Operational Research in Engineering Sciences: Theory and Applications 
 Vol. 2, Issue 2, 2019, pp. 12-23 
ISSN: 2620-1607 
eISSN: 2620-1747 

 DOI:_ https://doi.org/10.31181/oresta1902001k 

* Corresponding author. 
E-mail addresses: nkomazec@gmail.com (Komazec), ale_petrovic@live.com (Petrović) 

 
 

APPLICATION OF THE AHP-VIKOR HYBRID MODEL IN 
MEDIA SELECTION FOR INFORMING ABOUT THE 

ENDANGERED IN SITUATIONS OF EMERGENCY 

Nenad Komazec, Aleksandar Petrović* 

University of Defence, Pavla Jurišića 33, 11010 Voždovac, Belgrade, Serbia,  
 

Received: 17 April 2019  

Accepted: 12 July 2019  

First online: 24 July 2019 

 
Research paper 

Abstract: A distribution of information in situations of emergency represents a serious 
challenge for the expert services engaged in protection and rescue. The number of the 
people who need help in situations of emergency is rather large and the number of those 
who can really be helped depends on their availability to expert services. A large number 
of people, especially endangered groups, can be saved with the help of timely and 
qualitative information. In the conditions determined by a lack of time, the staff in 
charge of situations of emergency have to make a decision on informing the population 
about the incoming danger. In the paper, a hybrid model based on the analytic hierarchy 
process (AHP) and multi-criteria compromise ranking (VIKOR) is presented, as applied 
through the selection of the best medium for informing the population in situations of 
emergency. The AHP method is used to determine criteria weight coefficients, while the 
VIKOR method is applied in order to find the best media by means of making a selection 
amongst numerous concrete options – i.e. alternatives. 

Key words: media, situations of emergency, AHP method, VIKOR method. 

1. Introduction 

Situations of emergency represent the state of the high endangerment of a social 
community. The consequences of situations of emergency are manifold and have far-
reaching effects. Considering the size of a danger from various natural disasters and 
other accidents, and different categories of the endangered population, timely 
warning and informing are of great significance. Preventive acting through informing 
and alerting the population is the basis for reducing the consequences of situations of 
emergency. There are various population categories that need be informed about the 
incoming danger. The most endangered are persons with special needs, only to be 
followed by women and children, and, in the end, the persons who are able to save 
themselves on their own.  

mailto:ale_petrovic@live.com


Komazec and Petrović/Oper. Res. Eng. Sci. Theor. Appl. 2 (2) (2019) 12-23  
 

13 
 

In a situation of emergency, the problem of the functionality of the media may 
occur due to their territorial prevalence and the signal reception (for the television 
and the radio) or equipment supply and functionality (the internet). Informing by 
emergency sirens could be a solution in the areas with good coverage and a high 
population density. Otherwise, the effect of  emergency sirens can be very small. 
Depending on the type of the situation of emergency, electricity supply can also be 
problematic (Komazec et al. 2014).  

The manner in which people should be informed is mostly restricted by the effects 
of a situation of emergency and its diffusion, and very frequently, the availability of 
information to certain groups of people is the only criterion (Акимов & Порфирьев, 
2004). However, by carrying out a thorough analysis of the relevant factors, 
conclusions can be drawn which refer to the selection of the optimal medium (or 
media) for the purpose of informing as many people as possible in order to select those 
media that meet the created needs. In this paper, the Analytic Hierarchy Process (AHP) 
and the Multi-Criteria Compromise Ranking (VIKOR) methods  are applied to problem 
solving. The contribution of the paper reflects in the enhancement of the evaluation 
and selection methodologies regarding the media for the purpose of informing the 
population in situations of emergency through a new approach to the treatment of 
imprecision due to the fact that the application of this model or similar models in 
situations of emergency has not been reviewed in the existing literature. 

2. Problem Description  

The paper is focused on finding out the hybrid model which will enable the optimal 
selection of the media for informing the population in situations of emergency. 

