=Paper=
{{Paper
|id=Vol-3806/S_61_Shevchenko_Syvytskyi
|storemode=property
|title=
Computer Simulation Model of the Organization at the Stage of Transformation for the Purpose of Adaptation to New Projects
|pdfUrl=https://ceur-ws.org/Vol-3806/S_61_Shevchenko_Syvytskyi.pdf
|volume=Vol-3806
|authors=Yuriy Syvytsky,Viktor Shevchenko
|dblpUrl=https://dblp.org/rec/conf/ukrprog/SyvytskyS24
}}
==
Computer Simulation Model of the Organization at the Stage of Transformation for the Purpose of Adaptation to New Projects
==
https://ceur-ws.org/Vol-3806/S_61_Shevchenko_Syvytskyi.pdf
Computer Simulation Model of the Organization at the Stage
of Transformation for the Purpose of Adaptation to New
Projects
Yuriy Syvytsky 1, Viktor Shevchenko 1
1
Institute of Software Systems National Academy of Sciences of Ukraine, 40 Academician Glushkov Avenue, building 5, Kyiv,
03187, Ukraine
Abstract
The article is devoted to the topical issue of creating a computer simulation model that allows optimizing
the transformation processes of the organization in order to adapt to new projects. The purpose of the
article is to increase the efficiency of large organizational structures by creating computer models that, on
the one hand, have a sufficient level of adequacy, and on the other hand, have a visual interpretation of the
main input parameters, which allows them to be easily determined on the basis of empirical data. In the
work, the analysis of existing studies is performed, the relevance of the problem is substantiated. The
concept of elementary atomic structural model is introduced, its inputs and outputs are considered. Mixed
and hierarchical structural models of the organization are considered. Examples of different levels of the
hierarchy of the structural model of the organization, ways of building up hierarchical structural models, as
well as ways of transforming mixed models into hierarchical models are considered. The relationship
between model errors and its complexity is considered. Recommendations on the level of complexity of the
model are provided. Analysis of existing exponential and linear models was performed. Reasoned adequacy
of logistic models. Logistic models are defined as the most universal development models that allow
modeling development processes in various fields of human activity. The differential form of logistic
models is considered. The ordinary differential equation of the logistic model is solved in order to obtain
the integral form of the logistic equation. Computer model parameters are introduced that are easily
determined numerically based on empirical data. A mathematical model of the useful effect of the
organization in the conditions of transformation was created. It establishes the dependence of the useful
effect on the input resource (time). The model is created as a combination of several logistic dependencies,
each of which is responsible for increasing or decreasing the useful effect. The model takes into account
the dependence of the growth of the useful effect for the main and new technologies, the decrease of the
useful effect due to the moral obsolescence of the technology, and the gradual decrease of the useful effect
due to the directive shutdown of the old technology. The structure of the model allows its scaling to more
complex development scenarios. The concept of the degree of insensitivity of the useful effect to small
amounts of input resources at the initial stages of the organization's development is introduced. Investigate
the dependence of the initial result on the degree of insensitivity. The model is implemented in the MatLab
algorithmic language
Keywords 2
Computer model, simulation model, logistic dependence, useful effect, resource, management decision
support, automation, optimization
1. Introduction
Each organization implements its own business processes within the existing organizational
structure. In turn, the projects of the organization are usually implemented within the framework of
the existing structure of the organization and within the framework of the existing structure of
business processes. This is not always the rule. For example, during the implementation of
14th International Scientific and Practical Conference from Programming UkrPROGβ2024, May 14-15, 2024, Kyiv, Ukraine
*
Corresponding author.
β
These authors contributed equally.
ys@intecracy.com (Y. Syvytsky); gii2014@ukr.net (V. Shevchenko)
0009-0008-9947-6653 (Y. Syvytsky); 0000-0002-9457-7454 (V. Shevchenko)
Β© 2024 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).
CEUR
ceur-ws.org
Workshop ISSN 1613-0073
Proceedings
Enterprise Resource Planning (ERP) systems, it has long been concluded that the successful
implementation of an ERP project is possible only on the condition of preliminary study, analysis
and correction of the enterprise's business processes, taking into account the limitations and
additional capabilities of ERP systems [1]. Some adjustments to business processes entail
adjustments to the organizational structure of the enterprise.
