Optimasi Unequal Area Facility Layout Problem (UAFLP) Menggunakan Metode Hibrida Population-Based Simulated Annealing Dan Ant Colony Optimization (PSA-ACO)

Oktaviani, Aini (2025) Optimasi Unequal Area Facility Layout Problem (UAFLP) Menggunakan Metode Hibrida Population-Based Simulated Annealing Dan Ant Colony Optimization (PSA-ACO). Masters thesis, Institut Teknologi Sepuluh Nopember.

Text 6026231012-Master_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (4MB) | Request a copy

Abstract

Jumlah perusahaan manufaktur di Indonesia terus meningkat, mencapai 32.193 pada tahun 2023. Pertumbuhan ini memicu persaingan yang semakin ketat dan menuntut efisiensi operasional, salah satunya melalui perancangan tata letak fasilitas yang optimal. Facility Layout Design (FLD) berperan penting dalam efisiensi operasional, namun permasalahan desain tata letak fasilitas atau Facility Layout Problem (FLP) sering diabaikan karena membutuhkan waktu dan biaya besar. FLP bertujuan merancang tata letak fasilitas secara efisien dengan memperhatikan berbagai batasan operasional. Salah satu varian FLP yang kompleks adalah Unequal Area Facility Layout Problem (UAFLP), yang mempertimbangkan area fasilitas yang tidak sama. Berbagai metode meta-heuristik seperti Simulated Annealing (SA) dan Ant Colony Optimization (ACO) telah digunakan untuk menyelesaikan masalah ini. Modifikasi SA yang disebut Population-Based Simulated Annealing (PSA) mengubah pendekatan solusi tunggal menjadi berbasis populasi, dan terbukti unggul pada berbagai fungsi benchmark. Namun, belum ditemukan penerapan PSA untuk menyelesaikan UAFLP secara langsung. Sementara itu, ACO juga telah banyak digunakan untuk menyelesaikan permasalahan FLP, terutama yang melibatkan kendala kompleks dan lingkungan dinamis. Dalam penelitian ini, diusulkan metode hibrida yang menggabungkan PSA dan ACO untuk mengatasi tantangan UAFLP. Pendekatan ini bertujuan meningkatkan kualitas solusi dengan menghindari konvergensi prematur dan memperoleh tata letak yang lebih optimal. Hasil pengujian pada 15 dataset menunjukkan bahwa metode Hybrid PSA–ACO memberikan performa unggul, menghasilkan solusi dengan material handling cost minimum pada 13 dari 15 dataset. Selain itu, metode ini memiliki nilai rata-rata dan standar deviasi yang lebih baik di sebagian besar kasus. Walaupun waktu komputasi lebih tinggi, metode ini menghasilkan solusi yang lebih stabil dan optimal dibandingkan PSA atau ACO secara individu. Dibandingkan penelitian sebelumnya, metode ini juga menunjukkan performa yang kompetitif, sehingga dinilai efektif dan potensial dalam menyelesaikan UAFLP.
===========================================================================================================================================
The number of manufacturing companies in Indonesia continues to grow, reaching 32,193 in 2023. This growth intensifies competition and demands greater operational efficiency, one of which can be achieved through optimal facility layout design. Facility Layout Design (FLD) plays a crucial role in operational efficiency, yet the Facility Layout Problem (FLP) is often overlooked due to the significant time and cost required. FLP aims to design facility layouts efficiently while considering various operational constraints. One complex variant of FLP is the Unequal Area Facility Layout Problem (UAFLP), which takes into account the unequal area of facilities. Several metaheuristic methods such as Simulated Annealing (SA) and Ant Colony Optimization (ACO) have been widely applied to solve this problem. A notable modification of SA is Population-Based Simulated Annealing (PSA), which transforms the single-solution approach into a population-based algorithm, and has demonstrated superior performance on various benchmark functions. However, no prior research has been found that applies PSA specifically to solve UAFLP. Similarly, ACO and its various modifications have been extensively used to solve facility layout problems, especially those involving complex constraints and dynamic environments. This study proposes a hybrid method that combines PSA and ACO to address the challenges of UAFLP. The goal of this hybrid approach is to improve solution quality by avoiding premature convergence and discovering more optimal layouts. Experimental results on 15 datasets show that the Hybrid PSA–ACO method performs well, producing the lowest material handling cost in 13 out of 15 datasets. Additionally, this method demonstrates better average values and standard deviations in most cases. Although it requires higher computation time, it provides more stable and optimal solutions compared to PSA or ACO individually. Compared to previous studies, the proposed method also shows competitive performance, making it an effective and promising approach for solving UAFLP.

Item Type:	Thesis (Masters)
Uncontrolled Keywords:	Manufaktur, Unequal Area Facility Layout Problem, Population-Based Simulated Annealing, Ant Colony Optimization, Hybrid Population-Based Simulated Annealing–Ant Colony Optimization, Manufacture
Subjects:	Q Science > QA Mathematics > QA9.58 Algorithms T Technology > T Technology (General) > T57.6 Operations research--Mathematics. Goal programming T Technology > T Technology (General) > T57.84 Heuristic algorithms.
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Information System > 59101-(S2) Master Thesis
Depositing User:	Aini Oktaviani
Date Deposited:	24 Jul 2025 03:45
Last Modified:	24 Jul 2025 03:45
URI:	http://repository.its.ac.id/id/eprint/121033

Actions (login required)

View Item