The occurrence of the need for informing the population in the situations of natural 
disasters and technical-technological accidents depends on the level of the 
endangerment of the social community. The alert signal announcing a danger is 
activated by the authorities according to the law. The level of the danger, i.e. the 
endangerment, is the basis for the proclamation of a situation of emergency (Karovic 
& Komazec, 2009). A situation of emergency is proclaimed by the staff in charge of 
situations of emergency when risks and threats, or the resulting consequences are on 
such a scale and of such an intensity that they cannot be stopped or diminished by 
conducting the authorities’ regular activities, for which reason it is necessary that 
special measures, additional strengths and the means with an enhanced operation 
mode should be taken for the purpose of their mitigation and removal (Zakon o 
smanjenju rizika od katastrofa i upravljanju vanrednim situacijama [Law on Disaster 
Risk Reduction and Emergency Management], 2018; Pamucar et al. 2016). 

The proclamation of a situation of emergency follows immediately after becoming 
aware of the danger. This moment is a milestone in the protection and salvation of the 
endangered population, material goods and the environment. Namely, as long as the 
staff in charge of situations of emergency  are unaware of a danger, they cannot 
proclaim a situation of emergency, nor can they inform the endangered population 
about it; competent services (Republic Hydrometeorological Service of Serbia (RHMS), 
Republic Geodetic Authority (RGA), etc.) are, however, responsible for informing the 
population. The competent services usually inform people through the media (the 
television and the radio) and via the internet (posting warnings on relevant websites). 
In that period of time, the staff in charge of situations of emergency collect pieces of 



Application of the AHP-VIKOR hybrid model in media selection for informing about the 

endangered in situations of emergency 
 

14 
 

information and may perform such informing through local communities and 
responsible individuals. 

In some special situations, the electric-power industry, the water industry and the 
other business associations using hydro-accumulations and landfills are obliged to 
ensure that the population is timely informed about the incoming danger (Zakon o 
smanjenju rizika od katastrofa i upravljanju vanrednim situacijama [Law on Disaster 
Risk Reduction and Emergency Management], 2018). 

The members of the staff in charge of situations of emergency may use local radio 
and television stations. The transmission of information to the endangered population 
carried out by the Republic staff in charge of situations of emergency The law also 
envisages the obligation of mobile companies to transfer information to endangered 
people. All mass-media means are applicable when informing the population is 
concerned, even before situations of emergency occur (Petrovic et al. 2017). A special 
problem is a situation of emergency when a danger to the population, material goods 
and the environment has arisen. The conditions of all the people inside the endangered 
territory are such that they all fear for themselves, for their families, and for their 
material possessions. There is a similar situation in business companies which, 
simultaneously having to protect their own assets, also need to engage the employees 
whose families are jeopardized at that moment as well. In the case of a concrete 
problem, persons with special needs, the elderly, women and children are considered 
as special and specific groups of people.  

In the case of a particular problem, the means of mass communications are 
restricted to a segment of the mass media (the television, the radio and the internet – 
especially social networks and mobile communications)1 (Radojkovic & Miletic, 2005). 
There is a possibility of using print media, but this way of communication is restricted 
by the type and level of the influence of the concrete danger.  

2.1 Informing in Situations of Emergency  

The practical usage of effective informing is the basis for effective management 
(Moriarty et al. 2012). Informing in situations of emergency (and alerting, too) is an 
activity of great significance with respect to decreasing human casualties and 
mitigating damage to material goods and the environment. Namely, timely 
information provides a quick and right reaction of the endangered population to the 
danger. Timelessness depends on the type of danger (Dey, 2001). Situations of 
emergency and other accidents which may occur suddenly and develop rapidly are 
more complex to announce. Practically, their announcement is conducted after the 
moment of their occurrence. The possibility that the majority of the population will 
not receive information on time is most likely (Komazec et al. 2018). The dangers that 
occur in a longer period of time and develop gradually are much easier to announce, 
along with the appearance of the first indicators.  

The staff in charge of situations of emergency and the authorities’ specialized 
institutions (RHMS, RGA, etc.) have the legal obligation to provide information in 
situations of emergency. This approach is essential for controlling the information 
flow, the types of information, and the process of receiving information to as many 

                                                        
1The presence of social media is generally implied, due to the fact that they include mobile companies and 
their ability to transfer information as well. 



