20 July 2018
Review of the application of simulation modeling in the field of social planning, security, as well as modern machine learning
The report will provide an overview of the results related to the construction of simulation models in various fields. In particular, an agent model of crowd behavior in an emergency situation will be described, based on the proposed decision-making system and on the developed evolutionary algorithm of fuzzy clustering.
In the context of social planning, the Schelling’s model of decision making on the agent's choice of the area of residence, depending on preferences in relation to his surroundings will be considered. Depending on the initial conditions and the type of distribution curves of tolerance thresholds for agents of different types, the possibility of transition to an equilibrium state and arising cluster slices is investigated.
A simulation model will also be presented, designed to optimize the flow of loan applications in the interregional underwriting center of a large commercial bank. A distinctive feature of this model is the consideration of multiple factors affecting the processing time of loan applications by underwriters (specialists in assessing the probability of repayment of the requested loan).
All the mentioned models are integrated with machine learning modules for solving various optimization problems.
Last edited: 07 November 2018
International Institute for Applied Systems Analysis (IIASA)
Schlossplatz 1, A-2361 Laxenburg, Austria
Phone: (+43 2236) 807 0 Fax:(+43 2236) 71 313