China
Africa
Explore chapters and articles related to this topic
Firefly Algorithm
Published in A Vasuki, Nature-Inspired Optimization Algorithms, 2020
The fireflies have a natural habit of dividing themselves into groups, and, relating this behavior with our algorithm, it leads to grouping of fireflies around local optima. As the iterations progress, the groups cluster more and more around the optimum regions in the landscape and among them the global optimum can be identified. The parameter tuning gives flexibility in altering the performance of the algorithm as suited for the applications. The change in performance of the algorithm by varying the parameters has been explored [5]. The intermittent search strategy is one of the key components applied here where there is exploration of the landscape using Levy flights and intense exploitation of the search around regions of optimality [6]. Local search concentrates around regions where the global optimum is likely to be found. More exploration requires more iterations and hence convergence occurs later, whereas more exploitation requires fewer iterations which may lead to the global optimum or it could lead to premature convergence at a local optimum point. The idealized rules of the firefly algorithm have been combined with Levy flights to form the Levy-flight firefly algorithm (LFA) in [2]. There is a vast literature available on the firefly algorithm, its variants, and applications with hundreds of papers being published. In this chapter, a few variants and applications of FA have been discussed.
Dialectics of Nature: Inspiration for Computing
Published in Nazmul Siddique, Hojjat Adeli, Nature-Inspired Computing, 2017
Local search is a meta-heuristic method for solving search and optimization problems. Local search algorithms move in the solution space by applying small changes to a candidate solution until it satisfies a criterion of optimality. The local search allows determining the extremum (minimum or maximum) of a function f:R→R over a closed interval, say, [a,b]. It is assumed that the objective function f is unimodal, meaning f has only one local minimum or maximum on the interval of [a,b]. The local search methods have some common features such as local search, keep track of the single current state, move only to neighboring states, and use very little memory, and in this way, they very often are able to find reasonable solutions in large and infinite state spaces.
Cyber-attack feature processing approach based on MA-LSSVM
Published in Xiaoling Jia, Feng Wu, Electromechanical Control Technology and Transportation, 2017
Local search is an important concept of the memetic algorithm. It has an important influence on the performance of the algorithm. The basic concept of local search is based on the greedy algorithm, while constantly looking for a better solution at the current solution neighborhood. It is the process to select the excellent individual at local areas to enhance algorithm performance and individual fitness. TS is an optimization method for heuristic search. It has the following advantages: high search efficiency, memory function, and excellent climbing ability. The TS algorithm is adopted to avoid alternate searching through tabu list and corresponding taboo rules. The aspiration criterion is used to avoid taboo excellent status. However, TS performance relies on the initial solution to a great extent (Sun Yanfeng, 2006).
Hybrid meta-heuristic algorithms for optimising a sustainable agricultural supply chain network considering CO2 emissions and water consumption
Published in International Journal of Systems Science: Operations & Logistics, 2023
Fariba Goodarzian, Davood Shishebori, Farzad Bahrami, Ajith Abraham, Andrea Appolloni
In a hybrid algorithm, two or more algorithms are collectively and cooperatively solve a predefined problem (Talbi, 2015). In some hybrids, one algorithm may be incorporated as a sub-algorithm to locate the optimal parameters for another algorithm, while in other cases, different components of algorithms such as mutation and crossover are used to improve another algorithm in the hybrid structure (Blum & Roli, 2008). With regards to this nature, hybrid meta-heuristic algorithms can loosely be divided into two categories: Unified objective hybrids. Under this category, all sub-algorithms are utilised to solve the same problem directly; and different sub-algorithms are used indifferent search stages. Hybrid metaheuristic algorithms with the local search are an atypical example. The global search explores the search space, while the local search is utilised to refine the areas that may contain the global optimum.Multiple objective hybrids. One primary algorithm is used to solve the problem, while the sub-algorithm is employed to tune the parameters for the primary algorithm. In this paper, this category is used for the hybridisation of the meta-heuristic algorithms.
Modelling and solving algorithm for two-stage scheduling of construction component manufacturing with machining and welding process
Published in International Journal of Production Research, 2018
Ronghua Meng, Yunqing Rao, Yun Zheng, Dezhong Qi
Local search is an effective heuristic for solving optimisation problems. In order to emphasise exploitation ability, a local-best search algorithm is embedded into the HSA. We design a local search only for the best harmony vector πbest by applying move operations. The job at position j is moved to position i with i < j, while all jobs at location k with k = i + 1, …, j − 1, are moved to the position i + j − k. We design two stopping criterions. One criterion signifies that the new fitness is better than that of the former, while the other criterion denotes that the cycle index exceeds (m(1) * n(1))/5. The pseudo-code of the local search is as follows:
Hybrid evolutionary optimisation with learning for production scheduling: state-of-the-art survey on algorithms and applications
Published in International Journal of Production Research, 2018
Local search evaluated from early heuristic algorithms, which is represented by Simulated Annealing (SA) and Tabu Search (TS) methods. Kirkpatrick, Gelatt, and Vecchi (1983) proposed the SA method, it evaluated from the combination of physical process of solid annealing and Metropolis rules. Gaafar and Masoud (2005) applied SA with genetic algorithms to scheduling in agile manufacturing. Safaei, Banjevic, and Jardine (2012) proposed a parallel SA with multi-threaded architecture to solve a real bi-objective maintenance scheduling problem with conflicting objectives presents an application of the simulated annealing algorithm to solve level schedules in mixed model assembly line. Moriguchi, Ueki, and Saito (2017) explored an effective implementation of SA to optimise thinning schedules for single even-aged stands, with respect to the reliability of optimality. The advantage of local search is that it can find out the local best solution in shorter time, but is dependent of structural features and initial solutions of problems that makes it easily to local optimum but can’t get the global optimum. Some researchers combined the local search and the population-based optimisation method to improve the solution and computing time of scheduling problem.