Key points are not available for this paper at this time.
The composite architecture of the multi-agent bionic search system is proposed to solve the general distribution problem based on swarm intelligence and genetic evolution. Three approaches to the construction of such architecture are considered. The connecting link of this approach is a unified data structure that describes the solution of the problem in the form of a chromosome. The new principles and methods of coding and decoding of chromosomes for the representation of the general distribution problem considered in this paper exclude incorrect solutions, are distinguished by simplicity and linear estimates of temporal and spatial complexity. A modified paradigm of the particle swarm method is proposed. To organize the swarm movement of particles in hyperspace of solutions, a directed mutation operator has been developed. Experiments have shown that the quality of the solutions in the hybrid algorithm is 10 to 15% better than the genetic and swarm algorithms. The overall estimate of time complexity for any hybridization approach does not exceed the estimate of the time complexity of the genetic algorithm and lies within the range О(n2)–О(n3).
Лебедев et al. (Fri,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: