About methods of support for management decision-making under conditions of significant uncertainty V.I.Baluta1,2, V.P.Osipov1, O.Y.Yakovenko3 Vbaluta@yandex.ru | Osipov@keldysh.ru | Javdet@gmail.com 1 Keldysh Institute of Applied Mathematics, Moscow, Russia; 2 Plekhanov Russian University of Economics, Moscow, Russia; 3 Institute of applied modeling and forecasting, Moscow, Russia Attempts to improve the quality of managerial decisions by introducing modern advances in information technology in various areas of public administration (socio-humanitarian, strategic, foreign policy, etc.) do not give the desired effect, comparable to the effect of their implementation in the manufacturing sector. The solution to this problem requires qualitatively new approaches to the issues of information and analytical support for decision-making in conditions of significant uncertainty. This article highlights the difficulties of predictive management of social processes based on direct computer modeling of social systems, considers the disadvantages of expert decision support methods used in practice, and proposes new technologies for applying expert knowledge and competencies based on the use of computer modeling and research. Current approaches to working with experts can be described as methods of coordinating the opinions of a group of experts based on their personal views (models) on the issue under discussion. Our experience has shown that creating a common integrated model by a group of experts gives a much better result. While the traditional approach can be called "group" intelligence, the new approach is called "collective" intelligence. In addition, methods of decision support using artificial intelligence systems are currently being intensively developed. We propose to begin work on the creation of "hybrid" intelligence with the integration of these approaches to obtain a synergistic effect. Keywords: public administration, decision support, computer modeling, expert judgement, artificial intelligence, group intelligence, collective intelligence, hybrid intelligence development of general society and its institutions, to 1. Introduction predict the events and reactions of people, and therefore Many researchers note the fundamental problem of cannot serve as a support in the development of managerial insufficient compliance of traditionally slow-evolving decisions at the state level. Modern systems of government public administration systems with modern challenges of are forced to function in conditions of a significant level world development. of uncertainty, which is due to the high dynamics and poor Large-scale institutional transformations of the world predictability of possible trajectories of changes in the community in the course of scientific and technological current situation, affecting the dynamics of current progress have formed and continue to modify the new priorities in the military-political, economic, social and reality of human and society existence. An increase in the humanitarian spheres. number of connections and relations into which each Attempts to compensate for an insufficient element of the social world order is somehow interwoven understanding of the phenomena occurring in these areas due to the processes of globalization and information and with data volumes, as a rule, lead more to overloading of technological development increases the complexity of management systems than to improving the quality of developing and making managerial decisions by orders of decision-making. magnitude. The empirical knowledge previously In fact, modern civilization challenges can be accumulated in the field of social sciences and humanities interpreted as a consequence of the growing gap, on the in the new conditions is of little use, since it does not allow one hand, between the demands of social development us to explain with sufficient degree of certainty the nature and, on the other hand, the capabilities of management and tendencies of the changes that are taking place, derive systems to develop and implement adequate solutions (fig. their regularities and predict the further course of 1). Fig. 1. Generalized view of public process management problems Copyright © 2020 for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0) the field of applied social and humanitarian research in 2. The State of Research domestic science, several points of concentration of As a subject of scientific research, the field of public advanced scientific thought can be noted. So, at the administration appears as a system of tasks of developing Institute of Management Problems of the Russian managerial decisions and tasks of reproduction and Academy of Sciences, quite a lot of attention is paid to the development of management mechanisms in relation to development of general theoretical approaches to certain areas and directions, for example, such as politics, modeling various aspects of managing complex systems economics, defense, production, international relations, under the leadership of D. Novikov, methods for solving society etc. Obviously, each of these areas has its own