data yang dikumpulkan adalah waktu perakitan, suhu, dan kelembapan ruangan. Data yang telah diseleksi akan diolah menjadi waktu standar dan digunakan untuk menyeimbangkan lini perakitan. Solusi yang ditawarkan akan dilakukan praktek lapangan dan simulasi produksi. Hasil penelitian menunjukkan metode genetic algorithm lapangan memiliki performa yang lebih baik dari metode heuristik lapangan dalam parameter simulasi produksi dan indeks kelancaran. Metode genetic algorithm memiliki jumlah produksi 183 buah dan indeks kelancaran 47 detik dibandingkan metode rancangan awal 177 buah dan indeks kelancaran 52 detik serta metode heuristik tertinggi shortest processing time 158 buah dan indeks kelancaran terendah 53 detik. Hal ini menunjukkan bahwa metode genetic algorithm dapat diterapkan dengan baik dalam praktiknya. Meskipun metode genetic algorithm menunjukkan kinerja yang baik, terdapat kekurangan pada parameter efisiensi lini dan waktu menganggur yang berada pada urutan ketiga. Kekurangan lain dari metode genetic algorithm adalah solusi yang dihasilkan memiliki kemungkinan lebih besar dari waktu baku satu siklus. Faktor-faktor tersebut adalah parameter yang digunakan seperti jumlah iterasi, persentase mutasi, jumlah populasi, dan jumlah stasiun. Tetapi, metode genetic algorithm dapat menghasilkan beberapa solusi terbaik dan dapat dipertimbangkan dengan parameter performa lini. / The industrial development of each period has specific characteristics in changing the behavior and approach of the industrial sector. In Industry 4.0, these characteristics are indicated by the appearance of the Internet of Things and Service along with the Cyber-Physical System (CPS). One of the applications in the industry includes assembly line balancing. The purpose of this research is to conduct field practice and observe the assembly line balancing performance, including genetic algorithm and heuristic method, such as shortest processing time, largest candidate rule, and ranked positional weight. The method that is being used is observation of the initial method. The data that will be collected are assembly time, temperature, and humidity. Data that has passed the selection will be processed into standardized time and used for assembly line balancing. The suggested solution will be implemented through field practice and simulated production. The results of the study show that genetic algorithms in the field have a better performance than the heuristic methods in the parameters for simulated production and smoothness index. The genetic algorithm has a production amount of 183 pieces and a smoothness index of 47 seconds compared to initial methods with 177 pieces and a smoothness index of 52 seconds as well as the highest heuristic method shortest processing time with 158 pieces and the lowest smoothness index with 53 seconds. This indicates that genetic algorithms could be implemented well in the field. Although the genetic algorithm shows good performance, there are shortcomings in the parameters for the line efficiency and balance delay which are in third place. Another drawback of the genetic algorithm is the solution that will be provided has a high chance of being bigger than the standardized time for one cycle. The contributing factors are the parameters used such as number of iterations, mutation percentage, number of populations, and number of stations. However, the genetic algorithm can provide several best solutions and can be considered with line performance parameter.

Actions (login required)

View Item