99 

 IJAL 4 (2) (2019) 

 

International Journal of Active Learning 
Terakreditasi SINTA 4 

http://journal.unnes.ac.id/nju/index.php/ijal 

 

 

Analysis of Lecturers Competency Performance Evaluation using 

Fuzzy Modeling 
 

Jeffrey Nugroho,  Lilik Linawati, Tundjung Mahatma 

 

Universitas Kristen Satya Wacana Salatiga, Indonesia 

  

Info Articles 

___________________________ 
History Articles: 

Received 8 January 2019 

Approved 11 July 2019 

Published 1 October 2019 

___________________________ 
Keywords: 

fuzzy logic, lecturer 

performance, FIS Mamdani. 

__________________________________ 

Abstract
 

_________________________________________________________________________________ 
In the Indonesian 2005 Law (UU No. 14 Tahun 2005) about Teachers and Lecturers 

there are four competencies that must be owned by a lecturer in carrying out the duties 

–the Tridharma– namely pedagogic, personality, professional, and social competence. 

This research analyzes the evaluation results of lecturer's performance at the Faculty of 

Science and Mathematics using fuzzy modeling, employing the FIS Mamdani method.. 

The twelve aspects assessed in the questionnaire are grouped into four competencies 

with an assessment of 1 (not good), 2 (poor), 3 (good), and 4 (excellent). The average 

value of each aspect of each competency is used to derive the fuzzy value of each 

competency. The value of each competence is then used to get the final value of lecturer 

performance evaluation, to be back-processed into the crisp value which is translated 

through membership function. Based on the criteria set by the faculty, i.e. “Poor”, “Fair”, 

and “Good”, the results show that the performance evaluation of all lecturers are "Fair" 

towards "Good", but when the assessment uses the average value, all lecturer have a 

“Good” performance except one lecturer who shows “Fair” performance. 

 
 Address correspondence: 

Email: t.mahatma@staff.uksw.edu 
p-ISSN 2528-505X 

e-ISSN 2615-6377 

 

 

 

 



 

Jeffrey Nugroho et al. / International Journal of Active Learning 4 (2) (2019) 

100 

INTRODUCTION 

 

The performance evaluation is not only 

needed in the business office, but also in the 

education world to improve the quality of 

education in a sustainable manner. In improving 

the quality of education on an ongoing basis, 

universities conduct assessments, namely 

accreditation and internal evaluation of 

universities. Higher education strives to improve 

the quality including that of the lecturers. To 

maintain the quality of lecturers, monitoring and 

evaluation of the performance of lecturers are 

conducted every year/end of the semester by 

filling out questionnaires by peers, students, and 

superiors. With the results of the performance 

evaluation, lecturers are expected to improve or 

maintain the good performance that has been 

achieved. 

Many instruments with assessment aspects 

can be used to evaluate the performance of 

lecturers. Universitas Kristen Satya Wacana has 

conducted lecturer performance evaluation using 

the “Beban Kinerja Dosen” report –performance 

report based on credit units– as regulated by the 

Indonesian Ministry of Research and Higher 

Education / MENRISTEKDIKTI, for lecturers who 

have been certified. Every lecturer also has to 

report internally using the “BATA” report to 

evaluate the real workload every semester. The 

Quality Assurance unit (PMF) of the Faculty of 

Science and Mathematics designed a questionnaire 

to evaluate the performance of lecturers based on 

competency which refers to the Indonesian Law, 

“UU Nomor 14 Tahun 2005” concerning Teachers 

and Lecturers at the faculty level, involving 

supervisors, peers, administrative staff, and taught 

students concerned. The lecturers' competencies 

in question include pedagogic, professional, 

personality and social competencies. The 

performance evaluation based on lecturer 

competence in this paper hereinafter is referred to 

as lecturer performance evaluation. 

The assessment carried out so far has been 

in accordance with the lecturers' performance 

competence in general, but has not addressed each 

of the four lecturer performance competencies. 

Assessment of each aspect is considered highly 

subjective. Therefore, the assessment is carried out 

by the direct supervisor in question, fellow 

lecturers, supporting staff 

(administrative/laboratory staff), and students 

who took the course. The assessment uses a 

discrete scale of 1, 2, 3, 4, and the average value of 

the twelve aspects assessed, but this average value 

has not been able to describe the performance in 

each competency possessed by each lecturer. 

