Paper presented at the EMNet 2011
December 1-3, 2011, Limassol, Cyprus
V.N. Volkova, V.Ye. Lankin, A.V. Tatarova
INFORMATIONAL MODELS OF PARTNERSHIP ORGANIZATION PROCESS
St.-Petersburg, Taganrog
St-Petersburg State Polytechnical University,
Taganrog Institute of Technology – Southern Federal University
The main objective of companies integration is efficiency improvement in production, trade and other kinds of activity by means of mutual exchange of technologies, know-how, management experience and so on.
Evaluation of the expected partnership organization efficiency and forecasting of the joint activity outcomes are primary theoretical problems.
In this work we suggest for observation methods of complex expertise arrangement on the basis of informational approach to A.A. Denisov’s [1] analysis of systems, which allow to raise evaluation impartiality as compared with the traditional methods of expertise.
When companies consolidate, different variants are possible [2]:
1. Complete entry of a company, which can be called a donor-company, into a larger company – a recipient-company (amalgamation of companies) (fig. 1,а).
2. Partial integration of resources for reaching a new jointly formulated purpose – joint enterprise (fig. 1,b).
3. Company mergence for joint activities with the creation of a coordinating centre (fig. 1,c).
Difficulties with company integration efficiency are connected with the absence of statistical data required for making traditional calculations of economical efficiency. In this connection there is a necessity to control the process of integration, which is connected with intermediate evaluation of company mergence variants and development on its basis of recommendations on expediency to continue or to interrupt the realization of the selected variant, redistribution of the resources supposed for the project. It requires not only to compare different alternatives, but also to make appropriate evaluations at different stages of integration.
The examined peculiarities of the process of company mergence and difficulties in evaluation of their efficiency have initiated the development of new methods for complex expertise arrangement based on informational approach to analysis of systems.
Informational models of the 1-st type. Implementation of informational models of the 1-st type is based on evaluation of the degree to which the evaluated factors (organizational functions, technologies, resources etc.) influence the realization of enterprise’s objectives in the analyzed period of development. In accordance with the informational approach [1,3] for each factor there we introduce assessments of the degree of its compliance with the objectives рi¢ and probability of its realization and implementation qi, and we calculate the potential (value) of Нi factor:
Нi = - qi log (1 - рi¢), (1)
where рi' - is the probability of reaching the objective when using some factor;
qi - is the probability of this factor realization for reaching the appropriate objective. Here the habitual Shannon probability of non-reaching the objective (entropy) рi is substituted with the conjugate (1 - рi¢).
In a number of situations it is possible not to take into account qi, and then the informational measure (1) will accept the following look:
Нi = - log (1 - рi¢). (1а)
Implementing the characteristics of р¢, q and Н, we can obtain comparative assessments of factor complexes influence on reaching the objectives of a definite level (sub-objectives), and that of the sub-objectives on reaching higher level objectives etc. until the global objective (if there is a multilevel structure of objectives), and substitute complicated assessment of a constituent’s influence on the final (global) objective of step-by-step evaluation of “minor” ambiguities.
Calculation of Нi on the basis of assessments of рi¢ and qi provides the supposed approach with some advantages as compared with the traditional expertise evaluations and evaluations with the account of criteria weight coefficients, proposed in the PATTERN method.
These advantages consist in the following:
· Simplification of obtaining generalized assessments of the influence of the assessed components complexes, as Нi, is measured in bits, it is possible to simply sum up, but while processing probability assessments we have to use more complicated procedures;
· Opportunity to evaluate not only the degree (probability) рi¢ of the influence of i-factor on the realization of the objectives, but also the opportunity to take into account probability qi of the realization and implementation of this factor in specified conditions in the current period of time (which can be evaluated on the basis of statistical research in a number of cases).
In this situation it should be noted that not normative assessments are in the discourse, but comparative assessments of the confronted factors or their complexes measured in bits.
For the convenience of presentation of comparative assessments results it is expedient to implement the procedure of rationing, i.е. to obtain HS = Si Hi, and calculate the correlative investment of each observed factor - x = Hi / HS .
The algorithm of realization of the observed evaluation method with the account of the rationing procedure is shown in fig. 2.
Fig. 2
To illustrate the implementation of the observed method of expertise arrangement we will give a simplified example.
Suppose, we are going to accept the variant of the complete entry of one company into anther, illustrated in fig. 1,a.
Then the model for usefulness assessment of a donor-company inclusion into a recipient-company can be presented as in fig. 3.
…
In fig. 2 there are the following symbols: Cr - recipient-company; Cd - donor-company; P1, P2, … Pn – types of products issued by the recipient-company; Т1, Т2, … Тn – are technologies (including equipment, goods and raw materials, production technologies); Pн1, Pн2 – are types of products issued by the donor-company which can become innovations for the recipient-company with appropriate adaptation (renovation of technology, materials and so on); Тн1, Тн2 – are technologies of the donor-company which can be applied by the recipient-company to issue new types of products; рi' - the degree to which the resources and types of issued products before the company integration influence on reaching the objectives of the integrated company; рj' - is the degree to which the use of technologies, adaptation and product renovation of the donor-company influence the reaching of objectives of the integrated company; qj - is the probability product types renovation by means of realization of the donor-company technologies and production renovation of the company that enters the basic one.
As objectives of a renovated company it is possible to accept income rise by means of new products manufacturing, widening of product markets, for instance, by means of market segments of the donor-company where the recipient-company can introduce its own products.
