PENGEMBANGAN metaheuristicOpt: R PACKAGE UNTUK OPTIMASI DENGAN MENGGUNAKAN ALGORITMA POPULATION BASED METAHEURISTIC

Muhammad Bima Adi Prabowo, - (2019) PENGEMBANGAN metaheuristicOpt: R PACKAGE UNTUK OPTIMASI DENGAN MENGGUNAKAN ALGORITMA POPULATION BASED METAHEURISTIC. S1 thesis, Universitas Pendidikan Indonesia.

[img] Text
S_KOM_1503639_Title.pdf

Download (550kB)
[img] Text
S_KOM_1503639_Chapter1.pdf

Download (177kB)
[img] Text
S_KOM_1503639_Chapter2.pdf
Restricted to Staf Perpustakaan

Download (1MB)
[img] Text
S_KOM_1503639_Chapter3.pdf

Download (283kB)
[img] Text
S_KOM_1503639_Chapter4.pdf
Restricted to Staf Perpustakaan

Download (1MB)
[img] Text
S_KOM_1503639_Chapter5.pdf

Download (162kB)
[img] Text
S_KOM_1503639_Appendix.pdf
Restricted to Staf Perpustakaan

Download (3MB)
Official URL: http://repository.upi.edu

Abstract

Optimasi diterapkan diberbagai disiplin ilmu seperti teknik sipil, teknik mekanika, ekonomi, teknik elektro dan lain-lain. Karena optimasi diterapkan diberbagai disiplin ilmu maka optimasi sangatlah penting. Banyak sekali pendekatan yang dilakukan dalam melakukan optimasi salah satunya adalah population based metaheuristic. Di bahasa pemrograman R terdapat package optimasi menggunakan algoritma population based metaheuristic yaitu “metaheuristicOpt”. Algoritma-algoritma pada R package “metaheuristicOpt” memiliki dua kelemahan yaitu kompleksitas yang tinggi dan hyperparameter yang sedikit. Tujuan penelitian ini adalah mengembangkan R package “metaheuristicOpt” dengan menambahkan 10 algoritma baru yaitu clonal selection algorithm, differential evolution, shuffled frog leaping, cat swarm optimization, artificial bee colony algorithm, krill herd algorithm, cuckoo search, bat algorithm, gravitational based search dan black hole optimization untuk menutupi kelemahan algoritma sebelumnya. Dalam menambahkan algoritma ini kami menjaga konsistensi arsitektur package tersebut. Untuk menganalisis performa dari algoritma baru yang ditambahkan setiap fungsi diuji menggunakan 13 fungsi uji. Yang menjadi tolok ukur eksperimen adalah fitness dan waktu eksekusi. Berdasarkan eksperimen yang dilakukan beberapa algoritma baru memiliki kecepatan eksekusi yang lebih cepat dari algoritma sebelumnya dan beberapa algoritma baru juga memiliki fitness yang lebih baik dari algoritma sebelumnya. Optimization is applied in various scientific disciplines such as civil engineering, mechanical engineering, economics, electrical engineering and others. Because optimization is applied in various disciplines, optimization is very important. There are a lot of approaches used to optimize one of them is population based metaheuristic. In the R programming language there is an optimization package using the population based metaheuristic algorithm, namely "metaheuristicOpt". Algorithms in the R package "metaheuristicOpt" have two disadvantages: high complexity and few hyperparameters. Our goal is to develop the "metaheuristicOpt" package by adding 10 new algorithms namely clonal selection algorithm, differential evolution, shuffled frog leaping, cat swarm optimization, artificial bee colony algorithm, krill herd algorithm, cuckoo search, bat algorithm, gravitational based search and black hole optimization to cover up the weaknesses of the previous algorithm. In adding of these algorithms we maintain the consistency of the package architecture. To analyse performance of the new algorithm added to each function of the algorithm, experiments were carried out using 13 test functions. The benchmarks of the experiment are fitness and execution time. Based on experiments some of new algorithms added have a faster execution speed and better fitness than the previous algorithms.

Item Type: Tugas Akhir,Skripsi,Tesis,Disertasi (S1)
Additional Information: No. Panggil: S KOM MUH p-2019 ; Pembimbing: I. Lala Septem Riza, II. Enjun Junaeti ; NIM: 1503639
Uncontrolled Keywords: Metaheuristic algorithm, Optimization, R programing language, Software libarary, Swarm intelligence.
Subjects: L Education > L Education (General)
Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Divisions: Fakultas Pendidikan Matematika dan Ilmu Pengetahuan Alam > Program Studi Ilmu Komputer
Depositing User: Muhammad Bima Adi Prabowo
Date Deposited: 06 Dec 2019 02:46
Last Modified: 06 Dec 2019 02:46
URI: http://repository.upi.edu/id/eprint/38337

Actions (login required)

View Item View Item