Therefore, based on the description above, in this 

study fuzzy logic will be applied with the Mamdani 

FIS method to reprocess the evaluation results that 

have been done at the FSM UKSW. A similar 

research that has been done is to obtain the value 

of each competency to determine the prototype of 

the software (Mahmud 2013). Fuzzy logic is widely 

applied in various fields, because the concept of 

fuzzy logic is easy to understand and the 

mathematical concepts that underlie fuzzy 

reasoning are fairly simple. 

According to Kusumadewi (2004), the fuzzy 

set is a set that states an object that can be a 

member of several sets with different membership 

values (µ). The fuzzy set has two attributes, 

namely: Linguistics, i.e. naming groups that 

represent a particular situation using natural 

language, such as young, middle-aged, and old; 

Numeric, which is a value or number that indicates 

the size of a variable, such as 25, 35, 40, 50 and so 

on. 

The fuzzy set attribute is then stated in the 

membership function. The membership function is 

represented by a curve that shows the mapping of 

data input points into membership degrees that 

have intervals between 0 to 1. Some membership 

functions can be used, namely: Linear 

Representation 



 

Jeffrey Nugroho et al. / International Journal of Active Learning 4 (2) (2019) 

101 

 

Figure 1. Linear Representation Goes Up 

Membership Function: 

 
























bx

bxa
ab

ax

ax

x

;1

;

;0

             

…(1) 

 

Figure 2. Linear Representation Goes Down 

Membership Function: 

 

















bx

bx
b

xb

x

;0

0;
0             

…(2) 

 Triangle Curve Representation 

 

Figure 3. Triangle Curve Representation 

Membership Function: 

 





























cxb
bc

xc

bxa
ab

ax

cxatauax

x

;

;

;0



…(3) 

 

 Trapezoidal Curve Representation 

 

Figure 4. Trapezoidal Curve Representation 

Membership Function: 

 































dxc
cd

xd

cxb

bxa
ab

ax

dxatauax

x

;

;1

;

;0



…(4) 

 

Fuzzy logic was first introduced by Prof. 

Lotfi A. Zadeh in 1965. The basis of fuzzy logic is 

the theory of fuzzy sets, namely the important role 

of the degree of membership as a determinant of 

 

 



 

Jeffrey Nugroho et al. / International Journal of Active Learning 4 (2) (2019) 

102 

the existence of elements in a set. The degree of 

membership is the main characteristic in 

reasoning with this fuzzy logic. 

Fuzzy logic can be considered as a black box 

that connects the input space to the output space, 

as shown in Figure 5. 

 

Figure 5. Input-output mapping illustration.  

 

The black box contains a method or 

techniques that can be used to process input data 

into output in the form of good information. 

The method used is FIS Mamdani (Max-Min) 

which was introduced by Ebrahim Mamdani. Fuzzy 

Inference System (FIS) is a system used for 

reasoning with similar principles as humans do 

reasoning with their instincts. According to 

Kusumadewi (2010), there are several things that 

leads to the understanding of fuzzy systems, 

namely fuzzy variables, fuzzy sets, discussion 

universes, and domains. To get Mamdani's FIS 

output, 4 stages are needed: Formation of fuzzy 

sets; Application function implication; The 

composition of rules; Defuzzification 

Based on the description above, the 

formulation of the problem in this study is "How to 

analyze the results of lecturer performance 

evaluation using fuzzy modeling to determine the 

performance of lecturers?". The purposes of this 

study are: Obtain the results of the analysis of 

lecturer performance evaluation using Mamdani's 

Fuzzy Inference System (FIS) model; Obtain the 

results of a comparative analysis of the results of 

the lecturer performance evaluation obtained 

using fuzzy on average. 

 

METHODS 

 

This research is a research on performance 

evaluation based on the competence of lecturers at 

the FSM UKSW Salatiga. The study uses secondary 

data, namely even semester 2016/2017 data 

obtained from the “Penjaminan Mutu Fakultas” 

/Quality Assurance Section, Faculty of Science and 

Mathematics UKSW. In accordance with “UU 

Nomor 14 Tahun 2005” concerning Teachers and 

Lecturers (Chapter 5, Part Five: Guidance and 

Development, Article 69 paragraph 2) there are 

four types of lecturer competencies, namely 

Pedagogic, Professional, Personality, and Social. 