practical problems in the face of uncertainty - by V. specificity, its own range of problematic issues and Kulba's group, substantiation of the theoretical research methods, its own forms and methods of analyzing foundations of decision-making - by the group of the situation, developing and implementing solutions. The Aleskerov F. et al. The approaches to a comprehensive key tools for analyzing, developing and supporting the review of socio-political, economic, demographic implementation of solutions to date are various means of processes in their mutual influence are being developed by monitoring and visualizing data about the situation, a team of researchers at the Plekhanov Russian University including partially processed by relatively simple of Economics led by A. Kugaenko. analytical methods. 3. Basic approaches In addition to fairly simple analytical processing tools for monitoring data and their visual presentation forms, According to management theory, there are three basic recently in concepts and development plans for decision methodologies for improving the efficiency of developing support systems, much attention has been paid to the managerial decisions in relation to social and economic development of mathematical modeling methods and processes. These include full-scale experiments, expert various approaches to creating simulation models of assessments and mathematical modeling [26]. complex controlled systems. Based on such models, it is In the practice of public administration, targeted field supposed to conduct research and forecasting the behavior experiments are usually used to test the effectiveness in of the corresponding prototypes in various situations. As local pilot areas of any regulatory innovations or proposed the analysis of literary sources shows, the level of solutions in order to study the likely consequences of their development of modeling forms in various areas of public implementation. Although it is believed that field administration differs in their specificity and prevalence: experiments are most reliable, the cases of their if the economy has been using predominantly econometric application (unless, of course, the traditional “trial and and statistical models for a long time, and the military is error method” is referred to them) are extremely limited, persistently making attempts to introduce simulation, then in particular, they are unacceptable in crisis or urgent in social the humanitarian sphere still focuses on scenario situations, and most cases and simply impossible. modeling, and, as a rule, at the verbal level. An expert study of a problem situation is characterized Based on a relatively large number of publications of by the fact that general information about the situation is the corresponding orientation in foreign countries, we can limited to the expert’s personal knowledge. Although conclude that at present, there are multiple attempts to expert knowledge has the important property of focusing comprehend the possibilities and limits of the application on individual groups of alternatives, it is characterized by of traditional and new technologies of model research in an extremely high level of subjectivity and limitedness order to increase management efficiency in conditions of simply because of the characteristics of human thinking. significant uncertainty. To increase confidence in expert conclusions, groups of We can distinguish several among the most intensively experts are usually involved, but the factor of subjectivity developing areas modeling in the world. One of the main is not eliminated in that case. methods for analyzing the political situation remains Model studies, as a rule, are associated with the models based on the application of the principles of formalization of the description of the control object and equilibrium [1-3]. Methods based on the agent approach the current situation, the selection of criteria for the are getting a great deal of development, including their adequacy of modeling. A direct study of the situation on variety - multi-agent modeling (as a rule, requiring the model with an assessment of the possible significant computational resources, as well as significant consequences ends with an interpretation of the simulation costs for collecting and preparing data for modeling for results to redistribute the preference of alternatives. practical purposes) [4-6]. The method of system dynamics As the capabilities of computer technology and the remains quite popular, which over the past decades has development of model research methods have grown, the acquired many varieties, such as the use of technology of field of applicability of the results of model studies of cognitive maps, fuzzy modeling, etc. [7-19]. control objects and the environment has significantly Unfortunately, we have not come across work with the expanded to solve the problems of managing complex idea of simultaneous research of one object using a systems such as the state, economy, armed forces, etc. scientifically based of different models, which does not Thus, in practice, there is a competition between two allow us to evaluate their comparative effectiveness. fundamental approaches - an orientation toward the use of Although some generalizing works can be mentioned that information systems and ever-increasing complexity of could form the basis of such an approach [20-25]. models or at attracting the experience and knowledge of Regarding the development of modeling methods in experts. It is clear that this separation is somewhat contingent, since domain experts also participate in the experts uses their own ideas about the situation, on the creation of models, and any expert in solving the problem basis of which they form their own conclusions about its relies, inter alia, on the results of a model representation of development. To reduce the importance of the factor of the subject of study. specialization and subjectivity of experts, to date, many standard methods for rational construction of the processes 4. Problems of expert approaches of formation of expert conclusions and conclusions by One of the main difficulties in applying expert expert groups have been proposed. All of them are, to one approaches is the impossibility of a full-scale vision of the degree or another, based on methods of reconciling the analyzed picture in a holistic form by any of the experts resulting opinions by selecting the most preferred of them with the degree of detail that is inherent in each individual for some reason (acceptability, validity, authority, etc.). specialist in the field of his competencies. This does not This approach allows us to reach consensus and increase allow, within the framework of traditional approaches, a the level of confidence in the results, however, despite the qualitative analysis of multi-parameter problems with the group nature of the discussion, in fact even the adjusted complex nature of the interconnections within the systems final conclusion is based on the limited vision of one under consideration, which include most of the tasks of expert with whom the others agree, which virtually public administration. Moreover, it is obvious that the eliminates the possibility of original solutions appearing. impossibility of a holistic vision of the problem, situation, In other words, the technology used in practice (fig. 2), goals and objectives inevitably leads to a decrease in the which we will arbitrarily call the technology of “group” quality of management decisions. intelligence, involves procedures for agreeing on Even when solving problems in a group, each of the particular conclusions that do not allow going beyond the scope of voiced proposals and gaining new knowledge. Fig. 2. Conditional presentation of an “agreed” group decision of experts many of the approaches currently being developed are in 5. Problems of modeling social systems line with such a mechanistic approach - in the direction of From a research point of view, society, society is a the greatest possible detail in the model reproduction of variety of large complex systems with many elements that the system under consideration. As you know, the more are in a certain relationship and connections but with a complex the model, the more resources are needed for its large number of degrees of freedom. For this reason, creation and validation. Moreover, the more complex and various kinds of political, public, social, societal and accurate the model, the more difficult and costly it is to another processes are characterized by a significant degree provide it with satisfactory input data, that is, in addition of subjective uncertainty due to the high complexity of the to the development itself, additional resources are also behavior of individual elements of the system, the needed for collecting, validating, formalizing and entering complexity of the mechanisms for linking the behavior of large volumes of heterogeneous input data. the system and its elements and the objective limitation of When solving engineering, technical, and physical the subject in perception and understanding of reality. problems, where the behavior of modeling objects is To build a detailed model of such a system, it is objectively limited by the scope of well-studied physical necessary to describe the behavior of almost each of the laws, or in the production sphere, which is regulated by elements in all its relationships with others. Moreover, strict restrictions on resource balances and the rules of according to modern ideas about rational approaches to interaction of elements that reduce the degree of modeling, to obtain a qualitative model of the object of arbitrariness, such approaches are justified and give study, it is necessary to build a model of interaction of its satisfactory results. But in humanitarian spheres, objective components to the third structural level of the hierarchy of laws and restrictions that reduce the degrees of freedom of elements, which significantly increases the scale and elements have not yet been established (if they exist for complexity of the model. such systems at all with respect to the time and place scales Due to the increasing capabilities of computing tools, in question), so the creation of