Komazec and Petrović/Oper. Res. Eng. Sci. Theor. Appl. 2 (2) (2019) 12-23  
 

15 
 

people as possible. For the purpose of informing effectively and efficiently, the staff in 
charge of situations of emergency have several instruments at their disposal, namely: 

1. the television and the radio; 
2. the Internet-social networks; 
3. the early warning, informing and alerting system; 
4. print media and 
5. mobile telecommunications. 
The television and the radio belong to a group of highly widespread and available 

media. It is to be assumed that every single home has the ability to access them. The 
main issue in the usage of such media is the ability of the municipality staff in charge 
of situations of emergency to send information through the national television and 
radio network due to the fact that the national services are watched/listened to by a 
large number of individuals. Also, there is a problem in local services in the territory 
of the municipality and their availability throughout that territory.  

The Internet is also a widely applicable instrument for the transmission of 
information. There is a certain limitation when access to the internet is in question. It 
is possible to quickly transfer information to a large number of people throughout 
social networks, but the availability of those individuals to the staff in charge of 
situations of emergency may be an issue. A special problem implies those elderly ones 
who do not use the internet at all, or use it poorly.  

The early warning, informing and alerting system is directly available to the 
staff in charge of situations of emergency. The limitation lies in the operational 
correctness of the system, the territorial coverage and the ability of all endangered 
groups to understand the sent signals. 

Print media belong in the group of slower and mass means of information transfer. 
The limitation of their application lies in the fact that, in a situation of emergency, the 
distribution of such media may be stopped. Also, not every municipality owns its own 
print media, which may refer to the dependence on a publishing house, its distance 
and capacity.  

Mobile telecommunications represent a powerful, widespread and easily 
accessible medium for information transfer. A large number of people in all 
endangered groups own a mobile telephone. The main issue is the development of a 
database of telephone numbers, especially of the numbers of the endangered groups 
of individuals responsible for them.  

2.2. Description of the Media Selection Criteria 

In order to successfully apply the AHP and VIKOR methods to solving the research 
problem, it is necessary to identify the criteria common to all the listed and considered 
media of mass communications and among which a selection of the best media for 
informing the population in situations of emergency will be conducted. (Nenadovic et 
al. 2016). Taking this into account, the following criteria are identified: 

K1 – The frequency of informing – This criterion is expressed by the number of the 
repetitions of informing through the amount of time in order to achieve as good 
reception as possible by as many individuals as possible.  

K2 – Territorial coverage – This criterion is expressed in percentages and 
represents the ability to receive information in real time throughout the territory of 
the municipality.  



Application of the AHP-VIKOR hybrid model in media selection for informing about the 

endangered in situations of emergency 
 

16 
 

K3 –Presence in a target group– This criterion is expressed in percentages and 
reflects the presence of a concrete medium in a concrete target group, or the other way 
round – it reflects the percentage of the target group ‘consuming’ a particular medium. 

K4 – The availability of the medium – This criterion represents the coverage of the 
Republic of Serbia’s territory (by broadcasting or by distribution)2 and it is described 
by linguistic descriptors given in Table 1. 

Table 1. The descriptive scale of the linguistic criteria 

Linguistic descriptors Very poor Poor Medium Good Excellent 
Assigned numerical value 1 2 3 4 5 

K5 – The medium access price – This criterion is expressed by cash units and 
accounts for the amount which is necessary to pay in order to make the content of a 
particular medium available.3 

The characteristics of the listed criteria are presented in Table 2. 

Table 2. The criteria characteristics 

 ben4. cost5 Qualitative Quantitative  
К1 +   + 
К2 +   + 
К3 +   + 
К4 +  +  
К5  +  + 

3. Applied Methods  

The hybrid model used for the selection of the best media for informing the 
population in situations of emergency consists of the AHP and VIKOR methods. The 
AHP method is used to determine the weight coefficients of the identified criteria, 
while the VIKOR method is used to find a compromise solution, specifically for the 
selection of the optimal informing medium. 

