EDA-PSO: A Hybrid Paradigm Combining Estimation of Distribution Algorithms and Particle Swarm Optimization

Author(s):  
Endika Bengoetxea ◽  
Pedro Larrañaga
2012 ◽  
Vol 490-495 ◽  
pp. 524-528
Author(s):  
Jun Fei Zhuo ◽  
Xing He Wu ◽  
Guan Zhao Wu ◽  
Min Yao

Evolutionary computing is one of the important branches in computational intelligence. This paper mainly introduces four new branches of the evolutionary computation, i.e. Gene Expression Programming (GEP), Particle Swarm Optimization (PSO), Differential Evolution (DE) and Estimation of Distribution Algorithms (EDA).


Author(s):  
Amalia Utamima ◽  
Angelia Melani Andrian

Abstrak—Masalah penempatan fasilitas pada garis lurus dikenal sebagai problem Penempatan Fasilitas pada Satu Baris (PFSB). Tujuan PFSB, yang dikategorikan sebagai masalah NP-Complete, adalah untuk mengatur tata letak sehingga jumlah jarak antara pasangan semua fasilitas bisa diminimalisir. Algoritma Estimasi Distribusi (EDA) meningkatkan kualitas solusi secara efisien dalam beberapa pengoperasian pertama, namun keragaman dalam solusi hilang secara pesat ketika semakin banyak iterasi dijalankan. Untuk menjaga keragaman, hibridisasi dengan algoritma meta-heuristik diperlukan. Penelitian ini mengusulkan EDAPSO, algoritma yang terdiri dari hibridisasi EDA dan Particle Swarm Optimization (PSO). Tujuan dari penelitian ini yaitu untuk menguji performa algoritma EDAPSO dalam menyelesaikan PFSB.Kinerja EDAPSO yang diuji dalam 10 masalah benchmark PFSB dan EDAPSO berhasil mencapai solusi optimal.Kata kunci—penempatan fasilitas, algoritma estimasi distribusi, particle swarm optimizationAbstract—The layout positioning problem of facilities on a straight line is known as Single Row Facility Layout Problem (PFSB). Categorized as NP-Complete problem, PFSB aim to arrange the layout so that the sum of distances between all facilities’ pairs can be minimized. Estimation of Distribution Algorithm (EDA) improves the solution quality efficiently in first few runs, but the diversity lost grows rapidly as more iterations are run. To maintain the diversity, hybridization with meta-heuristic algorithms is needed. This research proposes EDAPSO, an algorithm which consists of hybridization of EDA and Particle Swarm Optimization (PSO). The objective of this research is to test the performance of EDAPSO algorithm for solving PFSB.  EDAPSO’s performance is tested in 10 benchmark problems of PFSB and it successfully achieves optimum solution.Keywords— facility layout, estimation distribution algorithm, particle swarm optimization


Sign in / Sign up

Export Citation Format

Share Document