The twelve aspects assessed in the questionnaire 

were then grouped into four competencies based 

on the “Buku Pedoman Sertifikasi Pendidik untuk 

Dosen Terintegrasi” issued by “Direktorat Jenderal 

Pendidikan Tinggi, Kementrian Pendidikan 

Nasional 2011” as follows: 

a. Pedagogic Competence 

 Mastery of media and learning 

technology (Aspect 2). 

 Objectivity in assessing students 

(Aspect 3). 

 Ability to guide students (Aspect 

4). 

b. Professional Competence 

 Mastery of their expertise (Aspect 

5). 

 Willingness to reflect and discuss 

learning problems with peers 

(Aspect 6). 

 Ability to follow developments in 

science and technology for 

updating learning (Aspect 7). 

c. Personality Competence 

 Communication skills (Aspect 1). 

 Wisdom in decision making 

(Aspect 8). 

 Being a role model in thoughts 

and behaviour (Aspect 9). 

d. Social Competence 

Input Space 
Output Space 

Input Variable 
Output Variable 

BLACK 

BOX 



 

Jeffrey Nugroho et al. / International Journal of Active Learning 4 (2) (2019) 

103 

 Ability to express opinions 

(Aspect 10). 

 Ability to accept criticism, 

suggestions, and opinions of 

others (Aspect 11). 

 Being sociable to colleagues, other 

employees, and students (Aspect 

12). 

 

Calculation of lecturer performance 

appraisal use an assessment in the form of discrete 

categorization of an ordinal scale, namely 1 = not 

good, 2 = poor, 3 = good, and 4 = excellent. To 

determine the results of the performance 

evaluation of the lecturers at the FSM UKSW, the 

questionnaire must be filled by the direct 

supervisor concerned, 3 fellow lecturers, 3 non-

faculty members (administrative staff/laboratory 

staff) and 5 students who have been taught by the 

lecturer (regardless of the type of course). This 

questionnaire was first used in the second 

semester of the 2016-2017 academic year. Data is 

processed using the Mamdani FIS method fuzzy 

system. The tool used is Matlab with fuzzy logic 

toolbox. The steps taken are: 

1. Formation of fuzzy sets (fuzzification) 

2. Application implication function: using the 

implication function Min. 

3. The composition of rules: using the Max 

method. 

4. Defuzzification: using the Centroid 

method. 

Data processing that has been done can be 

implemented in the following scheme: 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 6. Lecturer performance evaluation scheme. 

 

RESULTS AND DISCUSSION 

 

Identification of variables is carried out to determine the input and output variables and the 

universe of the discussion needed to calculate and analyze problems. 

 

Table 1. Formation of Input and Output Variables. 

Variable Fuzzy Set Discussion Universes Domain 

Input 

Pedagogic Not Good 0-4 [0, 2] 

Crisp Data 

Pedagogic Competence: 

 Aspect 2 

 Aspect 3 

 Aspect 4 
Professional Competence: 

 Aspect 5 

 Aspect 6 

 Aspect 7 
Personality Competence: 

 Aspect 1 

 Aspect 8 

 Aspect 9 
Social Competence: 

 Aspect 10 

 Aspect 11 

 Aspect 12 
 

FIS 

Mamdani 
Fuzzy values of 

each competency 

Final score (fuzzy) 

FIS 

Mamdani 

Crisp Value 

Membership 

Function 



 

Jeffrey Nugroho et al. / International Journal of Active Learning 4 (2) (2019) 

104 

Professional Poor [1, 3] 

Personality Good [2, 4] 

Social Excellent [3, 4] 

Output 

Performance 

Result 

Poor 

0-4 

[0, 2] 

Fair [1.5, 3.5] 

Good [2.5, 4] 

 

Next, a membership function is formed describing the input variable which consists of a trapezoidal 

curve, a triangle, and a linear rise. Whereas the output variable uses the shoulder curve. 

a. Membership Function for each Pedagogic, Professional, Personality, and Social Input 

Variable 

 

Figure 6. Fuzzy set for competency input variables. 

 

Based on the formula (4), (3), and (1) the membership function is shown as follows: 

 
























2;0

21;
12

2

10;1

x

x
x

x

x
NotGood

       

 ...(5) 

 





























32;
23

3

21;
12

1

31;0

x
x

x
x

xataux

x
Poor

      

 …(6) 

 





























43;
34

4

32;
23

2

42;0

x
x

x
x

xataux

x
Good

      

 …(7) 

 x  

x  

Not Good Poor Good Excellent 



 

Jeffrey Nugroho et al. / International Journal of Active Learning 4 (2) (2019) 

105 

 
























4;1

43;
34

3

3;0

x

x
x

x

x
Excellent

       

 …(8) 

 

b. Membership Function Output Variables Performance Results 

 

Figure 7. The fuzzy set for output variables is Performance Results. 