complex models for describing social processes encounters insurmountable summation or averaging of the finished conclusions of difficulties. individual experts, but a synergistic summation of the competencies of experts conducting a joint search 6. The lack alternate to expert methods solutions. If we go to the limit views, we can conclude that in the 7. The technology of "collective" intelligence field of public administration for solving practical problems it is impossible to build large-scale detailed When performing several search and research works models or provide them with initial data with details aimed at creating tools for predictive management of satisfying the necessary and sufficient requirements to complex systems under significant uncertainty, a new reduce the level of uncertainty in order to develop effective technology was tested - the technology of “collective” managerial of decisions. intelligence, combining expert analysis and modeling of However, one cannot deny the importance of modern complex systems (fig. 3). methods of computer modeling for studying complex systems, testing hypotheses, identifying obscure relationships and dependencies, etc. In other words, if direct modeling of specific processes in society is objectively impossible, then the use of various forms of modeling the systems under study in this area can be effective, first of all, to build up expert experience and knowledge. By the way, human evolution itself can serve as a confirmation of the higher efficiency of expert approaches in solving managerial problems in comparison with numerical modeling. There are facts about the unique abilities of individuals, for example, with absolute memory, able to perform ultrafast and voluminous calculations, etc. That is, the human brain is initially able Fig. 3. Conditional representation of the “collective” to do what computers are doing more and more well. And intelligence of experts physiologists believe that the potential of the human brain for certain types of information processing is almost The main contours of this technology were outlined in unlimited. Nevertheless, evolution did not follow the path the process of considering problematic issues of modeling of improving the mechanisms of deep analysis and and studying complex systems, searching for rational miscalculation of details, but the path of developing (primarily in terms of mathematical optimality) solutions mechanisms for a “broad” assessment of situations for an and predictive evaluation of their performance. The operational response in difficult conditions with high accompanying result of these works was a complex of uncertainty. various research and service software tools, some of which Strange as it may seem, despite the intensive have found application in various areas of solving development of multi-agent technologies, the prospects managerial problems, allowing to ensure their for the development of decision-making support systems implementation at a high scientific and technical level. based on computer technologies are also seen in the During the development, the foundations and the creation of “artificial intelligence”, that is, imitating the procedure for applying the technology were laid, its intellectual functions of a person in terms of “wide” and operability was verified, and its higher efficiency was operational analysis of reality and the development of confirmed in comparison with other expert methods for rational management decisions in conditions of high solving a certain range of tasks. uncertainty. From the above it follows that at present, the The general algorithm of work using the technology of use of expert approaches in management systems is seen "collective" intelligence involves the following set of as the most rational direction for the development of interdependent procedures: analysis and classification of a decision support systems. The only question is how to problematic issue; determination of the required most effectively use the progress in the development of composition of professional and target competencies of modern computer technologies to increase the the expert group; formation of a team of experts; teaching effectiveness of the application of these approaches and them the rules of technology; organization of coordinated their further improvement. interaction, including verification of psychological It seems that now the society is in a situation where a compatibility, distribution of roles, coordination of transition to new, more effective forms of information understanding of goals and perception of the subject of support for management is needed, based on a research; assignment of indicators and criteria of decision combination of the use of expert knowledge with the quality; the construction of a generalized cognitive model, possibilities of modeling situations, as well as on ways to as well as the selection of ready-made or the creation of a “objectify” the subjective knowledge of experts . In our set of express models of various subsystems and opinion, a transition from “group” intelligence to subprocesses; study of the model