It is also common knowledge that in each class of tasks (projects), different organizational
structures of the enterprise have different effectiveness. Therefore, the transition to new classes of
tasks (projects) may also require a change in the organizational structure of the enterprise. Small
projects are usually carried out within the framework of already existing organizational structures.
But for the implementation of large projects, based on the scale of financing and possible losses in
case of failure, new organizations are often created or organizations that already existed are
transformed. The goal of the transformation is to obtain the most effective organizational structure
within the framework of a specific large project. Decisions about transformation paths in the latter
case are often made based on experience, feeling, inspiration, etc. The larger the scale of the
organization and the scale of the project, the more difficult the task will be. A very deep justification
of decisions is needed in order to convince the people on whom the financing of transformation and
development projects of organizations depends. The problem is aggravated by the fact that the
losses of large projects in case of failure are too great. The presence of fierce competition with
external organizations and among the organization's insiders does not make the issue any easier.
That is why the issue of optimization and justification of decisions regarding the directions and
ways of transformation of organizational structures when opening new projects is an urgent task.
Accordingly, the urgent task is to create computer models of the transformation of organizations
that would be part of software systems for supporting management decisions.
2. Analysis of existing studies
The creation of a computer model should begin with a thorough study of the modeling object, that
is, the main types of organizational structures, their development and transformation processes.
The works [2, 3] analyzed the main types of organizational structures: hierarchical, matrix,
mixed, project, etc. Disadvantage: there are no numerical estimates and even approaches to creating
numerical characteristics of the considered types of organizational structures of enterprises.
General methodological approaches to the selection of the organizational structure of the project,
a detailed analysis of the characteristics and options for using possible organizational structures are
given in [4]. In [5], in addition, a structured sequence of factors that should be paid attention to
when choosing the organizational structure of the project is provided. But, as before, there are no
numerical estimates in these works. The way out of the situation should be the creation of models
that take into account the dependence of the useful effect of the project depending on the type of
project structure
General approaches to building computer models of dynamic processes are considered in [6].
Corresponding numerical methods in [7, 8, 9]. Disadvantage: general approaches do not take into
account the specifics of modeling organizational structures, as well as the resources consumed by
the organizational structure.
In [1], an approach for analyzing the effectiveness of various organizational structures in the
dynamics of development over time is proposed. Disadvantage: abstract structures that do not
correlate with standard types of structures were considered: hierarchical, matrix, mixed, etc. [2]. In
addition, the possibility of a sudden change in the structure of the project (organization) over time is
not taken into account.
Optimization modeling of a sudden change in the structure of a dynamic system was considered
in [10]. Disadvantage: as objects of structural changes, dynamic systems were considered using the
example of aircraft, the dynamics of which, by their very nature, are significantly different from the
dynamics of projects and organizational structures.
Thus, the analysis of existing research revealed a contradiction between the need for a
methodical apparatus of decision-making support software systems for the transformation of
organizational structures and the absence of a single approach that would guarantee obtaining the
best decision for transformation. The main tool for supporting such decisions is computer
simulation.
The purpose of the article: to increase the efficiency of large organizational structures through
the creation of computer models. These models should have a sufficient level of adequacy to the
simulated processes. On the other hand, the models should have a visual interpretation of the main
input parameters, which would allow them to be easily determined on the basis of empirical data.
3. Structural and logical model of the organization
Organizational structures are complex large systems. Practical verification of the consequences
of management decisions in the real operation of such systems can be risky and cost a lot.
Therefore, to forecast the development of such organizations and predict the consequences of
certain management decisions, it is appropriate to use simulation modeling. An additional
advantage of computer simulation modeling is that it is based on mathematical models that are
universal for the development processes of organizations, technologies, projects, etc. The patterns of
development of the organization (technology) according to the stages of the life cycle are usually
described as the dependence of the useful effect of the organization on the input resources spent on
its creation (Figure 1).
π(π₯)
Input Output
(Resource) (Useful Effect)
Figure 1: An elementary atomic model of the development of technologies and organizations, as a
process of transforming input resources into an output useful effect.
We upload some or a combination of resources (such as materials, money, personnel,
information, know-how, reputation, experience, time, etc.) to the input. At the output, we receive a
useful effect from the application of technology or from the activities of the organization (for
example, income, profit, volume of production, volume of mineral extraction, reputation, experience,
level of quality according to various indicators, speed of solving problems, etc.). Some indicators at
the input and output are related or coincide completely. This suggests that the output values of one
model can be input to another. It can also be argued that the considered model of transformation of
input resources into an output useful effect using a certain procedure π(π₯), can be used as an
elementary atomic model from which more complex models can be created.