 

Based on the formula (4) and (3) the membership function is shown as follows: 

 
























5.2;0

5.25.1;
5.15.2

5.2

5.10;1

x

x
x

x

x
Poor

       …(9) 

 





























5.35.2;
5.25.3

5.3

5.25.1;
5.15.2

5.1

5.35.1;0

x
x

x
x

xataux

x
Fair

     

 …(10) 

 
























5.3;1

5.35.2;
5.25.3

5.2

5.2;0

x

x
x

x

x
Good

      

 …(11) 

 

The data processing stage is done by 

building a fuzzy system using the Mamdani 

method. The Mamdani method fuzzy system was 

chosen because this method resembles a human 

mindset where the implication function between 

antecedents and consequences are the same in the 

fuzzy set. In this study, the output obtained is a 

Performance Result that can be used as an 

evaluation material. The stages are explained as 

follows: 

 

Formation of fuzzy sets (fuzzification) 

 x  

x  

Poor Fair Good 



 

Jeffrey Nugroho et al. / International Journal of Active Learning 4 (2) (2019) 

106 

In this stage, crisp values are taken, namely 

the performance value of lecturers from each 

aspect of the assessment to determine the degree 

of membership where the values become members 

of each fuzzy set that is appropriate and used as 

input when forming membership functions. Based 

on the membership function, if, for example, a 

lecturer gets a value of 3.42, it means that the 

value lies between "Good" and "Excellent". Stated 

in a fuzzy set in accordance with formulas (7) and 

(8) the degree of membership can be presented as 

follows: 

  58.0
1

58.0

34

42.34
42.3 






Good
 , 

  42.0
1

42.0

34

342.3
42.3 






Excellent
 . 

Whereas if a lecturer gets a value of 2.88, 

s/he gets a value that lies between "Fair" and 

"Good". Stated in a fuzzy set according to the 

formula (10) and (11) the degree of membership 

can be presented as follows: 

  62.0
1

62.0

5.25.3

88.25.3
88.2 






Fair
 , 

  38.0
1

38.0

5.25.3

5.288.2
88.2 






Good
 . 

As discussed at the variable identification 

stage, each aspect assessed is an input variable in a 

competency which later becomes the output 

variable of each competency. Furthermore, the 

value of each competency is used as an input 

variable to produce output that is the result of 

lecturer performance evaluation. Table 2 shows 

input data as well as the value of each competency. 

 

Table 2. Data processed with fuzzy for competency values. 

