Lecture
The evolution of the methodological foundations of such areas as operations research, systems analysis and decision theory is characterized by their mutual interpenetration. The integration of methods allows the DM
to achieve maximum effect in the study of complex, weakly structured objects and processes, which include socio-economic systems.
In the process of developing and analyzing the results of the practical implementation of a managerial decision, categories such as attributes, indicators, and criteria inevitably arise, which are used to formalize the properties and characteristics of the objects and phenomena under study, and to monitor the achievement of the set goals. When measuring the properties of an economic object, it is important to choose a measurement scale that would, on the one hand, provide the most informative assessment, and, on the other hand, — avoid an unjustified striving for high precision in cases where this could lead to a decrease in the quality of the information obtained about the object. For example, when conducting expert surveys, it is popular to use an ordinal scale, within which alternatives are ranked without a quantitative assessment of their characteristics, which best corresponds to the principles of human thinking and allows experts to express their opinion most correctly.
The need to make a managerial decision arises in the presence of a problem situation, i.e., when the state of the managed system is unsatisfactory from the point of view of the DM's goals. The goal of the DMP is to find the optimal (or best) managerial decision, i.e., such a decision that transfers the managed system from an unsatisfactory state to a state that satisfies the DM.
Modeling of problem situations for a specific DMP is often carried out using several available methods, then, by comparing the results obtained, the one that, from the point of view of experts or the DM, most adequately describes the original situation is chosen, which makes it possible to improve the quality of the managerial decisions made.
The general formulation of a multi-criteria DMP includes: finding the set of feasible decisions; constructing a vector criterion; describing the DM's preference relation; finding a subset of optimal (or best) managerial decisions based on the preference relation and the vector criterion, which reflect the goals of the management subject.
The mathematical model of a DMP provides for specifying the set of feasible decisions, the set of possible states of the environment, the set of possible results or outcomes obtained as a result of implementing the adopted managerial decision, and defining a criterion (evaluation) function and a preference relation.
Depending on the information that the DM has when making a decision regarding the state of the external environment, several main types of decision-making problems are distinguished: DMP under certainty; DMP under uncertainty (the random state of the environment is described with a given probability; the probability distribution over the set of environment states is unknown, but it can be estimated statistically or through experiment; the set of environment states is known, but there is no information about this set that would make it possible to assess the probabilistic properties of this set (DMP under complete uncertainty)); the randomness of the environment is determined by the purposeful actions of counterparties (DMP under game-theoretic conditions).
The classification of decision-making problems is closely related to the classification of decision-making methods, since the latter, in the form of requirements and constraints, take into account the specifics and conditions of the problems themselves.
The diversity of features and conditions of DMPs gives rise to a multitude of possible classifications of decision-making methods.
Knowledge of the specifics of particular problem conditions and the corresponding decision-making methods predetermines the possibility of their effective use by the DM to make the best decisions in applied economic situations. This is because the classifications of problems and decision-making methods make it possible to determine the range of decisions developed in specific divisions of the management apparatus and made at various hierarchical levels, the composition of the information used for this purpose, typical methods of information processing, the system for formalizing decisions made, as well as the processes of monitoring and stimulating their implementation.
Comments