You get a bonus - 1 coin for daily activity. Now you have 1 coin

The structure of a decision support system

Lecture



It is determined by

the degree to which the system covers methodological information (the second methodological aspect) and by the content of the specific tasks assigned to the system (Fig. 5.1).

Structure of a DSS considered in the broad sense.

Each DSS is developed or improved for the specific managerial or economic problem being solved and must take into account the features of the decision-making process that exist in the organization. Nevertheless, three main functional components of a DSS can be identified: a database (or knowledge base), a library of models, methods and tools, and an interactive user interface. This is precisely why, for the decision maker, the following subsystems are of particular importance in the structure of a DSS:

  • • Subsystem 1 «General informational and methodological support for decision-making»;
  • • Subsystem 2 «General information on economic-mathematical methods and tools for decision-making»;
  • • Subsystem 3 «Specific tasks of the subject domain».

The structure of a decision support system

Fig 5.1. Conceptual diagram of a DSS (in the broad sense)

The content of subsystems 1 and 2 should essentially represent reference information used in forming managerial or economic decisions, i.e., used when solving the specific tasks considered within subsystem 3. In fact, subsystems 1 and 2 provide the informational and methodological support used for the operation of subsystem 3; solving specific tasks within subsystem 3 presupposes the use of the knowledge that they provide. This is precisely why the volume of reference information in subsystems 1 and 2 will directly depend on the specifics of the tasks solved by subsystem 3.

Note that subsystem 3, at the same time, is in fact an independent DSS considered in the narrow sense, in which the questions of methodological and informational support for the task-solving process related to forming the final managerial or economic decision are taken «outside its scope».

Subsystem 1 «General informational and methodological support for decision-making» (Fig. 5.2). The purpose of this subsystem is to provide general methodological information that the DM must take into account when making specific managerial or economic decisions.

Since a DSS can be used both for processing information and for forming alternative and/or final decisions, the following sections can be identified within the general methodological information:

  • • Section 1 «Methodological information related to solving information-processing tasks»;
  • • Section 2 «Methodological information related to solving the tasks of forming alternative and/or final decisions».

Within Section 1 «Methodological information related to solving information-processing tasks», in turn, the following blocks can be identified:

• Block 1 — a general characterization of information as the main resource used in forming managerial or economic decisions;

The structure of a decision support system

Fig. 5.2. Structure of Subsystem 1 «General informational and methodological

support for decision-making»

  • • Block 2 — a list of possible sources of information;
  • • Block 3 — a list of the data that can be obtained from the information sources listed in Block 2.

Other additional blocks can also be identified.

Section 2 «Methodological information, related to solving the tasks of forming alternative and (or) final decisions» may include the following blocks:

  • • Block 1 — the content of the basic concept of utility maximization ;
  • • Block 2 — the content of the basic concept of bounded rationality ;
  • • Block 3 — a description of the normative model of the decision-making process ;
  • • Block 4 — a description of the descriptive model of the decision-making process ;
  • • Block 5 — a description of specific particular concepts used in descriptive models;
  • • Block 6 — a description of the algorithms used to find optimal and efficient solutions in normative decision-making models;
  • • Block 7 — a description of possible multi-criteria choice strategies used in descriptive decision-making models;
  • • Block 8 — a description of algorithms for applying alternative-preference rules for descriptive decision-making models;
  • • Block 9 — a description of methods for calculating the criterial evaluation for each of the alternatives in descriptive decision-making models;
  • • Block 10 — a description of the sources of information (sources of source data) used to form alternative and/or final decisions;
  • • Block 11 — a description of the procedure for making a managerial decision.

The list of blocks can be continued. Thus, into a separate block

methodological information can be singled out that is related to the specifics of constructing managerial or economic decisions for socio-economic systems, as well as to such notions as «modeling», «Pareto set», «criteria», «importance of criteria», «fuzzy sets», etc.

Subsystem 2 « General information on economic-mathematical methods and tools for decision-making» (Fig. 5.3), whose purpose is to provide methodological support for decision-making by supplying information on possible decision-making methods and tools.

The structure of a decision support system

Fig. 53. Structure of Subsystem 2 «General information on economic-mathematical methods and tools for decision-making»

Since economic-mathematical methods and tools can be used both for solving information-processing tasks and for forming alternative and/or final decisions, the information of this subsystem can also be systematized into two sections:

  • • Section 1 «Information on economic-mathematical methods and tools for decision-making related to solving information-processing tasks»;
  • • Section 2 «Information on economic-mathematical methods and tools for decision-making, related to solving problems of forming alternative and (or) final decisions».

