scispace - formally typeset
Search or ask a question
Topic

Extremal optimization

About: Extremal optimization is a research topic. Over the lifetime, 1168 publications have been published within this topic receiving 104943 citations.


Papers
More filters
Journal ArticleDOI
TL;DR: Numerical optimization was used to find the optimal parameter values for simulated annealing (SA), threshold accepting (TA), great deluge (GD), tabu search (TS), genetic algorithm (GA) and ant colony optimization (AC) when they are used for combinatorial optimization in forest planning.
Abstract: Heuristic methods are commonly used in complicated spatial forest planning problems to find the best combination of management alternatives for stands. The performance of heuristic methods depends on the parameters that guide their search processes. This study used numerical optimization to find the optimal parameter values for simulated annealing (SA), threshold accepting (TA), great deluge (GD), tabu search (TS), genetic algorithm (GA) and ant colony optimization (AC) when they are used for combinatorial optimization in forest planning. Ant colony optimization was implemented using the Max–Min Ant System, which was applied for the first time to forest planning problem. Solutions found by different heuristic methods for a non-spatial and a spatial forest planning problem were compared in a situation where the search time was restricted. The comparisons revealed that SA and TA were the best methods for fast search in both non-spatial and spatial problems. GA and AC were the least satisfactory methods, and GD and TS were between the best and the worst heuristics. The main reason for the poor performance of GA and AC was their slow search process. Differences between heuristic methods decreased when the allowed search time increased.

30 citations

Proceedings ArticleDOI
01 Mar 1992
TL;DR: The parallel gestetic algorithm performed consistently better than the parallel simulated annealing algorithm in all of the cases tested and in every case the edge recombination crossover function was superior.
Abstract: There are many combinatorial optimization problems for which there exists no director efficient method of solution. Simtdated annealtig (SA) and genetic algorithms (GA) are two promisirtg techniques for solving large optimization problems. The authorx have developed a parallel simulated annealing algorithm and a parallel genetic algorithm for a hypercube multiprocessor system. To compare the performance of these algorithms, we investigated two representative combinatorial optimization problems, the Traveling Salesman Problem (TSP) and the onedimensionrd Package Placement Problem (PPP). The parallel gestetic algorithm performed consistently better than the parallel simulated annealing algorithm in all of the cases tested. In addition, we tested five crossover functions on the sequential genetic algorithm for the Package Placement Problem and determined in every case the edge recombination crossover function was superior. There are some significant differences between genetic algorithms and simulated annealing that may account for the superior performanw of the parallel genetic algorithm for these ty~s of problems. We found it fairly easy to fine tune the parameters that drive a parallel GA for near optimal performance (population size, migration rate, and migration interval) compared to the parameters that drive a parallel simulated annealing algorithm. Furthermore, our parallel genetic algorithm is more mature than our newly developed parallel simulated annealing algorithm, Several future enhancements to the parallel simulated annealing algorithm are presented. “ Research partially supported by OCAST Grant ARO-038 md Sun Microsystems, Inc. Permission to copy without fee all or part of this material is grantad providad that the copias ara not mada or distributed for direct commercial advantage, tha ACM copyright notica and tha title of the publication and its date appear, and notice is given that copying ia by permission of tha Association for Computing Machinery. To copy otherwise, or to rapublish, requires a fae and/or specific permission. 01992 ACM O-89791-502-X/9210002/1031 ...$l .50 INTRODUCTION In the fields of Operations Research and Artificial Intelligence, there are many combinatorial optimization problems for which there exista no direct or efficient method of solution. Researchers have recently become interested in solvrng large combinatorial optimization problems using large numbers of processors. These algorithms offer optimal or near optimal solutions to many important optimization problems [14, 15, 18]. On a sequential machine for large n, timings from 0(n3) time to exponential time are commonly required. Since this is too expensive, leas time cmrtaumirtgalgorithms have been developed such as greedy algorithms. These less expensive algorithms rapidly find a “good” solution, but not an optimal solution or even near optimal solution. With the advent of parallel processors, new opportunities open up for effective solutions of combinatorial problems that were not available only a few years ago. Fmdmg a solution to many combinatorial optimization problems requirea an organized search through the problem space. An unguided search is extremely inefficient since many of these problems are NP-complete. Genetic algorithms and simulated annealing are promising tdtniques for solving large optimization problems. In this paper we look at two representative combinatorial optimization problems, the Traveling Salesman Problem (TSP) and the one-dimensional Package Placement Problem (PPP). Both problems were implemented and solved using a parallel simulated anneaIing algorithm and a parallel genetic algorithm developed by the authors. The rest of the paper is presented as follows. In Section 2 we review the Traveling Salesman Problem and the Package Placement Problem. In Section 3 the fundamentals of simulated annealing are reviewed. In Section 4 a parallel simulated annealing algorithm is described. In Section 5 the fundamentals of genetic algorithms are reviewed and in Section 6 a parallel implementation is described. Results and conclusions are presented in Scction 7. Future research issues are given in Section 8.

30 citations

Journal ArticleDOI
TL;DR: This paper presents a tabu search approach to a combinatorial optimization problem, in which the objective is to maximize the production throughput of a high-speed automated placement machine.
Abstract: Combinatorial optimization represents a wide range of real-life manufacturing optimization problems. Due to the high computational complexity, and the usually high number of variables, the solution of these problems imposes considerable challenges. This paper presents a tabu search approach to a combinatorial optimization problem, in which the objective is to maximize the production throughput of a high-speed automated placement machine. Tabu search is a modern heuristic technique widely employed to cope with large search spaces, for which classical search methods would not provide satisfactory solutions in a reasonable amount of time. The developed TS strategies are tailored to address the different issues caused by the modular structure of the machine.

29 citations

Journal ArticleDOI
TL;DR: A novel optimization algorithm based on competitive behavior of various creatures such as birds, cats, bees and ants to survive in nature and is an efficient method in finding the solution of optimization problems.
Abstract: This paper presents a novel optimization algorithm based on competitive behavior of various creatures such as birds, cats, bees and ants to survive in nature. In the proposed method, a competition is designed among all aforementioned creatures according to their performances. Every optimization algorithm can be appropriate for some objective functions and may not be appropriate for another. Due to the interaction between different optimization algorithms proposed in this paper, the algorithms acting based on the behavior of these creatures can compete each other for the best. The rules of competition between the optimization methods are based on imperialist competitive algorithm. Imperialist competitive algorithm decides which of the algorithms can survive and which of them must be extinct. In order to have a comparison to well-known heuristic global optimization methods, some simulations are carried out on some benchmark test functions with different and high dimensions. The obtained results shows that the proposed competition based optimization algorithm is an efficient method in finding the solution of optimization problems.

29 citations

Book ChapterDOI
01 Jan 2004
TL;DR: This chapter shows how the CE method can be easily transformed into an efficient and versatile randomized algorithm for solving optimization problems, in particular combinatorial optimization problems.
Abstract: In this chapter we show how the CE method can be easily transformed into an efficient and versatile randomized algorithm for solving optimization problems, in particular combinatorial optimization problems

29 citations


Network Information
Related Topics (5)
Genetic algorithm
67.5K papers, 1.2M citations
85% related
Optimization problem
96.4K papers, 2.1M citations
81% related
Artificial neural network
207K papers, 4.5M citations
80% related
Cluster analysis
146.5K papers, 2.9M citations
80% related
Fuzzy logic
151.2K papers, 2.3M citations
78% related
Performance
Metrics
No. of papers in the topic in previous years
YearPapers
20232
202213
20217
20209
201922
201815