for consistency and “collective” is needed, when in the process of analyzing consistency; conducting iterative studies on model the situation and making decisions, there is not a simple examples; adjustment and refinement of the model; conducting experiments; comparison, analysis and “artificial” intelligence technologies. interpretation of results; preparation of an opinion. It is As noted above, it is advisable to use the modern clear that, depending on the situation, some deviations capabilities of modeling complex systems to build up from the regulations described here are possible in the expert knowledge [27, 28]. One of the forms of using implementation process. “artificial” intelligence can be the use of an electronic As a result, a look at the constituent elements of the training ground with a model of the complex system under model and the forms of their relationship with respect to study deployed on it and constantly being improved. As an the competencies of each expert remains subjective, example, we can cite the results presented in [28, 29], however, the complex model obtained as a result of which relate to the methodology of training an intelligent collective work has the properties of an independent object agent at an “electronic training ground” with a layout of alienated from personalities, on the basis of which it is the environment that allows, during many experiments, to possible to study the behavior of a controlled system for determine the mechanisms of formation of rational subsequent collective analysis of the resulting results. trajectories of movement to a target in the phase space of To some extent, here we can draw an analogy with the an agent’s state from an arbitrary provisions. method of “abstracting from the problem” within the After training at the training ground and introducing framework of the so-called “Harvard School principal the function of assessing one’s current position by an negotiations”, the substantive essence of which is not a arbitrary number of features, one can use the rules direct search for a compromise, but preliminary agreement developed at the “training ground” to make decisions in on the principles of its search. the current situation, including the situation understood with a certain degree of uncertainty. Considering that an 8. Technology of "hybrid" intelligence “electronic polygon” is still some abstract model The technology of the "collective" intellect is representation of the environment, it is advisable to considered as an intermediate stage in the development of combine both the construction of the model on polygon decision support systems. It is expected that as the and the recommendations resulting from its application technology of “artificial” intelligence develops, the with expert assessment by specialists. Thus, with the help “collective” intelligence described here can evolve into of artificial intelligence technologies, it is possible to some form of “hybrid” intelligence, forming a symbiosis expand the knowledge base on the basis of which with such properties that will ensure that the control alternatives are evaluated and recommendations are made system meets the new challenges of civilizational on the most rational actions in the face of uncertainty. One development. The possibility and pace of evolutionary of the possible versions of the model of the “hybrid” development of management information support methods intellect based on the “collective” and “artificial” is in the direction of “hybrid” forms will be substantially presented in fig. 4. determined by the progress in the development of Fig 4. A possible scheme of the symbiosis of technologies based on hybrid intelligence to the search for compromises that are oriented towards 9. Conclusion traditional approaches for coordinating expert opinions. It can be summarized that in order to increase the The originality of the proposed approach lies in the effectiveness of information and analytical support for appropriate combination of social engineering methods government in the face of significant uncertainty, it is with modern advances in the field of information necessary to develop methodological foundations, technology. Implementation of the technology of “hybrid” principles of construction and architecture of a hybrid intelligence will allow to obtain the results of an expert automated decision support system by synthesizing the assessment of the situation and alternatives by collective intelligence of experts with artificial synergistically integrating expert knowledge with the intelligence based on methods computer modeling of capabilities of “artificial” intelligence. complex systems and solving optimization problems. The References development of forms of expert management support should be focused on obtaining qualitatively new results [1] Bischoff, I., and L.