=Paper=
{{Paper
|id=Vol-2318/paper2
|storemode=property
|title=A Procedure for Assessing the State of Cybersecurity of Power Grids
|pdfUrl=https://ceur-ws.org/Vol-2318/paper2.pdf
|volume=Vol-2318
|authors=Ihor Yakoviv,Vitaliy Tsyganok
|dblpUrl=https://dblp.org/rec/conf/its2/YakovivT18
}}
==A Procedure for Assessing the State of Cybersecurity of Power Grids==
https://ceur-ws.org/Vol-2318/paper2.pdf
A Procedure for Assessing the State of Cybersecurity of
Power Grids
Ihor Yakoviv1[0000-0001-7432-898X] and Vitaliy Tsyganok2[0000-0002-0821-4877]
1
Institute of Special Communication and Information Protection of National Technical Uni-
versity of Ukraine “Igor Sikorsky Kyiv Polytechnic Institute”, Kyiv, Ukraine
2
Institute for Information Recording of National Academy of Sciences of Ukraine, Kyiv,
Ukraine
iyakov52@gmail.com, tsyganok@ipri.kiev.ua
Abstract. Intensity of implementation, diversity, and damage from complex
cyber-attacks are constantly increasing. This process is further enhanced by on-
going expansion of attackers’ scope of knowledge of the weak spots of comput-
er networks and improvement of invasion technologies. Existing concepts of in-
formation system security assessment do not allow us to operationally consider
changes of factors, which determine the efficiency of protection.Cybernetic
segment of electric power grids has its own peculiar features, which, if properly
taken into consideration, can improve the efficiency of security assessment.
We suggest an approach to development of a procedure for operative as-
sessment of electric power grid cyber-security condition, based on knowledge
of the relevant threat level and of particular features of the processes related to
control of these grids.The paper includes the results of analysis of components
of the suggested assessment system, and defines the fundamental principles of
its functioning.
Keywords: Cyber-security, cyber-vulnerability, expert estimation, decision-
making support, electric power grid.
1 Introduction
A series of cyber-attacks on the control systems of electric power supply infrastruc-
ture of Ukraine (power grid) was detected in 2015-2016. On December 23, 2015,
when 225,000 clients of “Prykarpattyaoblenergo” joint-stock company got discon-
nected, we witnessed the first registered successful cyber-attack on the power system,
which brought it out of service. Electric power systems are an important component
of the national critical infrastructure. After these events many national security spe-
cialists from different states have to face the possibility of similar scenarios in their
respective countries.A special place within the complex of activities on power supply
infrastructure protection against cyber-attacks belongs to the procedure of current
cyber-security situation assessment (further referred to as Assessment). As a result of
this procedure a specialist, usually, has to get information (estimate) that would allow
him to provide answers to the following questions:
14
─ Is the level of protection adequate (inadequate) for the current level of cyber-
threats?
─ What is the extent of discrepancy between the current cyber-security level and the
required level?
─ Which actions should be taken in order to achieve the necessary level?
─ How to define the security assessment procedure for the specific system, and to
determine, whether it is adequate for the present situation?
Assessment procedure, in its turn, can be represented by the following intercon-
nected procedures:
─ Acquiring information on the current state of the assessment object (AO);
─ Normalization of the acquired information in accordance to estimation scales (ES);
─ Formation of estimates based on comparisons according to estimation criterion
(criteria).
In the situation of growing intensity of new complex cyber-attacks on the critical
infrastructure, based on constant expansion of knowledge of attackers on the vulnera-
bilities of computer systems, the problem of development of operational methods for
electric power grid cyber-security state assessment becomes extremely relevant.
In order to fill the Assessment concept with specific content (meaning), we
need to clarify the following aspects:
─ What is the assessment object (AO);
─ Which modern principles of cyber-security enhancement are used;
─ How existing approaches to IT-system security assessment can be implemented to
assess the cyber-security state of electric power grids.
Research objective is to define, analyze, and formalize the factors, which influ-
ence the system of operative evaluation of electric power grid’s cyber-security condi-
tion.
2 Analysis of assessment object
Modern power gridshave two basic components [1, 2]:
─ Power component (generating, distributing, supplying electric power to users);
─ Cybernetic (computer) component (monitoring of power processes and managing
their state).
