Processing may take a few seconds...

Chapter


Title

Computing Environment for Forecasting Based on System Dynamics Models

Authors

[ 1 ] Instytut Studiów Strategicznych, Wydział Bezpieczeństwa Narodowego, Akademia Sztuki Wojennej | [ P ] employee

Scientific discipline (Law 2.0)

[5.3] Security studies

Year of publication

2019

Chapter type

chapter in monograph / paper

Publication language

english

Keywords
PL
  • Metodologia
  • Modele matematyczne
  • Narkotyki
  • Prognozowanie
Abstract

EN The paper proposes the computing environment which allows building forecasts based on System Dynamics models. The environment is equipped with GUI for building dynamic models according to Forrester’s methodology. The unique feature of the application is that the created models can be calibrated with the help of optimization procedures tailored for solving nonlinear least squares problems with differential–algebraic equations. Furthermore, the application enables verification of decision rules inherited in System Dynamics models by solving problems associated with the models dynamic optimization—then the model with optimal decision rules can be simulated to build forecasts of interests. To illustrate the functionalities of the environment, the example of the model of drug prevalence is discussed in some detail.

Pages (from - to)

43 - 56

DOI

10.1007/978-3-030-26036-1_4

URL

https://link.springer.com/chapter/10.1007/978-3-030-26036-1_4

Book

Theory and Applications of Time Series Analysis : Selected Contributions from ITISE 2018

Presented on

International Conference on Time Series and Forecasting, 19-21.10.2018, Grenada, Hiszpania

Points of MNiSW / chapter

20.0