A.R. Bakhtizin's research while affiliated with Russian Academy of Sciences and other places

The problem of standardizing the description of agent-based models is considered. The foreign experience in solving this problem is studied in detail, and the authors' developments are proposed. In the course of this study, a critical analysis of the existing standards (the original ODD protocol and its updates and modifications) was carried out. An adapted translation of the ODD protocol for use in Russian-language publications was proposed. The need to develop a separate standard for describing agent-based models intended for approbation of management decisions was substantiated. The structure of the standard was proposed, and the prospects for its application were described. The scientific novelty and practical significance of this study consists in the development and justification of the first Russian-language standard for describing agent-based models with account for domestic experience, terminology, and the best examples of models’ description, as well as in the proposal to create an open registry of applied models that would provide interaction with potential users of agent-based decision support systems developed by research teams.

Under consideration are issues of using supercomputer technologies in the society development modelling. It has been shown that in the TOP 500 most powerful machines in the world, the USA takes the first place (150 systems), the second place – China (34 systems), Russia has 7 systems; ahead of all in performance is the USA (46%), Russia’s share is 1.4%. The supercomputers main tasks have been listed: deep neural networks customization for managing; composite materials development; drugs modelling; calculation of vehicle bodies optimal geometry and etc. The dynamic development of the situation has been noted in Russia, in particular by the increase of the Lomonosov supercomputer power. The МГУ-27 supercomputer has been distinguished, its main tasks: climate, drugs modelling, as well as the Digital Earth superproject. Three priority areas have been identified for implementing on the supercomputers within the framework of BRICS. Russia has announced climate and demography as priorities. In the CEMI RAS, an agent-based model – a Digital Twin of Russia and a Digital Twin of the Planet on the agent-based approach have been developed. The work has been performed jointly with the colleagues from the PRC. Most of the models made in the CEMIRAS have been implemented on the supercomputers. The work results have been published in the «Scientific Solutions of Complex Economic and Social Tasks Using Supercomputers» monography.

The results of the latest demographic forecasts from the world’s leading specialized centers (United Nations Population Division, the Wittgenstein Center for Demography and Global Human Capital, the Institute for Health Metrics and Evaluation) are considered, demonstrating a certain bias in favor of individual countries and their calculation methods. The second part of this article provides a description of a digital twin of the planet’s demographic system constructed by a Chinese−Russian team and implemented in China’s national supercomputer center. In addition, the results of some calculations carried out using this tool are described.

This article continues a series of works devoted to the creation of large agent-based models, built as an artificial society, and the development of software for their implementation-the MÖBIUS design system for scalable agent-based models. The basic core of the system is a demographic model that simulates the natural movement of the population. A new stage in the development of the work discussed in this article was the creation on the basis of this core of an agent-based model of Russia, which includes families as agents of a new type, hierarchically connected with human agents. In addition, objects of a new type were introduced into the model-projects that provide for the creation in an artificial environment of analogues of complex control actions aimed at stimulating fertility. Developed on the basis of simulating the reaction of individual families to the introduced regional support measures, the model makes it possible to track their impact on key demographic indicators. The agent-based model of Russia was tested on data for a long retrospective period using the example of the launch of maternal capital programs and showed good agreement with official statistics.

The author cites long-term demographic projections of the United Nations, the Institute for Health Measurement and Evaluation (IHME), the Joint Center of the Institute for Mathematical Research of Complex Systems (Lomonosov Moscow State University) and the Wittgenstein Center for Demography and Global Human Capital (IIASA + WCDGHC). According to the forecasts of these organizations, depopulation awaits Russia. The article presents calculations carried out by scientists of the CEMI RAS based on a demographic agent-­based model.The first scenario is associated with the initiation of the process of deurbanization, which involves an increase in the proportion of people living in rural areas, the second — with a gradual increase in the volume of medical services provided. The article also describes scenarios for the development of demographic systems in most countries of the world, obtained on the basis of an agent-­based model created by scientists from the CEMI RAS, and foreign experience in stimulating the birth rate. It is concluded that it is difficult to isolate the influence of a single factor on the increase in the total fertility rate. Accordingly, in order to really reverse the trend of depopulation, the simultaneous influence of many factors is needed.

The main goal of this paper has been to summarize selected developments in the field of artificial societies and agent-based modeling and to suggest how this fundamentally new toolkit can contribute to solving some of the most complex scientific and practical problems of our time. This paper consists of several parts. The first one considers the history of agent-based models. The second part is devoted to models of this class related to epidemiology, and the third considers agent-based modeling of pedestrian traffic and evacuation of the population. The fourth part is devoted to issues of demographic process modeling, and the fifth to the simulation of transport systems. Ecological forecasting is considered in part six, while the seventh is devoted to issues of land use, and the eighth to urban dynamics. The ninth part considers models used for the reconstruction of historical episodes, and in the tenth part, we will briefly touch on the issues of conflict simulation. In the eleventh part of the paper, we consider issues related to studying social networks using an agent-based approach. Agent-based models of economic systems are considered in the twelfth part. Undoubtedly, the scope of applications of agent-based models is wider, but we elected to focus on these indisputably significant areas.