A Framework for Adapting Population-Based and Heuristic Algorithms for Dynamic Optimization Problems

Abstract

In this paper, a general framework was presented to boost heuristic optimization algorithms based on swarm intelligence from static to dynamic environments. Regarding the problems of dynamic optimization as opposed to static environments, evaluation function or constraints change in the time and hence place of optimization. The subject matter of the framework is based on the variability of the number of algorithm individuals and the creation of feasible subspaces appropriate to environmental conditions. Accordingly, to prevent early convergence along with the increasing speed of local search, the search space is divided with respect to the conditions of each moment into subspaces labeled as focused search area, and focused individuals are recruited to make search for it. Moreover, the structure of the design is in such a way that it often adapts itself to environmental condition, and there is no need to identify any change in the environment. The framework proposed for particle swarm optimization algorithm has been implemented as one of the most notable static optimization and a new optimization method referred to as ant lion optimizer. The results from moving peak benchmarks (MPB) indicated the good performance of the proposed framework for dynamic optimization. Furthermore, the positive performance of practices was assessed with respect to real-world issues, including clustering for dynamic data.