For example, the general structural and logical model of the organization can be presented in the
form of a hierarchy (Figure 2), in the nodes of which there are atomic models of the efficiency of the
use of various types of resources (personnel, production equipment, materials) (Figure 1). Atomic
models are useful because they are easier to create than creating a detailed model of the entire
organization at once. After that, you can increase the complexity of the model, add new levels of the
hierarchy, expand the number of atomic models at each level of the hierarchy, etc. Atomic models
significantly simplify the creation of complete structural-logical models due to better structuring.
And already on the basis of a structural and logical model, we can create a computer simulation
model of activity, development or transformation of an organization or technology.
Figure 2: Structural and logical model of the organization: 2 levels
In addition to the hierarchical model (Figure 2), other variants of hierarchical organization
models are possible, in which models of the effects of development (activity) of individual units are
used as atomic models of the second and subsequent levels. And if we have enough information
about the dynamics of the development of subdivisions depending on input resources, then such a
model can also be adequate and useful. But, unfortunately, most often the task is precisely to
develop a simulation computer model to the level at which the useful effects of units and the
organization as a whole are related to the input resources. These relationships should be represented
by clear mathematical or algorithmic dependencies that allow full-fledged simulation modeling. The
difficulty lies in the fact that in order to implement the last thesis, the complexity of the model has
to be increased, in particular by the levels of the hierarchy, both vertically and horizontally (Figure
3).
Organization
Personnel Π roduction equipment
Quantity Material support Quantity Material support
Quality Quality
Figure 3: Structural and logical model of the organization: 3 levels.
The approach of using hierarchical models is not ideal because real objects are not often pure
hierarchies. More often, they contain additions to the existing links of the hierarchy. For example, a
number of horizontal connections can be added at one level (Figure 4a). Or there can be the addition
of connections from the higher levels of one branch to the lower levels of another branch of the
hierarchy (Figure 5a), like the well-known "thick tree" topology of telecommunication networks.
Such structures, in contrast to purely hierarchical ones, will be called mixed. But the authors'
research showed that almost any mixed architecture (except for the one with cyclic sections) can be
transformed into a purely hierarchical structure. At the same time, the primary mixed and
transformed to a hierarchical structure will be identical from the point of view of the regularities
that connect the inputs and outputs of these models (Figure 4b, Figure 5b).
a) b)
Figure 4: Transformation of a mixed model with horizontal connections a) to a hierarchical model
b) by increasing the depth of the hierarchy.
Output (Useful Effect)
A1 A1
B1 B2 B1 B2
C1 C2 C1 C1 C2
Input (Resources) Input (Resources)
a) b)
Figure 5: Transformation of a mixed model of the "thick tree" type a) to a hierarchical model b) by
horizontal expansion at certain levels of the hierarchy.
The appearance of mixed models is often the result of trying to take into account as many
influencing factors as possible (in this case, internal). But excessive model complexity also reduces
accuracy. Works [11] show that the dependence of the square of the error on the complexity of the
model has the form of a parabola (Figure 6). That is, with a model that is too simple, the error is
large because many important influencing factors are not taken into account. And with an overly
complex model, errors increase because:
1. It is difficult to provide a complex model with representative input data.
2. Calculation errors are increasing.
3. Secondary factors receive an influence that is commensurate with the influence of the main
factors.
Figure 6: Dependence of the modeling error on the complexity of the model.
Thus, for any model there is an optimal level of complexity at which modeling errors are
minimal. It is easy to implement when the complexity is ensured, for example, by increasing the
order of polynomials of the model [12]. But things become much more complicated when the
complexity of the model is ensured by the choice of the depth of model detail, in particular the detail
hierarchy. The depth of detailing of the hierarchy depends on the set of influencing factors that are
representative of the process under study. To create an adequate model, it is necessary to select a
representative set of factors that will be taken into account. For this, it is always necessary to
implement at least the following points of the system approach:
Identification of influencing factors.
1. Building a rating of influencing factors.
2. Discarding secondary factors that have a minor impact on the processes being modeled.
3. Numerical assessment of influencing factors and their importance.
Depending on the specifics of the problem formulation, points 3 and 4 may change places. That
is, the rejection of secondary factors can occur not only on the basis of expert assessments, but also
on the basis of objective numerical indicators that are calculated.