-H. Siemers. (2011). Biased Beliefs through the integration of expert competencies, as opposed and Retrospective Voting: Why Democracies Choose Mediocre Policies. Public Choice 156: 163-180. Dynamic Model of the Rivalry Process.” American [2] Young, J. 2005. Research, Policy and Practice: Why Journal of Political Science 55, no. 2 (2011): 263-75. Developing Countries Are Different. Journal of [20] Burton, R. M., & Obel, B. (2011). Computational International Development 17: 727-734. modeling for what-is, what-might-be, and what- [3] Rausser, G.C., J. Swinnen, and P. Zusman. (2011). should-be studies-and triangulation. Organization Political Power and Economic Policy: Theory, Science, 22(5), 1195-1202. Analysis, and Empirical Applications. Cambridge: [21] Chiu, C.-Y., & Qiu, L. (2014). Communication and Cambridge University Press. culture: A complexity theory approach. Asian Journal [4] Allen, T. T. (2011). Introduction to discrete event of Social Psychology, 17, 108-111 simulation and agent-based modeling: voting [22] Cioffi-Revilla, C. (2017). Introduction to systems, health care, military, and manufacturing. computational social science: Principles and London, UK: Springer applications (2nd ed.). New York, NY: Springer. [5] Blais, A., Erisen, C., & Rheault, L. (2014). Strategic [23] Clark, W. R., & Golder, M. (2015). Big data, causal voting and coordination problems in proportional inference, and formal theory: Contradictory trends systems: An experimental study. Political Research in political science? PS: Political Science & Politics, Quarterly, 67(2), 386-397 48(1), 65-70 [6] Fieldhouse, E., Lessard-Phillips, L., & Edmonds, B. [24] Saaty T. Theory and Aplications of the Analytic (2016). Cascade or echo chamber? A complex agent- Network Process: Decision Making with Benefits, based simulation of voter turnout. Party Politics, Opportunities, Cost and Risk. - Pittsburgh, RWS 22(2), 241-256 Publications, 2009. [7] Fisunoğlu, A. (2019). System Dynamics Modeling in [25] Casey K.L. Defining Political Capital: A International Relations. All Azimuth: A Journal of Reconsideration of Bourdieu’s Interconvertibility Foreign Policy and Peace, 8 (2), 231-253. DOI: Theory // Paper University of Missouri. - St. Louis, 10.20991/allazimuth.476884 2007. [8] Abdollahian, Mark A., et al. “Dynamics of Cultural [26] Petrovsky, A.B.: Decision Making Theory. Change: The Human Development Perspective.” Publishing Center Academiya, Moscow (2009). in International Studies Quarterly 56, no. 4 (2012): 827- Russian 42. [27] Baluta V.I., Osipov V.P., Yakovenko O.Yu. The [9] Akgul, Oner. “A Bibliographical Study on the environment of modeling, forecasting and expertise as Academic Research of Peace and Conflict in Turkey.” the intellectual core of support management complex Paper presented at the 7th Eurasian Peace Science systems. KIAM Preprint № 82, Moscow, 2015 Meeting, İstanbul, 2018. [28] M.A.Berberova, K.I.Chernyavskii, «Comparative [10] Chamberlain, Todd. “Systems Dynamics Model of assessment of the NPP risk (on the example of Rostov Al-Qa’ida and United States ‘Competition’.” Journal and Kalinin NPP). Development of risk indicators of Homeland Security and Emergency Management atlas for Russian NPPs», GraphiCon 2019 Computer 4, no. 3 (2007): 1-23. Graphics and Vision. The 29th International [11] Chen, Yu-Che, and Michael J. Ahn. Routledge Conference on Computer Graphics and Vision. Handbook on Information Technology in Conference Proceedings (2019), Bryansk, Russia, Government. New York: Routledge, 2017. September 23-26, 2019, Vol-2485, urn:nbn:de:0074- [12] Demir, Cenker Korhan, and Engin Avci. “Turkish 2485-1, ISSN 1613-0073, DOI: 10.30987/graphicon- Terrorism Studies: A Preliminary Assessment.” All 2019-2-290-294, http://ceur-ws.org/Vol- Azimuth 7, no.1 (2018): 21-44. 2485/paper67.pdf, p. 290-294. [13] Forrester, Jay W. Industrial Dynamics. Cambridge: [29] Baluta V.I., Osipov V.P., Chetverushkin B.N., The M.I.T. Press, 1961. Yakovenko O.Y. Simulation of dynamics and [14] Ghaffarzadegan, Navid, John Lyneis, and George P. adaptation in the space of conflict interaction of the Richardson. “How Small System Dynamics Models intelligent agents. KIAM Preprint № 111, Moscow, Can Help the Public Policy Process.” System 2019. Dynamics Review 27, no. 1 (2011): 22-44. [15] Goertz, Gary, and James Mahoney. A Tale of Two About the autors Cultures: Qualitative and Quantitative Research in the Baluta Viktor I. - Candidate of Technical Sciences, senior Social Sciences. Princeton: Princeton University researcher, Keldysh Institute of applied mathematics of the Press, 2012. 253 System Dynamics. Russian Academy of Sciences, Plekhanov Russian University of [16] Hjorth, Peder, and Ali Bagheri. “Navigating Towards Economics. E-mail: Vbaluta@yandex.ru Sustainable Development: A System Dynamics Osipov Vladimir P. - Candidate of Technical Sciences, Approach.” Futures 38, no. 1 (2006): 74-92. leading researcher Institute of applied mathematics. M. V. Keldysh, RAS, E-mail: Osipov@keldysh.ru [17] Jervis, Robert. System Effects: Complexity in Yakovenko Oleg Yu., Leading Researcher. Institute of Political and Social Life. Princeton: Princeton applied modeling and forecasting. E-mail: Javdet@gmail.com University Press, 1997. [18] Kirkwood, Craig W. System Dynamics Methods: A Quick Introduction. College of Business, Arizona State University, 1998. [19] Morey, David S. “When War Brings Peace: A