Lecturer 

Rated Aspects 

Pedagogic Professional Personality Social 

2 3 4 fuzzy 5 6 7 fuzzy 1 8 9 fuzzy 10 11 12 fuzzy 

D1 3.42 3.75 3.75 3.26 3.83 3.75 3.50 3.23 3.83 3.58 3.50 3.15 3.83 3.58 3.67 3.21 

D2 3.27 3.00 2.91 3.00 3.45 3.10 3.18 3.02 2.91 2.82 2.91 2.88 3.09 2.91 3.00 3.00 

D3 3.67 3.42 3.17 3.09 3.17 3.50 3.50 3.13 3.25 3.17 3.25 3.03 3.33 3.50 3.17 3.07 

D4 3.67 3.50 3.17 3.13 3.08 3.17 3.67 3.02 3.42 3.09 3.36 3.07 3.45 3.45 3.67 3.10 

D5 3.92 3.67 3.67 3.24 3.92 3.58 3.83 3.33 3.75 3.75 3.58 3.26 3.92 3.75 3.92 3.50 

D6 2.83 2.83 3.00 2.79 3.50 2.75 3.08 3.01 2.67 2.75 2.50 2.59 2.92 2.50 2.50 2.50 

D7 4.00 3.50 3.33 3.13 3.75 3.33 3.75 3.29 3.25 3.42 3.58 3.09 3.58 3.42 3.58 3.17 

D8 4.00 3.83 3.58 3.33 3.92 3.58 3.92 3.46 3.58 3.67 3.83 3.21 3.83 3.75 3.42 3.26 

D9 3.83 3.75 3.58 3.26 3.83 3.67 3.75 3.29 3.92 3.42 3.83 3.33 3.83 3.67 3.83 3.36 

D10 3.75 3.92 3.58 3.26 3.83 3.42 3.42 3.09 3.75 3.42 3.83 3.26 3.50 3.50 3.67 3.13 

D11 3.75 3.75 3.67 3.29 3.92 3.83 3.75 3.38 3.75 3.67 3.67 3.24 3.67 3.67 3.92 3.24 

D12 3.42 3.67 3.67 3.21 3.58 3.33 3.17 3.06 3.67 3.50 3.33 3.13 3.42 3.17 3.42 3.09 

D13 3.00 3.42 3.75 3.09 3.50 3.58 2.83 3.13 3.42 3.42 3.42 3.09 3.42 3.58 3.75 3.17 

D14 3.58 3.50 3.67 3.15 3.67 3.36 3.67 3.23 3.50 3.17 3.17 3.02 3.36 3.33 3.92 3.07 

D15 3.42 3.42 3.17 3.09 3.67 3.17 3.50 3.13 2.92 3.25 3.08 3.01 3.08 2.92 2.67 2.88 

D16 3.42 2.58 2.67 2.62 3.33 2.83 3.42 3.06 3.00 3.00 2.75 3.00 2.83 2.75 3.08 2.78 

D17 3.67 3.33 3.25 3.06 3.67 3.25 3.58 3.17 3.00 3.00 2.67 3.00 3.17 3.25 3.42 3.04 

D18 3.50 3.08 3.17 3.02 3.58 3.00 3.67 3.17 3.33 3.08 2.92 3.01 3.25 3.08 3.50 3.04 

D19 3.58 3.58 3.67 3.17 3.83 3.58 3.42 3.17 3.50 3.42 3.50 3.13 3.42 3.67 3.67 3.21 

D20 3.92 3.42 3.50 3.13 3.83 3.25 3.42 3.09 3.58 3.58 3.42 3.17 3.50 3.50 3.75 3.13 

D21 3.75 3.58 3.67 3.21 3.58 3.58 3.67 3.17 3.67 3.75 3.92 3.29 3.75 3.75 3.83 3.31 

D22 3.25 3.42 3.50 3.10 3.83 3.42 3.08 3.09 3.58 3.42 3.42 3.09 3.67 3.08 3.50 3.13 



 

Jeffrey Nugroho et al. / International Journal of Active Learning 4 (2) (2019) 

107 

D23 3.75 3.92 3.75 3.31 3.67 3.67 3.75 3.24 3.92 3.75 3.75 3.31 3.67 3.75 3.92 3.29 

D24 3.50 3.67 3.67 3.18 3.58 3.58 3.67 3.17 3.58 3.58 3.67 3.17 3.58 3.67 3.50 3.15 

 

a. Application implication function 

In this study, the implication function used is 

Min because this function will cut the output of the 

fuzzy set. 

b. The Composition of Rules 

The composition of the rules in this study 

uses the Maximum method. The first stage is to 

obtain the value of each competency; there are 64 

composition rules. Following are examples of some 

composition rules used. 

IF (“Aspect 2” is “NOT GOOD”) and (“Aspect 3” is “NOT 

GOOD”) and (“Aspect 4” is “NOT GOOD”) THEN 

(“Pedagogic” is “NOT GOOD”), 

IF (“Aspect 2” is “NOT GOOD”) and (“Aspect 3” is “NOT 

GOOD”) and (“Aspect 4” is “POOR”) THEN 

(“Pedagogic” is “NOT GOOD”), 

IF (“Aspect 2” is “NOT GOOD”) and (“Aspect 3” is “NOT 

GOOD”) and (“Aspect 4” is “GOOD”) THEN 

(“Pedagogic” is “POOR”), 

  
IF (“Aspect 2” is “EXCELLENT”) and (“Aspect 3” is 

“EXCELLENT”) and (“Aspect 4” is “EXCELLENT”) 

THEN (“Pedagogic” is “EXCELLENT”), 

Whereas in the next stage, 259 composition rules are 

employed to obtain performance results. 

Following are extracts of some composition rules 

that are used. 