Since any model is created in order to improve the management of certain processes, the issue of
optimizing this management may arise. In this case, it is necessary to formulate quality criteria,
according to which optimization will be carried out. In a significant number of cases, certain initial
values of the model can be directly taken as quality criteria. In other cases, aggregated indicators,
which are a combination of initial values, may appear as quality criteria. Aggregated indicators can
be abstract, or they can have a specific physical meaning that was not taken into account when
creating the model. In the latter case, the question arises of expanding the model to a state in which
physically understandable aggregated indicators will be calculated in certain blocks of the model. In
this case, the model will become completer and more adequate.
Modern processes of managing complex objects are based on well-known, so-called best
practices. In most cases, all the factors on the surface have already been taken into account.
Therefore, in order to obtain competitive advantages, the organization has to take into account an
increasing number of factors that were not taken into account before. For example, with regard to
personnel (Figure 7), the great potential of such factors is in the humanitarian sphere and requires
evaluation of not only objective, but also subjective indicators: morale, psychological type,
emotional intelligence, etc.
Personnel
...
A certain category of personnel
Quantity Quality Material support
Objectively: Subjectively:
- Experience in the position - Moral - psychological state
- General experience - Working conditions
- Education - Social package, etc.
- Class qualification
- Experience working in projects
- Course retraining
- ..
Figure 7: Structural and logical model of the organization: Personnel - 4 levels.
The problem is that models from the humanitarian sphere are very difficult to mathematical
formalization. Therefore, it is important to choose atomic models that are equally effective in
simulating processes in technical, economic, and social fields. The analysis of existing research
showed that logistic models are one of the most adequate models that can be equally successfully
applied in various fields of human activity.
4. Logistic models of development
Let's start with simpler models. In the conditions of the absence of restrictions on the
technologies used by the organization, the dependence of the useful effect on the spent resources
most often has the character of an exponent in the growth zone [1] (for example, the growth of
money placed on deposit or Moore's law - the growth of computing power). If the technology has
development limitations: due to limitations of the technology itself, limitations of production
scaling, limitations regarding accompanying input resources (for example, personnel), regulatory
limitations, etc., then the regularity of development often also has the character of an exponent, but
now it is already in the saturation zone, that is, approaching the asymptote, to which the
development process approaches from below. In general, part of the life cycle of development
consists of the stage of exponential growth and the stage of exponential entry into the saturation
zone. The transition between these two exponents is almost linear. To emphasize this property,
sometimes a separate section of linear development is added between the exponents, when the
useful effect on the output is strictly proportional to the amount of input resources. As mentioned
above, inputs can be materials, finances, personnel, intellectual property, image, etc. That is, the
input resource can be anything that can be turned into a useful effect.
If not limited to only one stage of the life cycle, researchers usually try to use more generalized
laws that cover both the stage of exponential growth and the stage of exponential saturation. In this
case, S-shaped dependencies are used, the most adequate, among which is considered logistic
dependence [1] (Figure 8). This dependence lies between two asymptotes and has the property of
central symmetry.
Figure 8: Logistic dependence of the useful effect of business (organization) depending on the cost
of the time resource.
Logistic dependence in differential form
ππ¦ (1)
= π (π¦ β π!"# ) π!"# β π¦
ππ‘
!"
has a clear physical interpretation. The left side of the equation corresponds to the growth
!"
rate of the useful effect π¦. The input resource is selected as a free variable. In our case, it is time π‘.
In the equation, there are two asymptotes π!"# and π!"# , which are located parallel to the
abscissa axis. The logistic dependence increases from the lower asymptote π!"# to the upper
asymptote π!"# . The growth rate of π¦ is proportional to the product of the distances y from the
lower π¦ β π!"# and upper π!"# β π¦ asymptotes. The scale of speed and, accordingly, the angle
of inclination of the logistic curve at the point of symmetry is determined by the coefficient π.
If the coefficients π, π!"# , π!"# do not change with respect to the free variable (time), then the
ordinary differential equation that specifies the logistic dependence has analytical solutions. To do
this, we will separate the variables, that is, we will place all the elements with the free variable π‘ to
the right of the equal sign, and we will place all the elements with the useful effect variable π¦ to the
left.
ππ¦ (2)
= π ππ‘.