The power component includes [2]:
─ Electric power stations (generating objects),
─ Electric power sub-stations (electric power transformation by transformers and
other equipment; distribution of electric power flows by commuting devices; con-
trol of electric power flow condition using different sensors and actuators);
─ Electric power supply networks, which connect power stations with sub-stations
and clients (consumers);
─ Subscribers’ metering devices.
15
The cybernetic component is generally called the Industrial Control System, or ICS.
Depending on control scales and tasks, ICS, in their turn, are divided into [3]:
─ Supervisory Control and Data Acquisition (SCADA) systems;
─ Distributed Control Systems, DCS;
─ Programmable Logic Controllers, PLC.
A PLC can be a part of a DCS, whilea DCS, in its turn, can be a part of a SCADA
system.
For management of electric power supply operator’s business processes separate
information technology systems of management (ITSoM) are used. Due to cyber-
security considerations, such systems should be distinguished from ICS [3], however
in practice this rule is often violated.Well-known cyber-attacks on electric power
grids of Ukraine were launched throughITSoM. This should be kept in mind in future,
when determining the essence of cyber-vulnerability level assessment.For example,
during clarification of assessment object composition we should distinguish between
the situations when ITSoM is within the cybernetic component of the power grid and
when it is not. At the first phase, it is appropriate to perform assessment for the case
when the “cybernetic component is just the ICS”, and at the second phase – for the
case when the “cybernetic component is both ICS andITSoM”.
3 Analysis of relevant cyber-security-ensuring principles
Analysis of existing corporate information and telecommunication system (ITC) pro-
tection practices allows us to single out two relevant strategies for cyber-attack coun-
teraction: reactive defense and proactive (preventive) defense.The common basis for
these strategies is provided by the following processes:
1) Observation (in real time) of the events within the assigned cyberspace segment;
2) Formation (using sensors), collection, and normalization of information on secu-
rity-related events in the unified operative processing center;
3) Analysis of events and making decisions on whether the cyber-attack is taking
place or not;
4) Making a decision on attack counteraction and implementation of this decision
through security actuators (executive security devices).
Under reactive strategy, decision on attack detection is made after the attack ends.
Counteractions might only help prevent a similar attack in future.Under proactive
(preventive) protection strategy, the attack should be detected before it ends. Insuch-
casethe time for implementation of activities to thwart (interrupt) the attack still re-
mains.
The key part of modern cyber-protection systems is the Cyber-Security Operations
Center (CSOC or SOC). Such centers, using operators and/or Security of Information
and Event Management (SIEM) tools, with different degrees of automation, imple-
ment the above-listed processes 1-4. Ukrainian normative and legal documents in the
field of information and telecommunication system protection neither explicitly de-
16
fine nor regulate real-time protection processes.As a rule, implementation of protec-
tion from unauthorized activities (UAA) within a complex information protection
system (CIPS)is performed based on reactive protection strategy. Information on se-
curity events is formed by the complex of means of protection (CPM)from unautho-
rized actions based on criteria (features), determined back at development
stage.Reaction (response) to security incidents is, as a rule, formed after the incident
is over.
Figure 1 represents the structure and key information processes of the operative
cyber-protection system.
database SOC
vulnerabilities CTI
(attacks)
Operators
2 3
Internet SIEM
1 security security
information command 4
Security sensors Security actuators
IT-system
SIEM – Security of Information and Event Management; CTI – cyber-threat intelligence
Processes: 1 – securityeventmonitoring; 2 – collection and analysis; 3 – decision on attack
detection; 4 – decision on response and its implementation
Fig.1.Model of processes of real-time cyber-protection system
Consecutive fulfillment of cyber-protection processes 1-4 represents a cycle that is
permanently repeated.In the beginning of the cycle Security sensors (based on Indica-
tors of Compromise, IOCs) track security events (or compromise). In case of an
event,security information is communicated to the SOC, where operators, using SIEM
software tools, analyze it as to compliance with security policy. In the case when a
cyber-attack is detected, a decision on counteraction is made. At the end of the cycle,
Security actuators, based on received security commands,implement this decision.The
background for sensor-based security monitoring isprovided by IOCs, detected by
Cyber Threat Intelligence (CTI). The sources of IOCs may be external databases of
threats and attacks (DataBase of Vulnerabilities/Attacks, DBV/A), or special methods
of external/internal CTI organization.
