"Development of an agent-based demographic model of Russia and its supercomputer implementation"
Makarov V.L., Bakhtizin A.R., Sushko E.D., and Sushko G.B.

The application of the agent-based modeling approach to the problem of natural human migration is considered. A demographic model of Russia is presented. This model takes into account the administrative division of Russia and simulates the processes of fertility, mortality and migration on the basis of modeling the behavior of individual members of the artificial society. In order to simulate the behavior of the artificial society as a whole, it is necessary to perform numerical experiments with the number of agents up to 109 and to use supercomputer technologies. In such experiments, an important problem is the implementation of an optimal automatic distribution of agents across the cluster processors. The application of model decomposition using the METIS algorithm with consideration of the main features of the agent model is shown. The obtained numerical results are discussed.

Keywords: agent-based modeling, demography, supercomputing technologies, graph-based decomposition using METIS

  • Makarov V.L. – Central Economics and Mathematics Institute, Russian Academy of Sciences; Nakhimovskii prospekt 47, Moscow, 117418, Russia; Dr. Sci., Professor, Academician of Russian Academy of Sciences, Scientific Supervisor, e-mail: makarov@cemi.rssi.ru
  • Bakhtizin A.R. – Central Economics and Mathematics Institute, Russian Academy of Sciences; Nakhimovskii prospekt 47, Moscow, 117418, Russia; Dr. Sci., Professor, Corresponding Member of Russian Academy of Sciences, Director, e-mail: albert.bakhtizin@gmail.com
  • Sushko E.D. – Central Economics and Mathematics Institute, Russian Academy of Sciences; Nakhimovskii prospekt 47, Moscow, 117418, Russia; Ph.D., Leading Scientist, e-mail: sushko_e@mail.ru
  • Sushko G.B. – Central Economics and Mathematics Institute, Russian Academy of Sciences; Nakhimovskii prospekt 47, Moscow, 117418, Russia; Ph.D., Senior Scientist, e-mail: gennady.sushko@gmail.com