(π¦ β π!"# ) π!"# β π¦
Let's transform the expression so that tabular integrals can be used
π(π¦ β π!"# ) π π!"# β π¦ (3)
β = π π!"# β π!"# ππ‘.
π¦ β π!"# π!"# β π¦
Integrate
ln π¦ β π!"# β ln π!"# β π¦ = π π!"# β π!"# π‘ + π. (4)
Convert to a more convenient look
π¦ β π!"# (5)
ln = π π!"# β π!"# π‘ + π.
π!"# β π¦
Find the exponent of both parts and write down the expression for the useful effect
π!"# + π ! ! !!"# !!!"# ! !! π!"# β π!"# (6)
π¦= ! ! ! !! ! !!
= π!"# + !(! ! !! ! !!)
.
1+π !"# !"# 1+π !"# !"#
Write the constant integration through already known constants
Ρ = β(π π!"# β π!"# Ξπ‘). (7)
Here, Ξπ‘ is the shift of the symmetry point along the abscissa axis. After substitution, we get the
expression for logistic dependence in integral form
π!"# β π!"# (8)
π¦ = π!"# + .
1+π ! ! !!"# !! !!" !!!!
In the future, we will use the generalized notation for part of the expression
1 (9)
ππΏ! π‘ β Ξπ‘ = ,
1+π ! ! !!"# !! !"# !!!!
which corresponds to the so-called SL-function [1]. The SL-function can be considered a
normalized logistic dependence that increases between asymptotes from 0 to 1.
Let's pay attention to the fact that the logistic dependence in the integral form also contains
many physically understandable components. But some of them are not quite convenient to find on
the basis of empirical data. Let's introduce additional notations:
π - upper asymptote,
π - lower asymptote,
π - the abscissa of the point of symmetry,
π - a logistic dependence constant.
The first three coefficients are only marked more succinctly. But the introduced additional
coefficient π significantly simplifies the calculation of the angle of inclination of the logistic
dependence at the point of symmetry. The coefficient π is equal to the lengths of the segments of
both asymptotes, that cut the perpendicular to the abscissa axis at the point of symmetry and the
tangent to the logistic curve at the point of symmetry. It is clear that under the condition of central
symmetry of the logistic curve, the lengths of these segments of both asymptotes are the same
(Figure 8).
This formalization of the integral form of recording logistic dependence significantly simplifies
finding the numerical values of the coefficients for the computer model of the organization's
development based on logistic dependence.
5. A computer model of the dynamics of the organization's development in
conditions of transformation
We will use logistic dependencies to build a general model of the development of the
organization at all stages of the life cycle: as at the stages of growth of the useful effect of the
organization
π¦ = π + (π β π) β ππΏ! π‘ β s , (10)
as well as at the stages of falling of the useful effect of the organization
π¦ = π β π β π β ππΏ! π‘ β s . (11)
In the second case, the logistic dependence is constructed with a negative value of the amplitude
(π β π).
In expanded form, the logistic dependence of the useful effect π¦ on time π‘ has the form
πβπ (12)
π¦ = ππΏ π‘ = π + ! .
! (!!!)
1+π !
The resulting useful effect of the organization's activity at any part of the life cycle (at any time π‘
of the life cycle) is found as the sum of various logistic components with positive and negative signs.
ππΏ! (π‘) = ππΏ! (π‘) + ππΏ! (π‘) + ππΏ!! (t) + ππΏ! (π‘), (13)
where
ππΏ! (π‘) - the dependence of the growth of the basic technology.
ππΏ! (π‘) - the dependence of the decline of the basic technology due to depreciation and moral
obsolescence.
ππΏ!! (π‘) - the dependence of the decline of the basic technology as a result of the management
decision to stop it to replace it with a more progressive one.
ππΏ! (π‘) - the dependence of the growth of new technology.
In this case, the argument of all dependencies is time π‘. Although in other cases, any other input
resource or a combination of different resources (material, financial, human, etc.) can be used as an
argument for the logistic dependence of the useful effect.
The general picture of the life cycle in the form of the dependence of the useful effect on the input
resource contains stages of growth of the useful effect, stages of decline, stages of repeated growth
after transformations aimed at eliminating the effects of moral obsolescence of the technologies used
by the organization. The moral obsolescence of technologies can be associated with the appearance of
new, more progressive technologies, with the actions of competitors, with the effect of consumers
getting used to certain technologies and the corresponding loss of interest, etc.