17
Broad-scale implementation of complex cyber-attacks of APT (Advanced Persis-
tent Threat) type against the national infrastructure became a powerful incentive for
development of proactive protection methods based on SOC. Characteristic features
of APTs are as follows.
─ The attack represents a complex set of malefactor’s actions, interconnected in
space and time. When taken separately, these actions might not seem suspicious;
─ The target action of the attack within the cyber-segment is prepared for a long time
(from several months to a year or more);
─ The combination of malefactor’s actions is a chain of tactic steps, allowing him to
achieve the goal of the attack. In spite of diversity of means, used in APTs, the set
of most tactics and their essence remain the same.
All these factors contribute to development of constructive methods for proactive
protection against APTs.
4 A model procedure for assessment of cyber-security
state of power grid
Above-mentioned analysis of modern real-time cyber-protection technologies allows
us to perform further formalization. For this purpose, let us consider the assessment
procedure as a system in the form of an information and functional structure (Fig. 2).
Such a structure (model) is the first level of assessment procedure formalization.It
allows us to specify the assessment problem through its decomposition into different
situations (scenarios) according to certain features.
For example, according to “cybernetic component composition”, the following list
of situations can be compiled (further on we refer to it as the scale of situations of
cybernetic component composition, SCCC, table 1).
Table1.
SCCC scale
CC-1 CC-2 CC-3 CC-4
ICS + + + +
ITS - + - +
SOC 24/7 - - + +
For situationCC-2 the information on the current CC state (Inf1) should reflect
more parameters than for CC-1 (NCC-2 >NCC-1, whereNis the number of parame-
ters). For CC-3 – NCC-3>NCC-2. Assessment for the situationCC-4 will be the most
complex from the standpoint of SCCC scale. In terms of this scale the issue of Inf1
relevance is also very important. Ongoing rapid development of CC intervention
technologies entails the need for increasing the assessment frequency, which can be
represented by assessment period (TAS).Situations when the assessment frequency
1/TAS = 2 per year, 1/TAS = 2 per month, and 1/TAS = 2 per hour are significantly
different from each other.In the first two cases there is time for implementation of
expert estimation methods to form Inf1 (that is, information based on knowledge of
18
the present situation, obtained from competent specialists, i.e. experts in the areas of
cyber-security).In the latter case, there is no time to use expert data-based approaches.
Power grid
(assessment object)
Cybernetic component
(CC)
SOC 24/7 ?
Norm. Compar. Estimate
Inf1
SS SA
ITSoM
ICS
? Inf2 Inf3 ?
Inf1 – information on the state of the cyber-component
S A Inf2 – estimation scale
Inf3 – estimation criteria
Power S – control sensors, A – control actuators
component SS – security sensors, SA – security actuators
ICS – industrial control system
SOC – security operations center
Fig.2.A model (information and functional structure) for assessment of cyber-security state of
electric power grids
At subsequent phase we have to addressthe question of semantic content of infor-
mation being used (Inf1, Inf2, Inf3–?). If information is anattribute (a set of attributes)
of an object, reflected in another object (attributive-transfer approach to the essence of
information, [4]), then the semantic (sense) of Inf2 (estimation scale) will be a set of
security parameters, characteristic for the given ICS, which were predefined based on
cyber-security policy. Semantic of Inf3 (estimation criteria) will be the set of values
of security parameters, predefined by security policy for the given ICS. According to
the logic of assessment structure under consideration, semantic meaning of Inf1 (cur-
rent information on the CC) for situation CC1 will be the current values of monitored
parameters in the estimation scale (Inf2).Once normalized, these current values are
compared to cyber-security criteria for the given ICS.Based on results of these com-
parisons, the overall estimate of cyber-security state is formed.
In addition, we should note that there is other necessary information, not reflected
in the structure, which, nevertheless, should be taken into consideration within as-
sessment procedure. We are talking about the list of possible security parameters,
19
based on which we can form estimation scales for different ICS configurations.Let us
call this information the “alphabet of ICS parameters”,Inf4.
