You are here

BENZETİMLE ENİYİLEME İÇİN ÇOKLU META-SEZGİSELLER

MULTI META-HEURISTICS FOR SIMULATION OPTIMISATION

Journal Name:

Publication Year:

Keywords (Original Language):

Abstract (2. Language): 
Optimisation with simulation is a happy marriage of two Operations Research methods. In the last decade, the research in this field has accelerated and many researchers have been interested in Simulation Optimization (SO). New techniques have been developed as a result of this interest. Almost all commercial simulation software contains an optimization module. Generally, these modules exploit meta-heuristic methods; however, they do not allow the analyst to choose the method. The performance of meta-heuristic methods may depend on the problem type and therefore the choice of method is crucial. In this paper, we aim to fill this gap and presented an open-source java-based simulationoptimization code library. The library includes three heuristic methods; genetic algorithm, tabu search, simulated annealing, as well as three enumeration based methods; partial and complete enumeration, and a new neighbourhood-based heuristic method. At the simulation side, Simkit, an event-based and open-source simulation library, is used. At the application side, we defined a fictional optimisation problem and used it to compare performances of the algorithms. Our results demonstrated the potential benefits of having multi metaheuristics available in SO.
Abstract (Original Language): 
Benzetim ile eniyileme iki yöneylem araştırması yönteminin mutlu bir evliliğidir. Son on yılda bu alandaki araştırmalar ivme kazanmış ve birçok araştırmacı Benzetimle Eniyileme (BE) alanına ilgi göstermiştir. Bu ilginin sonucu olarak yeni yöntemler de geliştirilmiştir. Hemen hemen bütün ticari benzetim yazılımları bir çeşit BE modülü içermektedir. Genel olarak bu modüller meta-sezgisel yöntemleri kullanmaktadır ancak analizcinin yöntem seçimine izin vermemektedir. Meta sezgisel yöntemlerin problem tipine bağlı olarak performansları değişebilir ve bu nedenle de yöntem seçimi önemlidir. Bu makalede bu açığı doldurmayı hedefliyoruz ve açık kaynak kodlu Java tabanlı bir BE kod kütüphanesi sunuyoruz. Kütüphane üç meta sezgisel; Genetik algoritma, yasaklı arama, simulated annealing, ve üç sıralı aramalı algoritma; parçalı ve tam sıralı aramalı, ve yeni bir komşuluk tabanlı sezgisel yöntemi içermektedir. Benzetim tarafında ise açık kaynak kodlu ve olay tabanlı bir kütüphane olan Simkit kullanılmıştır. Uygulama olarak hayali bir eniyileme problemi tanımlanmış ve algoritmalar karşılaştırılmıştır. Çalışmanın sonuçları BE’de çoklu meta sezgisellere sahip olmanın potansiyel faydalarını göstermiştir.

REFERENCES

References: 

[1] J. April, M. Better, F.Glover, P.J.Kelly (2004) New advances and
applications for marrying simulation and optimization. In Proceedings of the
Winter Simulation Conference, R .G. In-galls et al., Eds. IEEE, Piscataway,
NJ. 80–86.
[2] M.H.Alferaei, A.H.Diabat (2009) A Simulated Annealing Technique
for multi-objective simulation optimization. Applied mathematics and
computation.2009 215 pp 3029-3035
[3] F.Azadivar (1999) Simulation optimization methodologies.
Proceedings of the 1999 Winter Simulation Conference.
[4] A.H.Buss (2001) Basic Event Graph Modeling. Technical Notes,
Simulation News Europe, April 2001: 1-6.
[5] B.Dengiz, C. Alabas (2000) Simulation Optimization Using Tabu
Search. Proceedings of the 2000 Winter Simulation Conference. J. A.
Joines, R. R. Barton, K. Kang, and P. A. Fishwick, eds.
[6] M.C. Fu (2002) Optimization for simulation: Theory vs. practice.
INFORMS Journal on Computing 14:192-215.
[7] M.C. Fu, F.W. Glover, J.April (2005) Simulation Optimization: A
Review, New Develop-mens, and Applications. Proceedings of the 2005
Winter Simulation Conference. M. E. Kuhl, N. M. Steiger, F. B. Armstrong,
and J. A. Joines, eds.
[8] A.Homaifar, C. X. Qi, S.H. Lai (1994) Constrained optimization via
genetic algorithms. SIMULATION 62(4), 242–253.
[9] L. J. Hong, B.L. Nelson (2006) Discrete optimization via simulation
using COMPASS. Operations Research 54:115-129.
[10] L. J.Hong, B.L. Nelson (2007) A framework for locally convergent
random-search algorithms for discrete optimization via simulation. ACM
Transactions on Modeling and Computer Simulation (TOMACS) 17(4), 19.
[11] A.Law, M.McComas (2000) Simulation-based optimization.
Proceedings of the 2000 Winter Simulation Conference.
[12] A.M.Law (2007) Simulation Modeling and Analysis, 4th ed.
McGraw-Hill, New York.
[13] Z.Michalewicz (1992) Genetic Algoritms + Data Structures =
Evolution Programs, Springer Verlag, Berlin Hiedelberg.
Multi Meta-Heuristics For Simulation Optimization
31
[14] G. H.Neddermeijer, G.J. van Oortmarssen, N. Piersma, R. Dekker
(2000) A framework for response surface methodology for simulation
optimization. In WSC ’00: Proceedings of the 32nd conference on Winter
simulation, San Diego, CA, USA, pp. 129–136. Society for Computer
Simulation International.
[15] İ.Sabuncuoglu, E. Tekin (2004) Simulation Optimization: A
Comprehensive Review on Theory and Applications, IIE Transactions, Vol:
36, pp: 1067-1081, 2004.
[16] K.O.Willis, D.F. Jones (2008) Multi-objective simulation
optimization through search heuristics and relational database analysis.
Decision Support Systems 46 (2008) 277–286.
[17] L.Zhao, S. Sen (2006) A comparison of sample-path-based
simulation-optimization and stochastic decomposition for multi-location
transshipment problems, Proceedings of the 37th conference on Winter
simulation, December 03-06, 2006, Monterey, California.

Thank you for copying data from http://www.arastirmax.com