Figure 1 - Architecture and technological scheme of DSS
(animation: 4 frames, 7 of repeat cycles, size - 652×157, 5.81 kilobytes)

1. Relevance of the topic

In any company, any organization involved in management decisions a person (manager). Let's say there are already established staff, the manager is aware of the abilities of each. The allocation between employees (performers) tasks assigned manager can be subjective, because of their likes and dislikes, so the distribution of responsibilities will be incorrect manner. Or another case. The team received new employees, and the manager did not even know about their abilities, and the DSS already contains such information.

Another important aspect is the timely repair of equipment and its maintenance.

For these reasons is the actual development of a decision support system for managing objects prophylaxis regular periodic use.

2. The purpose and objectives of the study, expected results

Purpose - to improve management of preventive maintenance facilities regularly periodic use by the research and development of decision support systems, including a model of the formation plan for the organization and implementation of prevention in different contexts.

The main objectives of the study:

1. The study of the functioning of services of maintenance and repair of equipment.
2. Management of prevention in terms of certainty and uncertainty.
3. Development of the structure and algorithm of DSS.

The object of study: decision support system for managing preventive maintenance facilities regularly periodic use.

Subject of research: a model of formation of a plan for the organization and implementation of prevention in different contexts.

Methods of research: systems analysis, control theory, mathematical modeling.

As part of the master's work is to get the actual scientific results is the practice of DSS prevention of facilities management on a regular basis, occasional use.

3. Review of Research and Development

3.1 Model artist

Pre-make the following assumptions:

1. We believe that every artist has full knowledge of the technology works object (in defect-free and defect states) and its schematic or design concept.

2. The concept of situation x∈X, which is understood [2] the environment and the behavior of other artists, working in conjunction with the performer. In this environment - this space immediately adjacent to the object, and the parameters of its states can be, for example, temperature, tightness of the working space, the working equipment in the immediate vicinity of the object, etc. Because here is the model of one artist, then as the situation will take only the state of the environment.

3. We believe that the accumulation of experience artist produced "their" disaster recovery procedures, then there are some stereotypes (in the good sense of the word). Moreover, each of these stereotypes can be seen as an individual standard in the sense that the artist himself shaped the actions that it considers necessary to implement in a given situation. It is important to note that the individual rate does not go beyond the requirements established in the particular enterprise and recognized by all (for example, safety requirements). That said, we believe that every artist has formed "own" sequence of maintenance operations k-th element of the object, that is the procedure A_kn can be written as:

At the same time, we note that for some situations fixed sequence of operations strictly fixed, that allows to consider the procedure (3.1) as the cortege. In general, the sequence of operations during the procedure A _kn may vary depending on the particular situation, given the inherent property of human adaptation.

4. To determine the time t^°_k to distinguish the time axis t interval (t^°_k1, t^°_k2) boundary is the minimum time the procedure A_kn, due to physical disabilities artist t^°_k1, and the execution of this procedure, set by normative documents on the object t^°_k2 [4].

First, we consider the deterministic case. We believe that the n-th executor known x ∈X specific situation in which he would have to recover the k-th element of the object.

After receiving a task, n-th performer, as an active element of the system, planning their future behavior when restoring health k-th element of the object. To do this, he mentally loses procedure A_kn , taking up in relation to this play bystander. This process is called self-reflection [4]. This artist does not only form the program of their behavior, but also conducts self-esteem, that is, sets the time for yourself t^° _kn, which he estimated costs of performing the procedure A_kn. The arguments can be presented as artist [4]: I have performed the procedure A_kn in such circumstances and I know that I will need time, not less than t^°_{k 1}, but not more than t^°_k2; I realize this procedure in the time interval from about t^°_k1 to t^°_k2.

Since the interval (t^°_k1, t^°_k2) is set by the n-th performer, it can be argued that this interval is fully consistent with its capabilities and preferences. Therefore, this interval may be considered as a rule of the n-th performer [4]. This membership function variable t^°_kn to the concept (terms) Norma will be equal to:

(3.2)

Thus, n-th artist has positioned itself as a specialist, a norm of behavior is serving k-th element of the object under certainty is to perform the procedure A _kn during the time interval from t^°_kn1 to t^°_kn2 .