Adequacy of the suggested model (structure) can be determined if we compare it to
the existing model of information and telecommunication system security assessment
(this term is used in the national Ukrainian normative documents, and its meaning is
similar to the term “IT-system”). Procedure for assessing ITS protection against unau-
thorized actions is regulated by such documents as НД ТЗІ 2.5-004-99 and НД ТЗІ
2.5-005-99 (ND TPI – the normative document on technical protection of informa-
tion) [5, 6]. The essence of the existing ITS assessment procedure can be briefly de-
scribed as follows[7]:
a) there is a structured general set of protection services (SPS). Every service cor-
responds with a respective known threat. The services are implemented by the respec-
tive protection means.All protection services are divided into 4 groups, depending on
the types of basic threats: 1) information confidentiality breach (C); 2) integrity viola-
tion (I); 3) accessibility (availability) violation (A); 4) observability violation (O);
b) based on the general SPS and selected security policy, the functional set of
services for a specific ITS is formed (the so-called functional protection profile, FPP,
of theITS), which complies with the key protection tasks;
c) according to the FPP, protection system developer defines and installs certified
protection tools within the ITS;
d) the estimate of security state is formed as part of the national expert examina-
tion based on verification of correspondence between the developed protection system
and the previously developed FPP.
Table 2 presents the results of comparison of the suggested assessment model and the
existing ITS security assessment procedure.
Table2.
ITS security assessment procedure Amodelofpower grid security as-
sessment
1 Structured general set of protection Alphabet of ICS cyber-security para-
services (SPS) meters (Inf4)
2 Functional protection profile, FPP Estimation scale (Inf2)
3 Every service of the SPS should be Estimation criteria (Inf3)
implemented within the ITS protec-
tion system
4 Availability of the protection service Current information on ICSstate
in the protection system (informa- (Inf1).
tion is defined in the process of state
expert examination)
Each kind of information, defined within the suggested model (information and
functional structure) of electric power grid cyber-security assessment corresponds
with an information object from the existing procedure of ITS security assess-
ment.The difference of the procedures should be determined through additional stu-
dies, which would take the following aspects into consideration:
20
- peculiar features of power (PC) and cybernetic (CC) component structures;
- peculiar features of interaction between PC and CC;
- requirements to efficiency of assessment procedure for ICS.
Specific features of interaction between PC and CC can be analyzed based on the
model of information processes in a cybernetic system and criteria of their security
[8]. At the same time, approaches to formalized representation of cyberspace and
cyber-security should also be taken into account [9].
5 Technology for group decomposition of a problem and
expert evaluation
When the frequency of information change allows us to allocate the time for prelimi-
nary analysis (not in real-time mode), it is appropriate to use technologies, in which
knowledge of experts and knowledge engineers is engaged for construction of a sub-
ject domain model. Such a model can be a constructed as a hierarchy of criteria
through decomposition process [10].Based on this model it is possible to assess the
state of cyber-vulnerability of critical infrastructures.
5.1 The essence of technology
The technology is intended for use in weakly structured domains where information
for substantiate decision making is insufficient and knowledge in the field is signifi-
cantly limited, non-formalized, and, mostly, distributed among highly specialized
experts. The technology is currently implemented as a web-based distributed comput-
er system [11], which allows knowledge engineers (expert session organizers) and
experts to work together remotely and provide knowledge, required to build an ade-
quate model.
The technology involves the following stages:
─ Formulation of the main criterion of evaluationby expert examination organizer
(decision-maker);
─ Formation of a group of competent experts for the examination;
─ Decomposition of the main criterion by the expert group. This stage is divided into
the following sub-stages:
Formulation by each individual member of the expert group of a set of factors
that, in his / her opinion, significantly influence formation of the estimate ac-
cording to the criterion (the relative impact of a sub-criterion should amount to
at least 10%);
When each of the experts participating in the examination has formed a set of
components of the criterion (most significant from his / her point of view), the
expert examination organizer performs grouping of all experts’ formulations ac-
cording to their semantic similarity;
Group choice of (by voting for) the best formulation in each group of formula-
tions with the same semantics. Voting provides an opportunity to vote for one of
21
the formulations, or refuse to chooseany of the formulations provided by mem-
bers of the group (if, according to the voter, the influence of the respective fac-
tor upon the given upper-level criterion is not significant enough, he can refuse
to make a choice).
As a result of this stage, the group of experts, coming to a consensus, forms the set
of the most important factors influencing a certain criterion. In the beginning, this is
the main criterion, formulated by the decision-maker (organizer), and in the process
of subsequent decomposition, this may be any criterion, division of which into com-
ponents makes sense. Thus, interconnected components form the generalized criterion
for evaluation.