Then, to analyze the efficiency of company mergence we can use the following algorithm:
1. To assess рi' of each product types of the recipient-company which can be done on the basis of prior experience, sale statistics, market segments volumes, by transforming appropriate numerical values (accepted as the criteria of reaching the objectives) into relatives units of рi'; it is also possible to make expert evaluation of the potential investment into the total result of the company which is expressed in this approach with the help of informational measure of Hр. The experience of prior implementation of the observed approach for innovation assessment has shown that it is convenient to call this measure ‘the company potential’.
2. To calculate Нр = - Si log (1 - рi¢) in order to make it possible to compare the investment of your own company and that of the donor-company.
3. To assess рj' and qj for each component (types and kinds of products, technologies and so on) of the donor-company; where рj' and qj are assessed by single experts who work in separate competence spheres and who should be recommended to give a prognosis evaluation for a limited period of time, because it is considered easier for them rather than giving such evaluations for the current moment.
4. To calculate the potential of the donor-company
Нд = - Sj qj log (1 - рj¢).
It is now possible to manipulate the obtained evaluation results as usual numeric results, i.е. they can be summed, one can calculate the relative investments of companies potentials into the total result of the potential of the united company.
For example: to obtain HS = Нр + Нд;
to calculate the share of the donor-company investment xд = Hд / HS;
to define the relative share of the donor-company investment as compared with the former recipient-company potential e = Hд / Hр and so on.
Besides, if new types of product appear in the result of company mergence, then there is no need to evaluate the investment of all product types into the resulting potential of the united company using the method of rationing, it is sufficient to obtain additional evaluation result of investment of the new product types (NPT) Hнвп and to add it to the sum total potential. If it is interesting, it is possible to define the relative investment of NPT into the total potential.
Analogical model can be proposed in the case of equitable company integration. But in this case it takes to create a coordination body that would jointly produce the objectives and the criteria of reaching them.
To make a more thorough analysis of company integration expediency it is necessary to define the investment of specific kinds of products, kinds of technologies, materials, equipment and so on. In this case the general algorithm of conducting the expertise becomes more complicated.
To foster the procedure of conducting the complicated expertise automated and dialogue procedures are developed and applied [3].
In the observed example it was supposed that the experts give evaluation results to рi¢ or рj¢ on the basis of criteria which are clear only to them. At the same time it is advisable to specify the evaluation criteria. For this purpose it takes to form expert groups by different criteria and conduct a more complicated expertise obtaining results from the viewpoint of the correlative criterion and then to summarize these results. Besides, it is possible to divide the general objective into sub-objectives and to conduct evaluation of innovation impact by different criteria for different sub-objectives
The idea of evaluation with the account of different criteria is introduced in fig. 4.
I
Fig. 4
While working with such a model experts should single out the ways of evaluation of the influence degree of рjkn¢ of i-NPT by k-criterion on n-sub-objective, and probability qjkn of such evaluation realization.
For example (the example is shown in fig. 4 with brighter lines), to propose an expert to evaluate the degree of NPT1 influence on sub-objective-1 «Image rise» from the viewpoint of criterion K-1 «Profit rise for 25 %» - р1 C1 1¢, and then to evaluate analogically the probability of this evaluation realization - q1 C1 1.
To provide this idea to an expert of evaluation ways of рjkn¢ and qjkn it is convenient to use automated dialogue procedure, providing with combination of NPT, criteria and sub-objectives on the basis of theoretical and multiple procedure of “placement with repetitions”.
The generalized evaluation can be made on the basis of summing up the potentials obtained by different criteria, if the criteria supplement each other and the procedure of obtaining adaptive evaluation result is possible. But if the criteria are contradictive, i.e. can influence, for example, widening of product markets as well as narrowing of product markets, it is possible to introduce the evaluation results to the decision making people as column diagrams, the comparison of which would allow to compose agreeable or contradictive evaluation results and to make a decision taking into account compromises.
At the same time the observed way of using informational assessments does not yet solve all the problems of comparative evaluations of the components in the process of their implementation, and besides, there is still a necessity to obtain expert evaluation results of рi at the current moment, which is always difficult for experts, it is easier for them to give prognosis evaluation results of the degree of evaluated component influence on some perspective. That is why in a number of situations it is reasonable to supplement the observed evaluation method with informational model of the 2-nd type.
Informational models of the 2-nd type. The implementation of informational models of the 2-nd type is based on comparative analysis of complicated systems and problem situations during a certain initial period of their design (implementation, development) by way of comparing alterations of informational evaluation results in the flow of time. It is possible to use two ways of measuring Нi:
1) through probability рi¢ (1);
2) by means of determined features of the intercepted information:
at some moment (accepting simple average averaging-out, i.е. g =1):
Нi = Ji/ni; (2)
Using of the two ways to define Нi allows when there is known Нi (calculated through рi¢) and measured Ji to calculate:
ni = Ji/Нi. (3)
Then, having calculated the forecasted рik¢ by the end of the stage of the compared components implementation (new product kinds, technologies and so on), which is easier for a specialist to do rather than to give evaluation results рit¢ for the current moment while controlling their realization, it is possible to calculate Нik and ni, and then by assessments of criteria values (of direct or indirect features of the compared components) to define Jit at different moments of time (initial, current) and to calculate the values
Нit = Jit/ni (4)
Formation of objective (requirement)
structure
assessment рj¢ and qi
Calculation
Нi = - qi log (1 - рi¢)
Selection of DAi and
Averaging-out parameter g
Assessment Jik = Aik/DAk
Calculation ni
(при g =1 ni = Jik /Hik)
Assessment Jit= Aik /DAi
Calculation Hit = Jit/.ni
Рис. 5
for these moments of time by all the compared components, which can later be summed up to obtain generalized assessments of variants to calculate relative significance of separate components, i.е. to manipulate Н, expressed in universal relative units or bits as cost assessments.