                                                        
2 When electronic media (the television and the radio) are concerned, it is significant whether they are the 
media with a national frequency or the media covering only a certain region in Serbia, whereas when the 
press (newspapers) is concerned, it is essential whether they are the media distributed throughout the 
territory or the media distributed locally. 
3 When speaking about electronic media, the total amount represents the sum of all expenses, such as 
purchasing a television set or a radio receiver, the costs of electricity, broadcasting costs, a special fee for 
using a public service, etc., whereas in the case of the press, it accounts for the amount which has to be paid 
for certain newspapers, magazines and so forth. For the internet and mobile commu nications, it is the price 
for those services. 
4 The subset of the criteria with the benefit characteristics, which means that a higher value of the criterion 
is preferable, i.e. better. 
5 The subset of the criteria with the cost characteristics, which means that a lower value of the criterion is 
preferable, i.e. better. 
 



Komazec and Petrović/Oper. Res. Eng. Sci. Theor. Appl. 2 (2) (2019) 12-23  
 

17 
 

3.1. AHP 

The AHP method developed by Thomas Saaty at the beginning of the 1970s is a tool 
used in decision analysis, created for the purpose of providing assistance to decision-
makers in resolving complex decision-making problems in which many decision-
makers participate, numerous criteria and in various time periods. This process is 
based on the balance concept used in order to determine  the overall significance  of 
the set of relative attributes, activities or criteria, and relates to the analyzed decision-
making problem (Cupic & Suknovic, 2010). In the paper, this method is applied so as 
to determine the criteria weight coefficients regarding the selection of the media for 
informing the population in situations of emergency. Saaty’s standard nine-level scale 
presented in Table 3 is applied in order to carry out a pairwise comparison (Saaty, 
1980). Saaty’s scale is applied by the decision-makers or the analysts performing 
comparisons in pairs on the basis of the semantic preferences from the left-hand 
column of Saaty’s scale or by direct association. The numerical values stated in the 
columns 2 or 3 of Table 3, which correspond to the semantic preferences in the left-
hand column, are entered into the square comparison matrix, Equation (1). 

Table 3. Saaty’s pairwise comparison scale 

                  𝐶1  𝐶2  … 𝐶𝑛 

A = 

𝐶1
𝐶2
⋮

𝐶𝑛

[

𝑎11 𝑎12 … 𝑎1𝑛
𝑎21 𝑎22 … 𝑎2𝑛
⋮ ⋮ ⋱ ⋮

𝑎𝑛1 𝑎𝑛2 … 𝑎𝑛𝑛

]                                 (1) 

Since it is true that aij= 1/ aji and aii= 1 for every i,j = 1,2,...,n, the matrix A is positive, 
symmetrical and reciprocal. When applying Saaty’s classical scale, the relations in a 
pairwise comparison are strictly defined (Pamucar et al, 2016).  

3.2. VIKOR 

The VIKOR method was developed by Opricovic Serafim (Opricovic, 1998) based 
on the elements from compromise programming with the beginning at the “border” 
forms of Lp-metrics. These metrics represent the distance between the ideal point F* 
and the point F(x) in the space of the criteria functions (Petrovic et al. 2017). 

The first step in the VIKOR method is the initial decision matrix: 

                   𝑋1  𝑋2   𝑋3    ⋯  𝑋𝑛
                   𝑊1  𝑊2   𝑊3   ⋯  𝑊𝑛

 

Definition  Standard values Reciprocal values 
Equal importance  1 1 

Weak importance of one over 
another 

3 1/3 

Essential or strong 
importance 

5 1/5 

Demonstrated importance 7 1/7 
Absolute importance 9 1/9 

Intermediate values between 
the two adjacent judgments  

2,4,6,8 1/2, 1/4, 1/6, 1/8 



Application of the AHP-VIKOR hybrid model in media selection for informing about the 

endangered in situations of emergency 
 

18 
 

D =    

𝐴1
𝐴2
𝐴3
⋮

𝐴𝑚 [
 
 
 
 
𝑋11 𝑋12 𝑋13 ⋯ 𝑋1𝑛
𝑋21 𝑋22 𝑋23 ⋯ 𝑋2𝑛
𝑋31 𝑋32 𝑋33 ⋯ 𝑋3𝑛
⋯ ⋯ ⋯ ⋱ ⋮

𝑋𝑚1 𝑋𝑚2 𝑋𝑚3 ⋯ 𝑋𝑚𝑛]
 