Decomposition of the main criterion is carried out as a result of the repeated itera-
tive process, which is controlled by the expert examination organizer. This person, as
a knowledge engineer, decides, whether it is necessary to further decompose criteria,
that make up the components of the assessment system. For every such decomposi-
tion, a separate group of experts (who are most competent in the current issue under
consideration) may be formed [12]. The decision to stop criterion decomposition
process is made by the knowledge engineer in the case when it is possible to obtain
easily measurable, preferably, quantitative, indicator, characterizing the system ac-
cording to this criterion.
─ Group expert evaluation of the mutual influences of the criteria in the evaluation
system. At this stage, groups of experts, who conducted the respective particular
decomposition, evaluate relative direct impacts of factors upon higher-level criteria
in the hierarchy. As a result, relative impacts of all criteria in the hierarchy are de-
termined (using the methods of group expert evaluation).
Once this final stage is completed, the model of the evaluation (assessment) system
can be considered fully constructed.Now, based on this model, which is the structure
of interconnected factors of different significance, it is possible to determine the rela-
tive value of the estimate by the main criterion.
5.2 Anexample of power grid cyber-vulnerability state assessment
Let us illustrate the application of technology and the functioning of the systembythe-
following hypothetical example.
Let us assume, the knowledge engineer (a user, registered in the "Consensus-2"
system, with authority given by the expertise organizer (decision-maker)), formulated
the main criterion for assessing the state of cyber-vulnerability of electric powergrid
as "The level of protection of the cybernetic component of the power grid". The ex-
amination organizer, having appropriate rights in the system, authorizes a group of
users to work as experts.
After proper authentication, each expert, in accordance with his / her pertinent
knowledge level, formulates a list of the most important factors that influence the
level of protection of the cybernetic component of the power grid. Following the mu-
tual harmonization and unification (generalization) of the knowledge, provided by the
experts, the following formulations of the criteria-components of the evaluation sys-
22
tem are selected as the most appropriate ones: "preserving the integrity of informa-
tion", "protecting the confidentiality of information", and "ensuring the availability of
information". At this stage, the knowledge engineer has the opportunity to identify
these components of the system without consulting the experts, since such a decom-
position is provided in guidance documents on the organization of cyber defense [3].
Next, each of the above-listed criteria is decomposed by a separate group of ex-
perts. As a result, decomposition is completed. Within the graphical interface of the
expert session organizer the decomposition result might look as shown on Fig.3.
Fig. 3.Visual representation of the developed evaluation system modelwithin the interface of
"Consensus-2"system
Once the weights of the arcs in the graph are obtained (as values of relative impact
coefficients, determined through group expert evaluation), we get a weighted hie-
rarchy of criteria – a complete model for assessing the cyber-vulnerability of power
grids.
At the lower levels of the hierarchy of criteria, we have objective parame-
ters(indicators) of the system that the expert groups have identified as decisive, basic
ones for this evaluation. If we take such indicators for existing systems, and perform
calculations using available complex target-oriented dynamic alternative evaluation
methods [13], we obtain the relative indices of the state of cyber-vulnerabilities of
specific power grids. For the case when we have to evaluate one single power grid,
the evaluation is conducted in comparison with the established standard (benchmark)
– a hypothetical ideally protected system.
The described hierarchical network models allow us not only to perform compara-
tive assessment of the state of cyber-vulnerabilities in power grids, but also to optim-
ize allocation of resources, targeted at implementation of measures to improve cyber-
23
security of systems, and to build long-term cyber-defense organization plans for criti-
cal infrastructures [14, 15].
Conclusions
An original approach to the development of a procedure for assessing the state of
cyber-security of power grids is proposed. The approach is based on knowledge of the
level of actual threats and the peculiarities of the management processes in these sys-
tems.It is also based on real-time cyber-protection models and the system of cyber-
security state assessment, developed in the process of research.The models allow us to
classify the complexity of assessment procedures for different objects according to the
following properties:
─ Cybernetic component content;
─ Information of assessment procedure and its semantic meaning;
─ Time of assessment procedure information updating.
Based on the suggested approach, we can define the structure and content of cyber-
security assessment of a specific power grid, and verify the correctness of information
forming in assessment procedures.For instance, in grids, where cyber-component
includesreal-time protection system, information on system state should be formed
and processed with automation means (based on objective data, generated without
experts’ participation).However, experts can still participate in the formation of esti-
mation scale and criteria.
