ODMA: a novel swarm-evolutionary metaheuristic optimizer inspired by open source development model and communities

Abstract

In these recent years, several metaheuristic optimization algorithms have been developed to solve complex real-world problems. However, according to the no free lunch theorem, there is no metaheuristic algorithm which has the best suited for solving all optimization problems. Therefore, developing new metaheuristic algorithms is an open problem. Two major categories of metaheuristics are swarm intelligence and evolutionary algorithms. Both of them have their own advantages. In this paper, we present a naturally swarm and evolutionary algorithm in which its particles have memory, collaboration, information sharing, and competition inherited from one specific analogy. So, it has the advantages of both swarm and evolutionary algorithms without complications of hybridizing them. We formulate a new metaheuristic optimization algorithm called open source development model algorithm (ODMA) inspired by open source development model and communities, in such a way that each potential solution is considered as a software, and by evolution of the softwares, better solutions of the function that should be optimized are searched. The main operations of the algorithm include employing features and methods of leading softwares, evolution of leading softwares based on their history and forking from leading softwares. After a detailed formulation and explanation of its implementation, the algorithm is evaluated by 17 well-known benchmark functions, and the results are compared with PSO, ICA, and GA. The results show that ODMA outperforms GA, PSO, and ICA in terms of finding the global optimum and convergence.

EATMR: An Energy-Aware Trust Algorithm Based the AODV Protocol and Multi-path Routing Approach in Wireless Sensor Networks

“…There are many benefits to using both algorithms without the hassle of combining them. The open-source development model algorithm (ODMA) is a novel meta-heuristic algorithm that takes advantage of the combination of both categories [14]. ODMA considers each solution as an open-source software and evolves inspired by the open development model and user communities.…”

“…Approaching a leading software does not guarantee that a promising software will achieve a better position, so promising software may be in a better or worse position in society. In general, the main operations of this algorithm include (1) moving towards leading software, (2) evolving leading software based on its history, and (3) branching out from leading software [14].…”

“…Hitherto, various algorithms for clustering have been developed in WSN, among which evolutionary algorithms are very popular [1,5]. In this paper, ODMA [14] is used as a novel evolutionary algorithm for clustering. In the proposed scheme, in addition to the formation of clusters, the optimal number of clusters as well as CHs is determined.…”

A Hybrid Leader Selection Strategy for Many-Objective Particle Swarm Optimization

“…Then, all external archive members are sorted according to their fitness values using the selected objective. Finally, one archive member is randomly selected from the top 5% of the sorted archive [48], [49] and the selected member becomes the global leader of the particle. This 5% is to weight the focus away from the compromise solutions.…”

Modified Vortex Search Algorithm for Real Parameter Optimization

“…In [10][11][12][13][14][15], a number of recently proposed metaheuristics can be found. All of these metaheuristics have certain characteristics and thus each one may be more successful on a certain optimization problem when compared to the others.…”