 
 
 

                                                                                     (2) 

By the decision matrix, the m alternatives and the n criteria are defined. Every 
criterion is associated with its weight coefficient 𝑤𝑖. The weight coefficients of the 
criterion should follow the next condition: 

∑ 𝑤𝑖

𝑛

𝑖=1

=1                                                                                                                                             (3) 

After defining the decision matrix, the method is to be applied. The next step in the 
VIKOR method is the determination of 𝑥𝑖 

∗  and 𝑥𝑖
−, which is conducted by the following 

equations: 

𝑥𝑖
∗ = max (𝑥1, 𝑥2,…, 𝑥𝑛); i=1,2,…, n;            (4) 

𝑥𝑖
− = min (𝑥1, 𝑥2,…, 𝑥𝑛); i=1,2,…, n;            (5) 

The next step in the VIKOR method is the determination of the pessimistic (𝑆𝑗 ) and 

the anticipated (𝑅𝑗 ) solutions. 

𝑆𝑗 = ∑ 𝑤𝑖

𝑛

𝑖=1

(𝑥𝑖
∗ − 𝑥𝑖𝑗)  (𝑥𝑖

∗ −  𝑥𝑖
−)⁄ ;  𝑗 =  1,2, … , 𝑚                                                        (6) 

𝑅𝑗 =
max

𝑖
[𝑤𝑖 (𝑥𝑖

∗- 𝑥𝑖𝑗 ) / (𝑥𝑖
∗ −  𝑥𝑖

−); 𝑗 =  1,2, … , 𝑚                                                                      (7) 

After that, the next step is finding a compromise solution 𝑄𝑗 : 

𝑄𝑗 = 𝜈
𝑆𝑗− 𝑆

∗

𝑆−− 𝑆∗
 + (1–𝜈) 

𝑅𝑗− 𝑅
∗

𝑅−− 𝑅∗
; j=1, 2,…, m            (8) 

where 

𝑆∗= min 𝑆𝑗             (9) 

𝑆−= max 𝑆𝑗              (10) 

𝑅∗= min 𝑅𝑗              (11) 

𝑅−= max 𝑅𝑗              (12)  

𝜈 – the weight of the strategy satisfied according to the majority of the criteria, 𝜈 ∈
{0.25, 0.5, 0.75}. 

The last step in the VIKOR method is the ranking of alternatives. A set of 
alternatives can be ranked by the value of the function of the criteria assigned to each 
alternative 𝑄𝑗. The best alternative is the one that is the least distanced from the ideal 

value, i.e., the one that has the minimal 𝑄𝑗value, and vice versa. As relevant, the rank 

list is taken for the value 𝜈 = 0.5, but even though it is the first on the list, that action 
has to meet two more conditions (Petrovic et al. 2017), namely: 



Komazec and Petrović/Oper. Res. Eng. Sci. Theor. Appl. 2 (2) (2019) 12-23  
 

19 
 

1) There has to be sufficient advantage (more than the “minimum sufficient 
advantage”) related to the 2nd, 3rd, and other alternatives), which is established by 
applying the following expression: 

Q(𝑎‚) - Q(𝑎‚‚)≥ 𝐷𝑄             (13) 

where: 

DQ = min(0.25, 
1

𝑚−1
)             (14) 

where 𝑎‚ and 𝑎‚‚ represent the values of the 1st and the 2nd alternatives, respectively, 
by 𝑄𝑗(𝜈=0.5), and m represents the number of the alternatives. The minimum 

sufficient advantage is to be 0.25 in the cases when there is a small number of 
alternatives. 
2) It has to have a sufficiently stable position, i.e. position no. 1, according to the rank 
list QSj, or according to QRj, or according to Q for 𝜈 = 0.25 and 𝜈 = 0.75 where (Petrovic 

et al. 2017): 

Q𝑆𝑗 = 
𝑆𝑗 − 𝑆

∗

𝑆−− 𝑆∗
; j = 1, 2,…, m                                          (15) 

Q𝑅𝑗 = 
𝑅𝑗− 𝑅

∗

𝑅−− 𝑅∗
; j = 1, 2,…, m                           (16) 

4. Presentation of the Application of the Hybrid Model 

As already stated, the hybrid model consists of the AHP and VIKOR methods. The 
weight coefficients of the criteria are calculated by applying the AHP method in the 
Expert Choice program package and the results of that process are shown in Tables 4 
and 5. 

