Extremal optimization

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

PMACO: A pheromone-mutation based ant colony optimization for traveling salesman problem

Abstract: Traveling salesman problem (TSP) is a popular routing problem, which is a sub-problem of many application domains such as transportation, network communication, vehicle routing and integrated circuits designs. Among many approaches which have been proposed for TSP so far, evolutionary algorithms effectively applied to solve it, and attempt to avoid trapping in local minima. In this paper, a hybrid approach (named PMACO) is proposed for TSP, which is based on ant colony optimization (ACO) and the mutation operators in genetic algorithm (GA). In this manner, in the each iteration a hybrid mutation scheme is applied to search the neighborhood area of the solution corresponding with the two best ants (the best current ant, and the total best ant found so far). The population is divided to two parts: the first is regenerated according to the pheromone rule in ACO, and the last is generated by applying the proposed hybrid mutation scheme. Also in this paper, the initial population was not considered randomly; it was generated according to nearest neighborhood rule, which force the ants toward better solutions, and leads to reduce the running time. Finally, we compared the performance of PMACO, with ACO and GA separately. For each algorithm, the various parameters and operators were tested, and the bests were chosen to tune that. Experimental results were carried out on benchmarks from the TSPLIB for validation.

Genetical Swarm Optimization (GSO): a class of Population-based Algorithms for Antenna Design

Abstract: In this paper a new effective optimization algorithm called Genetical Swarm Optimization (GSO) is presented. This is an hybrid algorithm developed in order to combine in the most effective way the properties of two of the most popular evolutionary optimization approaches now in use for the optimization of electromagnetic structures, the Particle Swarm Optimization (PSO) and Genetic Algorithms (GA). This algorithm is essentially, as PSO and GA, a population-based heuristic search technique, which can be used to solve combinatorial optimization problems, modeled on the concepts of natural selection and evolution (GA) but also based on cultural and social rules derived from the analysis of the swarm intelligence and from the interaction among particles (PSO). Preliminary analyses are here presented with respect to the other optimization techniques dealing with a classical optimization problem. The optimized design of a printed reflectarray antenna is finally reported with numerical results.

A Heuristic for Boolean Optimization Problems

Abstract: A heuristic method is proposed for the solution of a large class of binary optimization problems, which includes weighted versions of the set covering, graph stability, partitioning, maximum satisfiability, and numerous other problems. The reported substantial computational experiments amply demonstrate the efficiency of the proposed method.

Reliability based generator maintenance scheduling using hybrid evolutionary approach

Abstract: Maintenance scheduling of generating units is very important for the reliable operation of units. This paper presents a hybrid evolutionary algorithm to tackle the Generator Maintenance Scheduling (GMS) problem. The paper assumes a reliability objective function for the GMS problem. A new local search method which is derived from Extremal Optimization (EO) and Genetic Algorithm (GA) is presented. The proposed method, Hill Climbing Technique (HCT) and EO are applied to different location in GA. The selected locations are initial population, mating pool, in the offspring created by the crossover operator and in the offspring created by the mutation operator. Combination of the proposed method with HCT is also applied to the selected locations in the GA. The discussed methods are applied to a test case study and implementation and performance of the applied methods are presented. The obtained results show that the proposed method in combination with HCT yields the best results in comparison with other local search methods.

Using Prediction to Improve Combinatorial Optimization Search

Abstract: This paper describes a statistical approach to improving the performance of stochastic search algorithms for optimization. Given a search algorithm A, we learn to predict the outcome of A as a function of state features along a search trajectory. Predictions are made by a function approximator such as global or locally-weighted polynomial regression; training data is collected by Monte-Carlo simulation. Extrapolating from this data produces a new evaluation function which can bias future search trajectories toward better optima. Our implementation of this idea, STAGE, has produced very promising results on two large-scale domains.