Computer simulation of the life cycle of the organization was performed for different initial
conditions and for different parameters of development dependencies. The main modeling scenario
included the growth of the useful effect of the organization after the introduction of certain innovative
technologies, the stage of saturation (the exit of the technology to the maximum of its possible
development), the fall of the useful effect due to moral aging, transformations regarding the change of
technology to a more progressive, repeated stage of saturation, a repeated drop in the useful effect.
More complex development scenarios in this model can also be implemented in a similar way. The
mathematical model was implemented in the form of a simulation model in the MatLab algorithmic
language. An example of simulation results is shown in Figure 9.
Figure 9: Constituent and resulting dependence of the useful effect of business depending on the
expenditure of the time resource.
As we can see, at the first stage of the growth of the effect, the dependence almost does not differ
from the usual logistic curve, since the main component of the development of the basic technology
ππΏ! (π‘) has the largest values. But already at this stage, the negative logistic dependence of moral
aging ππΏ! (π‘) technology is working, although its influence is still not very noticeable. In a certain
period of time, this leads to a local decrease in the resulting effect. This decline could last almost to
zero, but at a certain point in time, the transformation of the organization begins, which also implies
a change in technology to more advanced ones. That is, the new technology ππΏ! (π‘) starts working,
but at the same time the old technology ππΏ! (π‘) is turned off in a directive manner. At the software
level, this is implemented by adding the same value, but with a negative sign ππΏ!! π‘ = β ππΏ! (π‘).
As a result, the influence of the old technology is excluded ππΏ! π‘ + ππΏ!! π‘ = 0.
The depth of the drop in the useful effect due to the moral obsolescence of the technology
depends on the time of the beginning of the transformation of the organization π‘!"#$%& (transition
to a new technology). With the help of simulation, you can choose the moment of time π‘!"#$%& β
π‘! , π‘! , which will ensure the smallest drop in the useful effect
πππ₯ πππ ππΏ! (π‘) (
πΌ! = π‘
!"#$%& π‘ 14)
or the largest total useful effect during the period of operation π‘! , π‘! under the conditions of
transformation.
!! (
πππ₯
πΌ! = π‘ ππΏ! (π‘) ππ‘ . 15)
!"#$%&
!!
6. A study of insensitivity to small inputs
The initial periods of growth in the first stages of the life cycle have very slow growth. At the
same time, the reaction of the system to small amounts of input resources is almost the same as to
zero. For example, if there is an investment in the development of software production at the level
of 100 or 1000 dollars, then the output effect will be the same as for 0 dollars. To ensure at least
some noticeable useful effect at the level of the organization, the initial investment should be at the
level of several thousand or tens of thousands of dollars. This minimum investment will be different
for each type of technology. Similarly, spending time at the level of several hours or days, most
likely, will not bring a noticeable useful effect either. That is, there is a certain kind of insensitivity
of the system to small costs of input resources (financial, material, personnel, time). To study the
impact of insensitivity to small input values of resources, we will introduce the concept of
insensitivity, which will be measured as a percentage of the maximum possible useful effect. That is,
in fact, we will analyze not the value of the input resource, but the value of the useful effect ππΏ! (π‘)
to which it should lead. In the first stage of initial growth ππΏ! π‘ can be used instead of ππΏ! π‘ .
For example, the insensitivity of the system is defined at the level of Insensitivity = 0.1, and the
maximum possible value of the useful effect is equal to π¦!"# = π. To simplify the interpretation of
the result, π¦!!" = π = 0 was taken.
If for a certain input time resource, the useful effect is equal to ππΏ! π‘ = 0.05 π¦!"# , then the
initial value of the useful effect is taken to be equal to ππΏ! π‘ = 0.
If for a certain input time resource, the useful effect is equal to ππΏ! π‘ = 0.2 π¦!"# , then the
initial value of the useful effect is taken to be equal to ππΏ! π‘ = 0.2 π¦!"# .
Research on the dependence of the integral useful effect
!! (16)
πΈπππππ‘ = ππΏ! (π‘) ππ‘
!!
from the amount of insensitivity to small input resources Sensitivity for the entire period of
transformations π‘! , π‘! showed (Figure 10), that for Insensitivity = [0,0.5] the initial useful effect
almost does not change (it changes by no more than 4%). For setting development tasks at the
strategic level, errors are assumed at the level of 10-20% without a significant loss of the quality of
management decisions [1]. Therefore, a change in the useful effect at the level of 4% can be
considered equal to zero.