Table 4. The criterion pairwise comparison according to Saaty’s scale 

 K1 K2 K3 K4 K5 
K1 1.0 3.0 1/2 9.0 2.0 
K2  1.0 (4.0) 2.0 2.0 
K3   1.0 5.0 3.0 
K4    1.0 (3.0) 
K5     1.0 

CR=0.03 

Table 5: The values of the criteria weight coefficients 

K1   .300       
K2   .092       
K3   .406       
K4   .051       
K5   .151       

Inconsistency = 0.03        
with 0 missing judgments.        



Application of the AHP-VIKOR hybrid model in media selection for informing about the 

endangered in situations of emergency 
 

20 
 

For the purpose of applying the VIKOR method, a total of 6 different alternatives 
were chosen (from A1 to A6 ), by which the initial decision matrix was defined, which 
is accounted for in Table 6. 

The alternatives are as follows: Alternative A1–Local television and radio station; 
Alternative A2–National television and radio station; Alternative A3–Early warning, 
informing and alerting system; Alternative A4–Internet–social networks; Alternative 
A5–Print media, and Alternative A6–Mobile communications. 

The details regarding the listed alternatives are not presented in the paper in order 
to avoid a decrease in their positions in the media space and favoring certain media, 
too. 

Table 6. The initial decision matrix 

Criterion K1 K26 K3 K4 K5 
Alternatives 0.300 0.092 0.406 0.051 0.151 

A1 6007 50 45 5 6 
A2 300 48 60 4 5 
A3 600 5 35 3 0.3 
A4 750 4 40 2 0.35 
A5 480 22 10 4 2.4 
A6 430 18 8 1 1.8 
xi* 750 50 60 5 0.3 
xi- 300 4 8 1 6 

Characteristic of criterion max max  max  max  min 

By solving the equations from 4 to 12, the final solutions are obtained and they are 
presented in Table 7. 

Table 7. The final values of the alternatives 

Alternatives QSj QRj Qj (v=0.5) Qj (v=0.25) Qj (v=0.75) 
A1 0.463 0.252 0.357 0.305 0.410 
A2 0.353 0.576 0.465 0.520 0.409 
A3 0.040 0.156 0.098 0.127 0.069 
A4 0.000 0.000 0.000 0.000 0.000 
A5 0.938 0.936 0.937 0.936 0.937 
A6 1.000 1.000 1.000 1.000 1.000 

According to the results obtained, the final ranking of the alternatives is as follows: 
A4, A3, A1, A2, A5 and A6. 

                                                        
6 According to the last analysis of the media market in Serbia, conducted by Ipsos Strategic Marketing Agency 
in 2015 (Regulatory Body for Electronic Media, 2015), the television is the leading medium, with a market 
share of 53%, the press accounts for 20%, and the radio accounts for 4% (the other media account for a 
market share of 23%).  
7 The number of repetitions in one day (i.e. 24 hours).  



Komazec and Petrović/Oper. Res. Eng. Sci. Theor. Appl. 2 (2) (2019) 12-23  
 

21 
 

5. Sensitivity Analysis 

When applying methods of multi-criteria decision-making, it is crucial to examine 
the sensitivity of the mathematical model applied, so that decision-makers could have 
some kind of guarantee according to the rationality and quality of the obtained 
solution (Pamucar et al. 2016). The analysis of the sensitivity of the results obtained 
by the hybrid model implies the examination of changes in the weights of criteria and 
on the consistency of the solution with respect to a change in the measurement scale 
(Pamucar et al. 2017). 