In Section 1 «Information on economic-mathematical methods and tools for decision-making, related to solving problems of information processing», in turn, the following blocks can be identified:

  • • Block 1 — description of algorithms for processing source data into information acceptable to the DM and used by them to form the final decision;
  • • Block 2 — description of algorithms for converting the processed information into the form required by the DM.

Other additional blocks may also be identified in Section 1.

In Section 2 «Information on economic-mathematical methods and tools for decision-making, related to solving problems of forming alternative and (or) final decisions» information blocks can be identified taking into account the DM's adopted classification of decision-making methods and the available software.

As an example of an aggregated classification of decision-making methods, the following can be proposed:

  • • decision-making methods under certainty;
  • • decision-making methods under probabilistic uncertainty;
  • • decision-making methods under complete uncertainty;
  • • heuristic methods.

Examples of software used to implement various mathematical methods of information processing may include the software environments MATLAB , Mathematica , the analytical platform Deductor , and others.

Useful for the DM in Subsystem 2 may be not only a description of specific software for data processing and the implementation of particular economic-mathematical methods and computational algorithms used to support decisions, but also an account of its potential applications.

Subsystem 3 «Specific tasks of the subject domain» (Fig. 5.4), the purpose of which is to provide and use specific methods and tools for solving problems related to forming managerial or economic decisions.

The structure of a decision support system

Fig. 5.4. Structure of Subsystem 3 «Specific tasks of the subject domain» —

DSS in the narrow sense

This subsystem includes the description and solution of specific tasks passed to the decision support system. In doing so, the methods and tools systematized in Subsystems 1 and 2 are used.

The information provided by this subsystem is structured in content into two sections:

  • • Section 1 «Solving specific tasks related to information processing»;
  • • Section 2 «Solving specific tasks related to forming alternative and (or) final decisions».

Section 1 «Solving specific tasks related to information processing» should contain the following information blocks:

  • • Block 1 — general statement of the problem;
  • • Block 2 — the specific task within the problem that has been assigned to the DSS;
  • • Block 3 — description of the task by parameters, general methodological information on which is presented in Section 1 «Methodological information related to solving information-processing tasks» of Subsystem 1 «General information and methodological support for decision-making». For example, a specific list of information sources used and the data obtained from them may be indicated; an algorithm for processing and analyzing source data into information acceptable to the DM and used by them to form the final decision may be proposed and described; a description of the algorithm for converting source data into the form required by the DM may be given;
  • • Block 4 — software implementation of the task solution.

The list of blocks can be expanded.

Recently, the use of the DSS for processing large data arrays that change rapidly over time . The task of the DSS in this case comes down to processing and presenting them to the DM in a form convenient for analyzing the situation. An example is data from real sales receipts, which can be used to analyze situations regarding the volumes, locations, and timing of demand for particular goods and services.

Section 2 «Solving specific tasks, related to forming alternative and final decisions» may include the following independent blocks:

  • • Block 1 — general statement of the problem;
  • • Block 2 — the specific task within the problem that has been assigned to the DSS. This may be the task of processing source data preceding the task of forming alternative and (or) final managerial decisions , as well as the task itself of forming alternative and (or) final managerial decisions;
  • • Block 3 — description of the task of forming alternative and (or) final decisions with respect to parameters, general methodological information about which is presented in Section 2 «Methodological information related to solving tasks of forming alternative and (or) final decisions» of Subsystem 1 «General informational and methodological support for decision-making». Thus, for example, if the task is to form alternative and (or) final managerial or economic decisions, it is necessary to note which decision-making concept is implemented — utility maximization or bounded rationality; on the basis of which decision-making model it will be implemented — normative or descriptive; whether additional particular decision-making concepts will be used, if a descriptive model is taken as the basis; which algorithm will be used to search for optimal and efficient decisions, if the task is implemented within a normative decision-making model; which strategies will be used, if multi-criteria choice takes place within the descriptive decision-making model being implemented; which alternative preference rules are used and what the algorithm for applying them is, if the task is solved within a descriptive model; how the criterial evaluation is calculated for each of the alternatives, if the task is solved within a descriptive decision-making model; which information sources (source data sources) are used to form alternative and final decisions, etc.;
  • • Block 4 — software implementation of the task solution.

Additional blocks may also be identified.

Overall, Subsystems 1—3 implement three main functional components of the DSS: a database (or knowledge base), a library of models, methods and tools, as well as an interactive user interface.

Comments

To leave a comment

If you have any suggestion, idea, thanks or comment, feel free to write. We really value feedback and are glad to hear your opinion.
To reply

Lectures and tutorial on "Decision theory"

Terms: Decision theory