2. From the sub-multitudes possessing "system-significant" properties to ear-mark versions suitable for DMP, Customer and experts:
¾ a set of modes for selecting experts into the group [Prognosist, DMP];
¾ a set of modes for forming expert groups [Prognosist, DMP];
¾ a set of modes for interrogating experts [Prognosist, DMP].
DESCRIPTION OF THE STAGE:
From the beginning the modes of selecting experts into the group are discussed. An inclusion of the so called "potential experts" is one of the basic modes for selecting experts into the group. An advantage of a wide use of a preliminary (initial) selection of experts is obvious. A formation of a group of experts in two stages (as, for example, in the situation analysis procedure) is especially stipulated in a number of expert procedures.
Numerous modes for selecting experts into a group are based upon the methods of taking the expert competence level into account.
In our opinion, a definition of the experts competence and a competence level of experts is an extremely difficult problem.
In the opinion of G.G. Azhaldov [1] concerning the expert competence level this "level should exceed an average level of the competence of a specialist in the problem".
From our point of view, it is most expedient to solve this problem on the basis of a posteriori analysis of opinions of experts on the problems being appraised.
When selecting the candidates into a group of experts it is necessary to take their socio-physiological and psychological characteristics into account in addition to their competence.
For example, provided in the group of experts there would be a direct informative interaction then it is necessary to consider their compatibility.
In case when an unprejudiced association (methods of a "brain attack" type) is desirable during an analysis of an object of prognostication then it is desirable to include one or two experts of a schizoid type into the group (where a psychologically normal expert will see something normal then, on the contrary, a schizoid is able to create a new stereotype).
In the opinion of G.G. Azhaldov, while selecting experts it is necessary to take the fact into account that a candidate should "according to his moral-ethic qualities be accepted as a third person capable of presenting his opinion on the raised problem before him independently on external influences and his own benefits".
While summing up the above said let's note what exerts an influence upon the selection of the expert as an Organizer, DMP and Prognosist.
First of all they are individual characteristics of the expert and also a specific character of the object of prognostication and a methodology being used.
The works [15,30,100] are devoted to the methods of a well-grounded selection of experts.
Let's enumerate the basic methods of forming expert groups [79,100]: appointment; mutual recommendations "snow ball"; consecutive recommendations; promotions by collectives of scientific divisions; documentational one; testing one.
NOTES:
1. Under the term a "group of experts" we understand at least two co-operating experts.
2. In a number of situations it is possible and expedient to enroll one expert. The number of enrolled experts depends, in particular, on a competence correlation of the number of candidates into experts available. (There are interesting algorithms specifying an estimation of one expert).
3. There are polar points of view on the problem of determining the competence level. The first one: "in spite of the fact that the problem of expert commissions forming possessing a sufficient level of competence is being discussed for a long time it may be considered rather raised than solved" [50]. And the second one: "this problem is actually solved on the basis of methods of the classic probability theory" [1].
4. From our point of view, the use of such concepts as a "general totality of experts" and "warranted probability of a collective expert estimation" is always practically incorrect and hampers the solution of problem of forming expert commissions.
5. The "data banks" in computers where the results of the work of experts involved earlier for the development of analogous prognoses could be of a great help for Prognosists and DMP when forming groups of experts.
6. It is desirable to register expert groups formed in one or another way with an order or instruction of the management which organizes an elaboration of a prognosis.
