Designing modules of system dynamics in decision support systems

Objectives. When creating models of system dynamics, the basic construct at the design stage is the representation of the process under study in terms of a causal relationship consisting of a positive feedback loop and a negative feedback loop. The construction of a model of a dynamic environment can experience a number of difficulties in using feedback. This work shows the possibility of designing modules of system dynamics for decision-making systems based on the situational-activity approach. The study proposes the gap in knowledge about models of system dynamics to be filled with a conceptual model of an act of activity, by means of which an expert system can be implemented based on production rules. In this context, conceptual models are applied to human reasoning with reference to certain types of activity. The objective of the study was to investigate the possibility of applying the situational-active approach to designing models of system dynamics of infectious diseases based on particular representations of the conceptual structure of the act of activity.

Methods. By synthesizing Bolotova's situational algorithm and Shchedrovitskiy's system-activity approach, the conceptual structure of the act of activity is presented as a methodology of the situational-activity approach. The analysis of this structure leads to the construction of a plan of processual structure and a plan of analytical relationships. The article proposed a hypothesis that the process representations describe the notation of flows and levels, and the analytical relationships implement differential equations. In order to prove this hypothesis, the subject area of infectious diseases was investigated.

Results. Based on the set of these plans, a graphic image was synthesized for constructing models of system dynamics, which is identical to the diagram of flows and levels of development of the SIR process. However, the problem of constructing conceptual structures is nontrivial, complex, and laborious. Therefore, the Designer-Solver-Interpreter software suite was implemented. The software tools enable a visualization of the conceptual structures and implementation of the knowledge bases for expert models of system dynamics. It also tests the completeness and viability of the model.

Conclusions. To date, there is no single conceptual structure for designing expert systems and situational and simulation dynamic models. The proposed method and software tools allow these problems to be resolved using the situational-activity method. Various types of dynamics in expert systems interact, thus confirming the reliability of knowledge in the models of system dynamics. The conceptual structures of the act of activity are the core part of designing expert systems, while he derivative process and analytical representations of the act of activity are the core part of developing modules of system dynamics.

1. Brehmer B. Dynamic decision making: Human control of complex systems. Acta Psychol. (Amst.). 1992;81(3): 211-241. https://doi.org/10.1016/0001-6918(92)90019-A

2. Andrianova E.G., Golovin S.A., Zyko S.V., Lesko S.A., Chukalina E.R. Review of modern models and methods of analysis of time series of dynamics of processes in social, economic and socio-technical systems. Rossiiskii tekhnologicheskii zhurnal = Russian Technological Journal. 2020,8(4):7-45 (in Russ.). https://doi.org/10.32362/2500-316X-2020-8-4-7-45

3. Yarygin O.N., Korostelev A.A. Systemic dynamics as the basis of modern managerial competence. Aktual'nye problemy ekonomiki i prava = Actual Problems of Economics and Law. 2014;4:196-205 (in Russ.).

4. Forrester J.W. Information sources for modeling the national economy. Journal of the American Statistical Association, 1980;75(371):555-574. https://doi.org/10.2307/2287647

5. Sterman J.D. All models are wrong: reflections on becoming a systems scientist. System Dynamics Review. 2002;18(4):501-531. https://doi.org/10.1002/sdr.261

6. Gonzalez C. Learning to make decisions in dynamic environments: Effects of time constraints and cognitive abilities. Hum. Factors. 2004;46(3):449-460. https://doi.org/10.1518/hfes.46.3.449.50395

7. Nakamura G.M., Cardoso G.C., Martinez A.S. Improved susceptible-infectious-susceptible epidemic equations based on uncertainties and autocorrelation functions. R. Soc. Open Sci. 2020;7(2):191504. https://doi.org/10.1098/rsos.191504

8. Gavrilova T.A. Ontological approach to knowledge management in the development of corporate automation systems. Novosti iskusstvennogo intellekta = News of Artificial Intelligence. 2003;2:24-30 (in Russ.).

9. Verkhoturova Yu.S. Model of the subject domain in the language of ontology description. Vestnik Buryatskogo gosudarstvennogo universiteta. Matematika, informatika = BSU Bulletin. Mathematics, Informatics. 2013;9:63-68 (in Russ.).

10. Shchedrovitskii G.P. The methodological meaning of the opposition of the naturalistic and system-activity approach. Voprosy metodologii. 1991;2:143-154 (in Russ.).

11. Rodrigues da Silva A. Model-driven engineering: A survey supported by the unified conceptual model. Comput. Lang. Syst. Struct. 2015;43:139-155. https://doi.org/10.1016/j.cl.2015.06.001

12. Ganter B., Obiedkov S. Conceptual exploration. Berlin, Heidelberg: Springer; 2016. 315 p. https://doi.org/10.1007/978-3-662-49291-8

13. Tan R.P., Zhang W.D., Chen S.Q., Yang L.H. Emergency decision-making method based on case-based reasoning in heterogeneous information environment. Control and Decision. 2020;35(8):1966-1976.

14. Chen D.-Y., Zhao H., Zhang X. Semantic mapping methods between expert view and ontology view. J. Softw. 2020;31(9):2855-2882.

15. Sorokin A.B., Smol'yaninova V.A. Conceptual design of expert systems of support of decision. Informatsionnye tekhnologii = Information Technologies. 2017;23(9): 634-641 (in Russ.).

16. Sorokin A.B., Brazhnikova E.V., Zheleznyak L.M. Designing a knowledge base for the development of intelligent models based on the conceptual structure of activity act. J. Phys.: Conf. Ser. 2020;1615:012023. https://doi.org/10.1088/1742-6596/1615/1/012023