When examining change in the weight of the criteria, a total of six scenarios were 
developed (Table 8) (A – the equal importance of all the criteria, B – the absolute 
dominance of K1, C – the absolute dominance of K2, D – the absolute dominance of K3, 
E – the absolute dominance of K4, F – the absolute dominance of K5). Within the 
framework of the independence analysis regarding change in the measurement scale, 
a total of two scenarios were developed (Table 9). In the first scenario, the qualitative 
criterion (K4) was given by the two different scales (S1 and S2) connected by a positive 
affine transformation (y = 2x – 1). In the second scenario, the quantitative criterion 
(K5), which represents the media access cost, expressed in cash units is was given by 
the two different scales: (S1) in RSD (Republic of Serbia’s Dinar) and (S2) in euros. 

Table 8. The sensitivity analysis of change in the weights of the criteria 

Scenario A B C D E F 
Alternatives Alternatives rank  

A1 4 4 3 5 3 2 
A2 1 2 5 3 4 3 
A3 5 6 2 4 1 6 
A4 3 1 1 1 5 1 
A5 6 3 6 6 6 4 
A6 2 5 4 2 2 5 

Table 9. The independence analysis of change in the measurement scale 

Scenario 
Scenario 1 Scenario 2 

S1 S2 S1 S2 
Alternatives Alternatives rank 

A1 5 5 5 5 
A2 3 3 3 3 
A3 2 2 2 2 
A4 1 1 1 1 
A5 5 5 5 5 
A6 6 6 6 6 

6. Discussion and Conclusion  

According to the results obtained by conducting a sensitivity analysis of the 
developed hybrid model for the purpose of the selection of the best medium for 
informing the population in situations of emergency and with respect to the research 
studies (Pamucar et al. 2018), a conclusion can be drawn that the hybrid AHP-VIKOR 



Application of the AHP-VIKOR hybrid model in media selection for informing about the 

endangered in situations of emergency 
 

22 
 

model is completely applicable in the cases of solving the treated problem and satisfies 
the set goal. The sensitivity analysis of change in the weights of the criteria shows that 
the hybrid model is sufficiently sensitive and that it keeps alternative priorities (in this 
particular case, Alternative A4 is favored). Furthermore, checking the consistency of the 
solution by changing the measurement scale shows that the model is stable and that it 
generates sustainable solutions. By an analysis of all of the results obtained, it is possible 
to conclude that the application of the AHP and VIKOR methods can significantly help 
decision-makers to come to the necessary solution. 

The proposed model examined in the paper represents an integration of the AHP and 
VIKOR methods, where the AHP method is used to determine the weight coefficients of 
criteria within the process of the selection of the best medium for informing the 
population in situations of emergency, whereas the VIKOR method is used to rank the 
obtained alternatives and find the optimal solution. The model has been verified through 
the media selection process inside the territory of a certain municipality by six different 
alternatives. The results obtained by the application of the model show that Alternative 
no. 4 is the best solution in all the scenarios with different values of the criteria. In 
comparison with the hybrid model, Alternative no. 4 has a priority engagement. Taking 
into consideration the fact that situations of emergency are concerned in this case, it is 
not only enough to depend on one single medium, but the competent staff in charge of 
situations of emergency will engage all available media. This means that informing the 
population would certainly be performed through the television, the radio, print media, 
and mobile communications. The early warning, informing and alerting system would be 
used for signal transmission. The sensitivity analysis has shown the stability of the 
results obtained by the application of the model in all of the considered scenarios. 

The presented application of the hybrid model provides an unbiased aggregation of 
experts’ choices by taking into consideration all the inconsistency and subjectivism of 
group decision-making. Apart from the expressed contribution, it is essential to 
emphasize the authors’ attempt to apply this model in  situations of emergency, which 
are characterized by uncertainty and a lack of time as well, the large amount of 
information and crisis decision making. The development of such models contributes to 
the literature in which the theoretical and practical application of multi-criteria 
techniques is subjected to review. The suggested model enables the evaluation of 
alternatives despite the imprecision and lack of quantitative information in the decision-
making process. By applying the developed approach, problems concerning multi-
criteria decision-making and the evaluation and selection of a medium for informing the 
population in situations of emergency can easily be dealt with. The model can be applied 
to making various decisions. It is also applicable in the process of decision-making within 
the staff in charge of situations of emergency in situations of emergency. The flexibility 
of the model is proven by the fact that its verification can be performed by applying any 
type of multi-criteria decision-making methods.  

