Let’s summarize the results.
First, we considered the composition and structure of the control process, highlighting its five components (the object of management, the purpose of management, the control action, the model of the system, the subject of management) and discussing the interactions between them.
Second, we noted that after performing two initial operations (finding the desired control action on the system model and its execution on the system), the reaction of the system may be different. This requires specific management actions in each case, which allowed us to identify seven situations with special types of management behavior: program management, trial and error method, regulation, management by structure, management by goals, management in time deficit, and management in the absence of information about the ultimate goal. Each of them describes in detail the control algorithms and the limits of their capabilities.
Special attention should be paid to the special definitions of “complex system” and “large system” that we have introduced. They are the pride of the Tomsk school of systems analysis because before that different authors used these terms ambiguously, as synonyms, then used only one of them, then gave them a partially overlapping meaning. Clear distinguishing of these terms solved a few problems.
The first — the indication of different causes of complex and large systems has allowed us to offer different specific ways to overcome the difficulties associated with them: to combat the complexity additional information is needed to turn a large system into a small, accelerating decision-making.
The second — these qualities of the system can be combined in all four possible variants, requiring different approaches. The fuzzy division of the concepts “complex” and “large” can lead to difficult situations, according to the academician V. M. Glushkov, who did a lot for the development of computer technology and its implementation in the governance of the national economy of the USSR. At all levels, he argued that the main reason for the already manifested inefficiency of total economic planning is the insufficient capacity of the fleet of computers. When the Prime Minister A. N. Kosygin invited him to submit an application for any type and number of computers to solve this problem, it became clear that this action alone cannot solve the problem since the Soviet economy was not only a large but also a complex system.
Additionally, the classification of control types given in this chapter cannot be absolute and universal (as, indeed, any classification). As a model, it simply describes the variety of real-world control options. In life, there may be cases when the management of one system is used simultaneously or alternately a combination of different types of control. On the other hand, as a model, this classification is targeted, and other classifications may be required for other purposes. For example, in some cases there are automatic, semiautomatic (automated), and manual controls; these types are used in the control of machines, aircraft, manned spacecraft, etc.
Another classification will be required to distinguish management from other types of control: it is not a single-purpose management, not the management of the technical system, not the administrative management, not the automatic control, etc.
In conclusion, we note that in the Russian language the word “управление” has a very broad meaning. It includes such concepts as administration, command, management, and control of a technical device (machine tool, car, weapon, missile, etc.). It is interesting to know that in English there are words for specific types of control (government, management, control, administration, guidance, driving, etc.), but the general term equivalent to Russian “управление” is absent (the closest but not identical is processing).
Questions and Tasks
- 1. Which five components ensure the implementation of the control process?
- 2. Under what conditions is the search for control action on the system itself unreasonable and unacceptable?
- 3. What is a simple system? What is the reason for simplicity ?
- 4. Which system is called complex? What is the reason for the complexity?
- 5. Describe the algorithm of trial and error. What features does it have?
- 6. What distinguishes the method of trial and error from the “poke method”?
- 7. List the functions of a regulator.
- 8. What is management of goals? Under which conditions is this type of control applicable?
- 9. What is the “large system”? What are the options to manage it?
- 10. Give examples of systems that w'ould be both small and simple, small and complex, large and simple, large and complex.