Now suppose that the n-th Executive unknown situation in which he would have to recover the k-th element of the object. But, as seen specific service object, then with full confidence it can be argued that at least the boundary values of the situation are known a priori. Moreover, it can be argued that x _min Je X (the most favorable conditions), and x _Max Je X (the least favorable) are constants. Thus, we have the classic problem of decision making under uncertainty interval.

It is known that the only objective of this class for this task is the so-called guaranteed result, i.e. a result that can be obtained under the worst conditions.

We assume that the n-th Executive also holds this principle. Next, assume that the time interval (t^°_k1, t^°_k2) covers all possible situations related to environment. Finally, we note that in determining ratings t^°_kn Executive reasoned thus: I have performed the procedure A_kn in such conditions ... . Thus, it can be assumed that the interval (t^°_kn1, t^°_kn2) is constant, but is shifted within an interval (t^°_k1, t^°_k2) depending on the situation.

When taking into account all the above put that corresponds to the worst conditions of a range of estimates (t^°_kn3, t^°_k2) ∈ (t^°_k1, t^°_{k 2}), the rate of the n-th Executive (guaranteed result) in a Uncertainty is the time

Therefore, under conditions of uncertainty the n-th artist has positioned itself as a specialist, which is normal behavior for maintenance k-th element of the object is to perform the procedure A_kn during the time interval from t^°_kn1 to t^°_{kn 2}, closest to the right border interval (t^°_k1, t^°_k2).

These models are formed without the type of artist, that is without taking into account its level of self-esteem. At the same time, it is reasonable to assume that the n-th artist is characterized by excessive, or low self-esteem. It is understood that in the case of self-inflated self-evaluation interval (t^°_kn1, t^°_kn2) is shifted toward the lower limit of the interval (t^°_k1, t^°_k2), and at low self-esteem - to the upper limit of the range (t^°_k1, t^°_k2).

By definition, made earlier, maintenance personnel to service the top SUTS block diagram is presented in the form of block implementation of the BWI management. In order to form a model of STB at this stage, we assume that the maintenance staff is implementing the same type. Given the model of any n-th artist and model designations adopted by STB for a particular situation x ∈ X take the form shown in Table 3.1.

Table 1 - Model STB

3.2 Model Centre

In the theory of control of organizational systems [5–8] under the center generally understand the hierarchical structure, provide leadership (control) agents (performers), and members of the hierarchy are called managers. Because at this stage is not considered functional interaction of all managers, members of the top management of the service, as the center will take a lot of managers M = {m ₁ , ..., m _r , ..., m _s}, directly issue the performers.

In forming the models take into account the following management [9]:

1. Every r-th manager, as executor, has full knowledge of the technology works object (in defect-free and defect states) and its schematic or design concept.
2. Every r-th manager, unlike the artists may be known to the concrete situation, that is known to x ∈ X.
3. Every r-th manager has knowledge of maintenance procedures every k-th defect A _k .
4. Every r-th manager is able to control the work of artists.

According to [5–8] relations between the center and agents in the process of co-operation are modeled using three hierarchical games D ₁ , D ₂ and D ₃ . It is known that the center is not your own (non-mediated agent) result of the activities, that is the result of his work is the result of the agent. Therefore, common to these games is the desire center "force" the agent is to carry out those actions that are most beneficial to the center, that is the whole system. For this center, using his right "first mover" uses various types of controls? institutional, motivation and information, or a combination of both [5–8, 10].

Thus, all other things being equal, the only effective way to improve the efficiency of work on the top is to get the "right" solution assignment problem.

Suppose that in the problem considered here Center offers the performers make an application for execution of the k-th job. A quantitative measure of such applications are self t^°_kn, that artists are supposed to, according to the center. As a result, the center receives information about the model of STB (see. Table 1). It goes without saying that this information is subjective, as formed by the performers.