IF (“Pedagogic” is “POOR”) and (“Personality” is “POOR”) 

and (“Professional” is “POOR”) and (“Social” is 

“GOOD”) THEN (“Rating Result” is “POOR”), 

IF (“Pedagogic” is “POOR”) and (“Personality” is “POOR”) 

and (“Professional” is “POOR”) and (“Social” is 

“EXCELLENT”) THEN (“Rating Result” is “FAIR”), 

IF (“Pedagogic” is “POOR”) and (“Personality” is “POOR”) 

and (“Professional” is “FAIR”) and (“Social” is 

“FAIR”) THEN (“Rating Result” is “FAIR”), 

  
IF (“Pedagogic” is “FAIR”) and (“Personality” is “POOR”) 

and (“Professional” is “POOR”) and (“Social” is 

“POOR”) THEN (“Rating Result” is “POOR”). 

All rules were made having considered the 

results between antecedents and consequence. For 

example, if a lecturer gets a minimum score of 

GOOD on pedagogic and professional competencies 

and gets a minimum score of FAIR on social 

competence and personality, then the lecturer will 

get a GOOD assessment. 

c. Defuzzification 

To determine the crisp output, a 

defuzzification method is used that matches the 

Mamdani FIS. The crisp output obtained is a score 

of the performance assessment results. In this 

study, the Centroid (Composite Moment) method 

is used. 

Furthermore, testing and simulation are 

carried out to calculate the score of the 

performance assessment of lecturers by using 

fuzzy logic. Simulation was carried out using the 

Matlab software along with fuzzy logic toolbox 

which consists of two stages, namely: 

1. Stage 1 is to get the value of each 

competency, based on each aspect 

assessed in the questionnaire. In this 

stage, the 64 rules were used. The 

following are the figures that show the 

work in stage 1. 

  



 

Jeffrey Nugroho et al. / International Journal of Active Learning 4 (2) (2019) 

108 

Figure 8. FIS editor membership 

functions  input variable aspects of the 

assessment.  

 

Figure 9. FIS editor membership function 

output variable competency. 

  

 

Figure 10. Rule editor for R1-R64. 

 

To test whether the rules is in accordance 

with the objectives to be achieved, the value of 

each aspect in Table 2 were entered to get the 

competency value. Examples of results of 

competency output variables (Pedagogic, 

Professional, Personality, and Social) for L2 are 

presented in the following figure: 

 

Figure 11. Examples of output from 

Pedagogic Competencies. 

 

Figure 12. Examples of output from 

Personality Competencies. 



 

Jeffrey Nugroho et al. / International Journal of Active Learning 4 (2) (2019) 

109 

 

Figure 13. Examples of output from 

Professional Competencies. 

 

Figure 14. Examples of output from 

Social Competencies. 

 

2. Stage 2 is to get the results of the lecturer performance assessment, based on the value of 

each competency from the previous stage. In this stage, 259 rules were used. The following 

are the figures that show the work in stage 2. 

 

Figure 15. FIS editor membership 

function competency input variable. 

 

Figure 16. FIS editor membership 

function output variable Performance Results. 

 

 

Figure 17. Rule editor for R1-R259. 

 



 

Jeffrey Nugroho et al. / International Journal of Active Learning 4 (2) (2019) 

110 

To test whether the rules are in accordance 

with the objectives to be achieved, the value of 

each competency was entered in Table 3 to get the 

results of the lecturers' performance. Examples of 

the results of the lecturer performance output 

variable for L2 are presented in Figure 18. 

 

 

Figure 18. Examples of outputs of lecturer performance. 

 

Based on the testing and simulation, data 

analysis was done by comparing the scores of the 

results of the lecturers' performance evaluation 

which were calculated on average and using fuzzy 

logic. The simulation resulted in the calculation of 

the output of Lecturer Performance Results as 

presented in Table 3. 

 

Table 3. Data from assessment processing based on FIS and average assessment results. 