Further research studies regarding this paper should refer to the application of this 
and similar models in combination with other methods and the development of a new 
method, which would lead to the enrichment of this highly applicable scientific area.  

Situations of emergency are the state of the endangerment of social stability with 
great implications for the life and health of people, the state of material goods and the 
environment. Therefore, every contribution to the improvement of the decision–making 
system of the staff in charge of situations of emergency is also a contribution to 
prevention and reaction in case a danger occurs.  



Komazec and Petrović/Oper. Res. Eng. Sci. Theor. Appl. 2 (2) (2019) 12-23  
 

23 
 

References 

Cupic, M., & Suknovic, M. (2010). Odlucivanje [Decision making]. Belgrade: Faculty of 
organisational sciences. 

Dey, P. K. (2001). Decision support system for risk management: a case study. 
Management Decision, 634-649. 

Karovic, S., & Komazec, N. (2009). Krize i krizni menadzment [Crisis and crisis 
management]. Novi Glasnik. 

Komazec, N., Mladenovic, M., & Dabizljevic, S. (2018). Etiology of the notion of event in 
terms of decision-makingand determination of organizational system risk conditions. 
Decision Making: Application in Management and Engineering, 165-184. 

Komazec, N., Pavlovic, S., & Ilic, M. (2014). Upravljanje rizikom u funkciji prevencije u 
vanrednim situacijama [Risk management in the function of prevention in emergency 
situations]. 16. Nacionalna konferencija "Kvalitet - uslov za bolje poslovanje". Kopaonik: 
AKSS. 

Moriarty, S., Mitchell, N., & Wells, W. (2012). Advertising&IMC: Principles and Practice. 
Yorkshire: Pearsons. 

Nenadovic, G., Kostic - Stankovic, M., & Vucinic, D. (2016). Integrisane komunikacije u 
funkciji promocije vojne profesije [Integrated communications in the function of 
promotion of the military profession] . Vojno delo, 68(7), 330-349. 

Opricovic, S. (1998). Visekriterijumska optimizacija sistema u gradjevinarstvu 
[Multicriteria system optimization in the construction]. Belgrade: Gradjevinski fakultet. 

Pamucar, D. S., Bozanic, D. I., & Kurtov, D. V. (2016). Fuzzification of the Saaty's scale and 
a presentation of the hybrid fuzzy AHP-TOPSIS model: An example of the selection of a 
brigade artillery group firing position in a defensive operation. Vojnotehnicki glasnik, 
66(4), 966-986. 

Pamucar, D., Bozanic, D., & Komazec, N. (2016). Risk Assessment of Natural Disasters 
Using Fuzzy Logic System of Type 2. Management, 23-34. 

Pamucar, D., Bozanic, D., & Randjelovic, A. (2017). Multi-criteria decision making: An 
example of sensitivity analysis. Serbian journal of management, 11(1), 1-27. 

Pamucar, D., Stevic, Z., & Zavadskas, E. K. (2018). Integration of interval rough AHP and 
interval rough MABAC methods for evaluating university web pages. Applied Soft 
Computing, 67, 141-163. 

Petrovic, A., Sakovic, R., & Marinkovic, Z. (2017). Izbor lokacija vatrenog polozaja 
artiljerijske grupe primenom AHP i VIKOR metode [Selection of artillery group firing 
position by aplication of AHP and VIKOR method] . Safety for the future 2017 (pp. 560-
568). Obrenovac: RASEC. 

Radojkovic, M., & Miletic, M. (2005). Komuniciranje, mediji i drustvo [Communication, 
media and society]. Novi Sad: Stylos. 

Saaty, T. (1980). The analytic hierarchy process. New York: McGraw-Hill. 

Zakon o smanjenju rizika od katastrofa i upravljanju vanrednim situacijama [Law on 
disaster risk reduction and emergency management] (Vol. 87). (2018). Belgrade: Sluzbeni 
glasnik. 

Акимов, В. А., & Порфирьев, Б. Н. (2004). Кризисы и Риск: К Вопросу Взаимосвязи 
Категорий [Crises and Risk: Towards a Question of Linking Categories]. Проблемы 
Анализа Риска, 38-49.