In the range of values of Insensitivity = [0.5,0.9], the change in the useful effect is very noticeable
and almost linearly decreases by 30%. The last 70% of changes in the useful effect of the system
occur in the range of Insensitivity = [0.9,1]. Most real projects correspond to the range of
Insensitivity = [0,0.5]. Very rarely, the development of projects can take place in the range of
Insensitivity = [0.5,0.9]. And almost never occurs in the range Insensitivity = [0.9,1].
Let's pay attention to the fact that the simulation results shown in Figure 9 correspond to the
indicator of Insensitivity = 0.
Figure 10: Dependence of the integral effect over a period of time on the level of insensitivity of the
system goal of development to small levels of the initial useful effect
7. Conclusions
The paper examines the main approaches to creating a computer model of the organization's
development in conditions of transformation of its structure (technological changes). The model is
intended for forecasting the development of the organization and finding optimal solutions for the
beginning of the transformation period.
It was found that computer models based on logistic dependencies are the most adequate for the
adopted formulation of the problem. Such models make it easy to find numerical values of
parameters based on empirical data.
The model as part of the management decision support software system allows to predict the
consequences of various management decisions in order to choose the best one.
The paper uses the criterion of maximizing the lowest level of the organization's useful effect
over the entire forecasting period. The integral criterion of the total useful effect that the
organization will receive in the event of the implementation of a specific transformation scenario is
also used.
The computer model is implemented in the MatLab algorithmic language.
Directions for further research: improvement of the model by expanding the list of factors taken
into account during modeling.
8. References
[1] V.L. Shevchenko, Optimization modeling in strategic planning, Military and Strategic Research
Centre of National Defence University of Ukraine, Kyiv, 2011.
[2] A Guide to the Project Management Body of Knowledge PMBOK, 6-th edition, Project
Management Institute, Inc., 2017.
[3] A Guide to the Project Management Body of Knowledge PMBOK, 7-th edition, Project
Management Institute, Inc., 2021.
[4] L. Nozdrina, V. Yashchuk, O. Polotaj, Project management: a textbook [for students of higher
educational institutions], Ministry of Education and Science of Ukraine, Ukoopspilka; Lviv,
commerce Acad.; Center for Education, Kyiv, 2010. ISBN 978-966-611-01-0030-4.
[5] P. Mykytyuk, Project management. Training manual [for higher education closed.], Ternopil,
2014.
[6] V.L. Shevchenko, V.I. Mirnenko, D.S. Berestov, R.M. Fedorenko, A.V. Shevchenko, Imitation
modeling. Mathematical modeling of processes: Study guide., in: V.L. Shevchenko (Ed.), Taras
Shevchenko National University of Kyiv, Kyiv, 2020.
[7] Sri.Nandakumar, Numerical methods. Study material. Core Course. B Sc Mathematics. VI
Semester. University of Calicut. School of distance education., 2011.
[8] V.L. Shevchenko, D.S. Berestov, M.V. Tkachenko, R.M. Fedorenko, Computing methods: Study
guide, in: V.L. Shevchenko (Ed.), Taras Shevchenko National University of Kyiv, Kyiv, 2019.
[9] Todd Young, Martin J. Mohlenkamp, Introduction to Numerical Methods and MatLab
Programing for Engineers, Ohio University, 2015.
[10] O.I. Lysenko, O.M. Tachinina, S.O. Ponomarenko, O.H. Guida, Theory of optimal branched
trajectories, National Technical University of Ukraine βIgor Sikorsky Kyiv Polytechnic
Instituteβ, Kyiv, 2023. ISBN 978-617-549-163-8.
[11] Yu.P. Zaichenko, Fundamentals of designing intelligent systems: [learning. manual], Slovo
Publishing House, Kyiv, 2004.
[12] Vladyslav Haydurov, Bohdan Yailymov, Andrii Shelestov, Air quality assessment model based
on satellite data based on the method of group consideration of arguments, International
Scientific Technical Journal Problems of Control and Informatics. 68. (2023) 93-106.
doi:10.34229/1028-0979-2023-5-8.
https://www.researchgate.net/publication/375158627_Model_ocinki_akosti_povitra_za_suputni
kovimi_danimi_na_osnovi_metodu_grupovogo_urahuvanna_argumentiv.