Lecturers 

Competency 
Rating Result 

(fuzzy) 

Rating Result 

(average) 
Pedagogic Professional Personality Social 

fuzzy fuzzy fuzzy fuzzy 

D1 3.26 3.23 3.15 3.21 3.42 3.66 

D2 3.00 3.02 2.88 3.00 3.45 3.05 

D3 3.09 3.13 3.03 3.07 3.45 3.46 

D4 3.13 3.02 3.07 3.10 3.45 3.45 

D5 3.24 3.33 3.26 3.50 3.38 3.84 

D6 2.79 3.01 2.59 2.50 3.06 2.99 

D7 3.13 3.29 3.09 3.17 3.42 3.70 

D8 3.33 3.46 3.21 3.26 3.39 3.83 

D9 3.26 3.29 3.33 3.36 3.41 3.82 

D10 3.26 3.09 3.26 3.13 3.42 3.74 

D11 3.29 3.38 3.24 3.24 3.40 3.84 

D12 3.21 3.06 3.13 3.09 3.43 3.59 

D13 3.09 3.13 3.09 3.17 3.44 3.47 

D14 3.15 3.23 3.02 3.07 3.43 3.59 

D15 3.09 3.13 3.01 2.88 3.45 3.41 

D16 2.62 3.06 3.00 2.78 3.40 3.22 

D17 3.06 3.17 3.00 3.04 3.44 3.34 

D18 3.02 3.17 3.01 3.04 3.44 3.45 

D19 3.17 3.17 3.13 3.21 3.43 3.62 



 

Jeffrey Nugroho et al. / International Journal of Active Learning 4 (2) (2019) 

111 

D20 3.13 3.09 3.17 3.13 3.44 3.68 

D21 3.21 3.17 3.29 3.31 3.42 3.77 

D22 3.10 3.09 3.09 3.13 3.45 3.62 

D23 3.31 3.24 3.31 3.29 3.42 3.82 

D24 3.18 3.17 3.17 3.15 3.44 3.74 

 

The main focus of the data analysis is the 

fuzzy evaluation results and the results of the 

assessment using average. Before comparing the 

two assessment results, each assessment is first 

examined. The lecturer performance appraisal 

scale on the average value of lecturer performance 

set by the “Penjaminan Mutu Fakultas” is 0 to 4 

with the following details: 

 "Poor", for the score of between 0 to 2. 

 "Fair", for the score of between 2 to 3. 

 "Good", for the score of between 3 to 4. 

Based on the assessment results using the 

average (Table 3), all lecturers has a performance 

title of "Good". However, D6 has the title "Enough" 

because it gets a value of 2.99 which is on a scale of 

2-3. Calculated using the average, the final result 

shows as if D5 and D11 are lecturers who perform 

the most "Good" because they have a value of 3.84 

but, in fact, other lecturers who have grades 

between 3-4 also have a "Good" performance. The 

results of the average assessment appears as such 

that it may be interpreted as the sequence of 

lecturers who get the highest value to the lowest. 

This study using fuzzy modeling is expected 

to eliminate the assumption of the lecturer 

sequence from the highest value to the lowest, or 

vice versa. With fuzzy logic and modeling that has 

been created, results of evaluations using fuzzy 

should be observed in Table 3. By observing Figure 

7 and equation (9) - (11), if one of the lecturer 

values is taken, namely D6 with a value of 3.06 and 

substituted into equation (9) - (11), it yields: 

  006.3 
Poor

   

  44.0
1

44.0

5.25.3

06.35.3
06.3 






Fair
    56.0

1

56.0

5.25.3

5.206.3
06.3 






Good
  

 

Figure 19. D6 performance results. 

 

Figure 19 explains that D6 has performance 

results between "Fair" and "Good". The calculation 

results from D6 are obtained and, it can be 

interpreted that D6 has a performance that is more 

than "Fair", closing to "Good" performance. 

As it can be seen in Table 3 and Figure 7, D2 

with a value of 3.45 is in two fuzzy sets, namely 

"Fair" and "Good". To find out which predicate is 

appropriate, it is first substituted into equation (9) 

- (11) as follows: 

  045.3 
Poor

  

  05.0
1

05.0

5.25.3

45.35.3
45.3 






Fair
    95.0

1

95.0

5.25.3

5.245.3
45.3 






Good
  

x  

 x  

Poor Fair Good 



 

Jeffrey Nugroho et al. / International Journal of Active Learning 4 (2) (2019) 

112 

   

Figure 20. D2 performance results. 

 

The calculation results from D2 are 

  05.045.3 
Fair

  and   95.045.3 Good  can 
be interpreted that D2 has a performance more 

than "Fair" closing to "Good" performance, and 

also shown in Figure 20. If D2 has a value greater 

than or equal to 3.50, then the lecturer has a 

"Good" performance with a membership degree of 

value 1. Simple fuzzy modeling in this study cannot 

be ordered from the highest value to the lowest 

value because the output will return to the very 

fuzzy set. 

