SEARCH
 

Search

 

Результаты поиска 1 - 3 из 3
Начало | Пред. | 1 | След. | Конец


The technology to support agent-based modeling for supercomputers

Makarov V.L. Central Economics and Mathematics Institute, Russian Academy of Sciences, Moscow, Russian Federation ( makarov@cemi.rssi.ru )

Bakhtizin A.R. Central Economics and Mathematics Institute, Russian Academy of Sciences, Moscow, Russian Federation ( albert.bakhtizin@gmail.com )

Sushko E.D. Central Economics and Mathematics Institute, Russian Academy of Sciences, Moscow, Russian Federation ( sushko_e@mail.ru )

Journal: National Interests: Priorities and Security, #1, 2016

Importance The article discusses specialized software to implement agent-based models in supercomputers, and describes Supercomputer Technology for Agent-oRiented Simulation – STARS.
Objectives The research sets up a multi-agent demographic model, which has the interface to demonstrate changes in the main demographic indicators of the region's population. We devise a methodology to effectively reflect the computing core of the multi-agent system in relation to the architecture of modern supercomputer.
Methods We apply the internode communication technology through the active messaging technology that allows to significantly increase the productivity of multi-agent models.
Results STARS facilitates the implementation of online modeling and online visualization of the simulation process within the estimated time. Currently, it is possible only if there is a monopolistic access to supercomputer. To make large-scale experiments with more complicated agents, a more productive and data intensive supercomputer will be needed.
Conclusions and Relevance As the largest IT-market actors (Microsoft, Wolfram, ESRI, etc.) got more interested in AOM, the instrument proves to have good prospects and promising future. Supercomputing technologies and their application turn out to be reasonable and practicable, considering exponential growth in total data volume and the need to build analytical systems for obtaining new generation data required to predict social phenomena.


Simulation modeling of the Smart City System: The concept, methods, and cases

Makarov V.L. Central Economics and Mathematics Institute of Russian Academy of Sciences (CEMI RAS), Moscow, Russian Federation ( makarov@cemi.rssi.ru )

Bakhtizin R.A. Central Economics and Mathematics Institute of Russian Academy of Sciences (CEMI RAS), Moscow, Russian Federation ( albert.bakhtizin@gmail.com )

Beklaryan G.L. Central Economics and Mathematics Institute of Russian Academy of Sciences (CEMI RAS), Moscow, Russian Federation ( glbeklaryan@gmail.com )

Akopov A.S. Central Economics and Mathematics Institute of Russian Academy of Sciences (CEMI RAS), Moscow, Russian Federation ( akopovas@umail.ru )

Journal: National Interests: Priorities and Security, #2, 2019

Subject The research investigates key processes of urban life and its maintenance, including food supply, infrastructure, fire security, quality and accessibility of medical services, etc. The article also discusses the creation of a system supporting the Smart City decision-making process.
Objectives The research develops methods and tools to manage the Smart City system through system dynamics and agent-based modeling.
Methods Using simulation modeling, namely system dynamics and agent-based modeling (supported via Powersim and AnyLogic), we evaluate how multiple guiding parameters influence crucial characteristics of the Smart City system.
Results We devised an approach to designing the Smart City system through methods of system dynamics and agent-based modeling (supported via Powersim and AnyLogic) intended to streamline the decision making process for reasonable urban planning.
Conclusions and Relevance We propose the consolidated architecture of the Smart City decision-making system integrating the simulation models, data storage and city monitoring subsystem. The article describes the cases of simulation models implemented via Powersim and AnyLogic to support rational urban planning. The simulation models will significantly improve the quality of urban environment, satisfy the demand for food products, provide access to healthcare services and ensure effective rescue actions in case of emergency.


An agent-oriented social-ecological-economic model of a region

Makarov V.L. Central Economics and Mathematics Institute, RAS, Moscow, Russian Federation ( makarov@cemi.rssi.ru )

Bakhtizin A.R. Central Economics and Mathematics Institute, RAS, Moscow, Russian Federation ( cgemodel@yandex.ru )

Sushko E.D. Central Economics and Mathematics Institute, RAS, Moscow, Russian Federation ( sushko_e@mail.ru )

Journal: National Interests: Priorities and Security, #3, 2015

Importance In recent years, the Russian Federation authorities to the greater extent realize the need for regional development planning. At the same time, it is important to underline that environmental problems of territories become more acute, and the parameters of external environment affecting the economy of regions are subject to fluctuation due to various kinds of crisis in the global economy.
     Objectives The article aims to develop an instrument enabling to obtain high-quality forecasts of parameters of the socio-economic system depending on undertaken measures, as well as the estimation of dynamics of these parameters with possible environmental changes.
     Methods We have developed a multi-agent model of a region, which represents an artificial society, within the scope of which several individual models are integrated: natural environment of a region, the socio-demographic structure of its population, and the structure of its economy. This model is integrated in order to simulate the relationship of processes occurring in these areas.
     Results The model enables to demonstrate the dynamics of socio-economic and environmental characteristics of a region as a result of interaction of many independent actors (agents). The multitude of the agents simulates a real social structure of a region. In the model, the agents act in their own interests at pre-assigned institutional constraints.
     Conclusions and Relevance The design of the given model enables to simulate not only the influence of human activities on the economy and ecology of a region, but also the reverse influence of the quality of life of people on their health, working ability and behavior. The model can be used in the regional planning process to find a compromise between current and strategic goals to prevent over-exploitation of natural resources and to ensure a balanced economic growth.


Результаты поиска 1 - 3 из 3
Начало | Пред. | 1 | След. | Конец


Отсортировано по релевантности | Сортировать по дате