Essence of Technology

     Considered here Technology (IT AC DTIP) is designed to solve the problems on effective control of DTIPs of various purposes. Each DTIP, that enters in the scope of action of Technology, is characterized by the following seven properties:

1)   stationarity, limitation in time or instantaneousness of its flow;

2)   availability for control;

3)   limited aftereffect or absence of any aftereffect in the description of its current state, i.e. presence in the DTIP phase state vector (formula (1), see subsection "Properties of DTIPs" of this Sec.) of digital and analog parameters with limited ranges of their possible values;

4)   representing it as a controlled service system (Fig. 2, see the same subsection "Properties of DTIPs"), which includes interacting objects of three possible categories: 1 — devices of service; 2 — demands and 3 — mediums (devices have resources and are intended for servicing­ demands, and mediums describe conditions of functioning in the system of devices and demands);

5)   discrete moments of possible entry to the system of demands and devices, their leaving from it, as well as coming into interaction between them;

6)   limited set of controls (instant and non-instant) available to apply to objects of the system in each its state;

7)   additive quality criterion, that quantitatively describes the efficiency of functioning the controlled service system.

Definition. Methodologyalgorithm of search a goal, a set of techniques, methods, tools, methods, principles of achieving the goal.

     The basis of Technology is a uniform scientific methodology of numerical optimization of stationary discrete processes of service, which provide fulfillment of the following three interconnected procedures:reduction of initial DTIP to a stationary discrete process of service (DPS) with a finite number of phase states, which is reached due to the use to the first process of incrementing or Markov approximation and its possible subsequent regeneration (in case of its limitation in time or instantaneousness)

  • reduction of initial DTIP to a stationary discrete process of service (DPS) with a finite number of phase states, which is reached due to the use to the first process of incrementing or Markov approximation and its possible subsequent regeneration (in case of its limitation in time or instantaneousness);
  • synthesis of optimal tabular strategy of control (TSC) of created DPS, which is reached by applying to it one of known algorithms (schemes) of recurrent optimization (in the presence ofincrementing the laboriousness of performing this procedure can be adjusted by changing an increment value);
  • evaluation ofreal effectiveness of the foundTSC applied to the initial DTIP, the need for which arises due to unaccounted factors during synthesis of the strategy or its possible improvement at the stage of its application.

     This methodology allows to synthesizenumerically of optimal TSCs or close to them for DTIPs, which satisfy the above properties. Explain its essence on the following three examples.

     Example 1. Was solved the task of optimizing the delivery of a passenger by public transport. In this case, the initial DTIP represents a time-limited process. On its basis was constructed a stationary controlled DPS by regenerating the first one through artificially introduced increment of a unit duration (first procedure described above). Further, to this DPS was applied the main scheme of recurrent optimization and as a result was found numerically the optimal tabular strategy of its control (second procedure). Then this strategy was represented in terms of initial task in user-friendlyform as list (third process).

     Example 2. Was solved the task of optimizing expenses. In this case, the initial DTIP represents an instant process of selecting choices. To construct on its basis a stationary controlled DPS, we had to add to it one fictitious state with a staying time in it a unit time interval. As a result, this process became a lasting in time one, which allowed to form from it by regenerating the stationary DPS (first procedure). Then there were fulfilled two remaining procedure performed in the same manner as in the first case.

     Example 3. Was solved the task of optimizing reformation of a military formation. In this case, the initial DTIP is a time-limited process. On its basis was constructed a stationary controlled DPS by applying incrementing to the initial process and its subsequent regeneration (first procedure). The incrementing here is necessary in order to, firstly, limit the number of DPS phase states, and secondly, specify the moments of control this process. Further, to DPS was applied the main scheme of recurrent optimization, as a result was found numerically the optimal tabular strategy of its control (second procedure). Then this strategy was represented in user-friendly graphic form (third process). At that, occurredan improvement of the found strategy at the stage of its applying due to the fact that were eliminated downtimes, which took place in the controlled DPS between the moments of completing service of demands (moments of completing reformation of separate military units of the military formation) and the closest moments of incrementing, in which are permitted controls.

     Implementation to life of IT AC DTIP will provide a great economic and social effect from using of highly effective strategies for control of discrete processes in various areas of human activity. This conclusion of the founder of Technology is based on the following obvious statements:

  • currently does not exist a similar information technology, that makes in every particular case either to apply obviously nonoptimal (heuristic) strategies for control of corresponding DTIPs, or to spend a lot of money on the development of mathematical means and program software for synthesis of such optimal strategies;
  • DTIPs are widespread in various fields of human activity, that is why solving the problem of effective control of such processes is very actual;
  • proposed Technology is based on a uniform scientific methodology of numerical optimization (see above), which allows to train specialists on its practical use in various fields of human activity;
  • there is a high level of automation in realizing many procedures of Technology due to their standardization and universalization.