D2 and D6 both show performance that 

moves away from "Fair" and approaching "Good" 

performance with their respective membership 

degrees. However, when compared with D2, D6 

must further increase the value of each 

competency to achieve "Good" performance, so 

that the membership level can be one or close to 

one.  

 

CONCLUSION 

From the results of the research and 

discussion in this study, it can be concluded that by 

using FIS Mamdani, each lecturer shows a 

performance evaluation that is more than "Fair" 

and approaches the "Good" performance based on 

the criteria set by the faculty. If the average value 

is used, all lecturers have a "Good" performance, 

except for one lecturer who shows "Fair" 

performance. Fuzzy modeling can thus be used as a 

comparison in processing lecturer performance 

evaluation results which are calculated using the 

average value. To improve or maintain the 

predicate of lecturers' performance evaluation, 

each lecturer can look at the values of each 

competency that is still lacking. 

 

ACKNOWLEDGEMENT 

 

The authors wish to thank the Faculty of 

Science and Mathematics UKSW for providing the 

research data, and the e-SisTem FSM UKSW that 

has supported this research. 

 

REFERENCES 

 

Abdulrahman, U. F. I., Shamsu-Deen, S., & Abaidoo, E. K. 

2017. Assessing Lecturers' Performance using 

Fuzzy Logic. International Journal of Computer 

Applications, 160(1). 

Andani, S. R. 2015. Fuzzy Mamdani Dalam Menentukan 

Tingkat Keberhasilan Dosen Mengajar. In 

Seminar Nasional Informatika (SEMNASIF) (Vol. 

1, No. 4). 

Furqan, M. 2017. Penerapan Fuzzy Inference System 

(FIS) Untuk Menentukan Kinerja Dosen Bidang 

Pengajaran STT NURUL JADID. Jurnal Teknik 

Informatika, 6(01). 

Kusumadewi, Sri., Purnomo, Hari., 2004 : Aplikasi Logika 

Fuzzy Untuk Pendukung Keputusan, Edisi 1, 

Penerbit Graha Ilmu, Yogyakarta. 

Kusumadewi, Sri., Purnomo, Hari., 2010 : Aplikasi Logika 

Fuzzy Untuk Pendukung Keputusan, Edisi 2, 

Penerbit Graha Ilmu, Yogyakarta. 

Mahmud, M. 2013. Prototype Logika Fuzzy Inference 

System dalam Penilaian Kinerja Guru: Studi 

Kasus SMK XYZ. In Seminar Nasional Informatika 

(SEMNASIF) (Vol. 1, No. 4). 

Nugroho, J. 2018. Analisis Hasil Evaluasi Kinerja 

Berdasarkan Kompetensi Dosen Dengan 

Pemodelan Fuzzy. Skripsi. UKSW. Salatiga 

 x  

x  

Poor Fair Good 



 

Jeffrey Nugroho et al. / International Journal of Active Learning 4 (2) (2019) 

113 

Nuryadin, S. 2013. Sistem Pendukung Keputusan dalam 

Menentukan Penilaian Kinerja Dosen dengan 

Metode Fuzzy Database Model Mamdani. 

Electrans, 12(2), 161-170. 

Purwanti, W. N., & Patria, L. 2013. Penerapan Logika 

Fuzzy Pada Penilaian Kinerja Pegawai: Studi 

Kasus Universitas Terbuka. 

Simajuntak, M. 2017. Penilaian Kinerja Dosen dengan 

Menggunakan Metode Sugeno. Journal 

Information System Development (ISD), 2(1). 

Widaningrum, I. (2013). Evaluasi Kinerja Dosen 

Menggunakan Metode Fuzzy Multi-Attribute 

Decision Making (FMADM) dengan 

Pengembangan (Studi Kasus: Universitas 

Muhammadiyah Ponorogo). Semnasteknomedia 

Online, 1(1), 09-61. 

Widaningrum, I. (2015). Sistem Pendukung Keputusan 

Evaluasi Kinerja Dosen Menggunakan Analitycal 

Hierarchy Process (AHP) Dan Fuzzy Analitycal 

Hierarchy Process (Fuzzy-AHP). 

Semnasteknomedia Online, 3(1), 2-2. 

Yazdi, M. (2012). Perancangan Sistem Pakar Fuzzy 

Untuk Menentukan Kinerja Dosen di Perguruan 

Tinggi. SESINDO 2012-Jurusan Sistem Informasi 

ITS, 238.