VolRC RAS scientific journal (online edition)
RuEn

Journal section "Socio-economic research"

New in the Methodology of Studying a Social Space, or What Is Agent-based Modeling?

Rossoshanskaya E.A.

4 (6), 2016

Rossoshanskaya E.A. New in the Methodology of Studying a Social Space, or What Is Agent-based Modeling?. Social area, 2016, no. 4 (6). URL: http://socialarea-journal.ru/article/2016?_lang=en

Abstract   |   Authors   |   References
The article raises the issue of development of the methodology of studying a social space as an abstract multidimensional space of social positions. The author raises the issue of the prospects of using agent-based modeling in the study of a social space. The article summarizes the theoretical and methodological foundations of agent-based modeling; in particular, it describes its essence, specifics and advantages, and presents the concept of an agent-based model. The definitions of the term used in the works of leading Russian experts are discussed: V. L. Makarov, A. R. Bakhtizin, E. D. Sushko and others. The meaning of “an agent” as a central and most important category of agent-based modeling is revealed. Its proximity to the concept of “a social agent” used in sociology of space is highlighted. The basic properties and attributes of the agents in agent-based modeling are outlined. The stages of modeling and the requirements for knowledge, skills and qualifications of the modeling subject (the developer of agent-based models) are described. The article presents a brief overview of the Russian experience of agent-based modeling of socio-economic processes which are represented by models such as a demographic model of a notional region, a multi-agent model of Russia, an agent-based regional model “Governor”, a complex agent-based model of urban socio-economic development (CABMUD), an agent-based interregional inter-industry “input-output” model, a multi-agent model of territorial system development. Comparative analysis of the software developed so far is conducted in order to implement agent-based models. Special attention is paid to the Russian modeling system AnyLogic as one of the best software solutions enabling the construction of agent-based models without much experience in programming. The author emphasizes the importance of interaction between physical and social spaces through the support of GIS¬-technology by some modeling systems. In conclusion, the article summarizes the main opportunities provided by the use of agent-based modeling in the study of a social space: first, the creation of agents close as possible to real social agents; second, the consideration of two-way relations of agents and space due to the implementation of interactions of two types: “agent-agent” and “agent-environment”; third, providing real tools for modeling physical and social spaces taking into account their unity and connection

Keywords

social space, agent, agent-based approach