We turn now to the priorities that, in fact, reflect the reputation that has developed at the center of a particular artist. Reputation is undoubtedly vague (fuzzy) characteristic of the artist, because in her estimation using such gradation (terms), such as low, medium, high, very high, etc.

In accordance with the method of paired comparisons assume that the center has the ability to answer questions like: How many times does the reputation of the i-th Executive superior reputation j-th artist?. The results of the responses can be represented as some superiority coefficients [11]:

(3.4)

where α_i - i-th reputation of the performer; α_j j-th reputation of the performer.

For the numerical evaluation of α_ij is the most suitable transitive scale type [11]:

• α_ij - slight superiority;
• α_ij·α_ij - strong superiority;
• α_ij·α_ij·α_ij - a very strong superiority;
• α_ij·α_ij·α_ij·α_ij - the absolute superiority, etc.

Note that statements made here such as, for example, a strong superiority are quite natural for managers.

As a basis for quantifying the factors of excellence in [11] proposed to use the α_ij = 1.5 or α_ij = 2. As previously it was assumed that all the performers are the top service are approximately the same qualifications, for the formation of reputation is a very suitable base score equal to 1.5.

For each k-th element of the object introduce the following notation:

• α_kij coefficient superiority serving k-th element;
• α_ki i-rate reputation of the Executive serving k-th element;
• α_kj reputation index j-th artist serving k-th element.

We believe that the reputation of the indicators are positive relative value, ie 0 ≤ α_kn ≤ 1, and therefore for each k-th element of the equality:

(3.5)

Furthermore, we note that

(3.6)

So, let r-th manager as a result of the pairwise comparisons artists expressed their opinions in the form of coefficients of excellence (α_kij) _r . In accordance with the formulas (4) and (6) define

(3.7)

Then, using the formula (4), define α_kjr >. After such calculations for K and N defects performers have a model reputation, which can be represented as in Table 2.

Table 2 - Model reputations α^y _kjr

Likewise, reputations are built models of other (s - r) managers, and hence the center of the model there s reputation models.

Conclusion

Note

In writing this essay master's work is not yet complete. Final completion: December 2015. The full text of work and materials on the topic can be obtained from the author or his manager after that date.

References

1. Резников В.А., Суворова А.М. Качественные модели системы управления техническим состоянием оборудования // Искусственный интеллект. –2011. – № 1. – С. 229-235.
2. Бурков В.Н., Коргин Н.А., Новиков Д.А. Введение в теорию управления организационными системами. – М.: Либроком, 2009. – 264 с.
3. Садовин Н.С., Садовина Т.Н. Основы теории игр: Учебное пособие. Йошкар-Ола: Изд-во МарГУ, 2011. – 119 с.
4. Резников В.А., Темник А.М. Нечёткая модель исполнителя системы технического обслуживания и ремонта в условиях определенности // Радiоелектроннi i комп’ютернi системи. – 2012. – № 4. – С. 201-205.
5. Бурков В.Н., Новиков Д.А. Как управлять организациями: Учебное пособие для вузов. – М.: Синтег, 2004. – 400 с.
6. Новиков Д.А. Теория управления организационными системами. – М.: МПСИ, 2005. – 584 с.
7. Воронин А.А., Губко М.В., Мишин С.П., Новиков Д.А. Математические модели организаций: Учебное пособие. – М.: ЛЕНАНД, 2008. – 360 с.
8. Бурков В.Н., Коргин Н.А., Новиков Д.А. Введение в теорию управления организационными системами. – М.: Либроком, 2009. – 264 с.
9. Резников В.А., Сандул Ю.О. Модель организации профилактического обслуживания сложных технических объектов // Искусственный интеллект. – 2014. – № 4. – 151 с.
10. Новиков Д.А. Институциональное управление организационными системами. – М.: ИПУ РАН, 2004. – 68 с.
11. Черноруцкий И.Г. Методы принятия решений: Учебное пособие. СПб.: БХВ-Петербург, 2005. – 416 с.