For performing assessment,we provide an opportunity to use group decomposition
technology, implemented within a distributed decision support system. The technolo-
gy allows us to utilize the subjective experience of the experts.
References
1. NISTIR 7628 Revision 1. Guidelines for Smart Grid Cybersecurity. National Institute of
Standards and Technology Interagency Report 7628 Rev. 1, Vol. 1 290 pages (September
2014).
2. Yuning Jiang; Manfred Jeusfeld; YacineAtif; Jianguo Ding; ChristofferBrax; Eva Nero. "A
Language and Repository for Cyber Security of Smart Grids", 22nd IEEE Enterprise
Computing Conference", 2018 10. 16-18, Stockholm, Sweden.
3. Stouffer Keith, Falco Joe, ScarfoneKaren Guide to Industrial Control Systems (ICS) Secu-
rity. Recommendations of the National Institute of Standards and Technology.NIST Spe-
cial Publication 800-82Rev. 2 (as of August 12, 2015)accessed
onhttp://dx.doi.org/10.6028/NIST.SP.800-82r2 .
4. IhorYakoviv, “The communication channel from the position of attributive-transfer nature
of the information”, Information technology and security, vol. 1, iss. 2, pp. 84-96, 2012.
Kyiv, Ukraine: Institute of special communications and information security NTUU “Igor
Sikorsky KPI”. [Online]. Available: http://its.iszzi.kpi.ua/issue/view/2838 .
24
5. ND TPI 2.5-004-99 Criteria for assessing the security of information in computer systems
from unauthorized access. Approved by order of DSTSZI of the Security Service of
Ukraine dated April 28, 1999 No. 22. (in Ukrainian НДТЗІ 2.5-004-99)
6. ND TPI 2.5-005-99 Classification of automated systems and standard functional profiles
of protection of the processed information from unauthorized access. Approved by order of
DSTSZI of the Security Service of Ukraine dated April 28, 1999 No. 22. (in Ukrainian
НДТЗІ 2.5-005-99)
7. Igor Yakoviv, Fundamentals of building a comprehensive information security system for
the information and telecommunication system. Tutorial. - Kyiv: National Technical Uni-
versity of Ukraine "KPI named after Igor Sikorsky", 2016 – 88p. (in Ukrainian)
8. Igor Yakoviv, “The base model of informational processes of management and safety cri-
teria for cybernetic systems”, Information technology and security, vol. 3, iss.1(4), pp.68-
73, 2015. Kyiv, Ukraine: Institute of special communications and information security
NTUU “Igor Sikorsky KPI”. Available: http://its.iszzi.kpi.ua/article/view/57735/53977 .
9. IhorYakoviv, “Infocommunication system, conceptual model of cyberspace andcyberse-
curity”.Information technology and security, vol. 7, iss.1(4), pp.68-73, 2017. Kyiv,
Ukraine: Institute of special communications and information security NTUU “Igor Si-
korsky KPI”. Available: http://its.iszzi.kpi.ua/article/view/57735/53977 .
10. Saaty T.L. The Analytic Hierarchy Process: planning, priority setting, resource allocation /
N.Y.: McGraw Hill. – 1980. – 287 p.
11. Свідоцтво про реєстраціюавторського права на твір №75023. Комп’ютернапрограма
„Система розподіленогозбору та обробкиекспертноїінформації для систем
підтримкиприйняттярішень – «Консенсус-2»” / Циганок В.В., Роїк П.Д.,
Андрійчук О.В., Каденко С.В. // від 27/11/2017.
12. Тоценко В.Г. Методы и системы поддержки принятия решений. Алгоритмический
аспект / ИПРИ НАНУ. – К.: Наукова думка, 2002. – 382с.
13. Циганок В.В. Удосконалення методу цільовогодинамічногооцінювання альтернатив
та особливостійогозастосування.Реєстрація, зберігання і обробкаданих. 2013. т.15,
№1.– С.90-99.
14. Tsyganok V, Kadenko S, Andriychuk O, Roik P. Usage of multicriteria decision-making
support arsenal for strategic planning in environmental protection sphere. J. Multi-
Crit.Decis. Anal. 2017; pp. 227–238.
15. Tsyganok V.V., Kadenko S.V. &Andriichuk O.V. Using different pair-wise comparison
scales for developing industrial strategies.International Journal of Management and Deci-
sion Making. – 2015. – vol. 14, issue 3. – P. 224-250.