Implementation of Fuzzy TOPSIS on Expert Systems

for Core Competency Detection

Sutiah

, Supriyono

, Indah Aminatuz Zuhriyah

, Zainul Arifin

1,3

Islamic Education Department, Faculty of Tarbiyah and Teaching Training, State Islamic University of Maulana Malik

Ibrahim, Malang, Indonesia

Informatics Engineering Department, Faculty of Science and Technology, State Islamic University of Maulana Malik

Ibrahim, Malang, Indonesia

Psychology Department, Faculty of Psychology, State Islamic University of Maulana Malik Ibrahim, Malang, Indonesia

Keywords: fuzzy TOPSIS, core competencies, expert system, application, learning.

Abstract: The expert system on core competency detection is a fairly important system development. The study aims to

implement the fuzzy TOPSIS (Technique For Others Reference by Similarity to Ideal Solution) in the core

competency detection expert system, and the student learning achievement was majoring in Islamic religious

education in Islamic Religious College. In the process of learning an Indonesian National Qualification

framework based curriculum designed to detect the level of development of core competencies and student

achievement in the Faculty of Tarbiyah and Teaching Training or Islamic Education study program at Islamic

Religious College. The leading product development of the study is an expert system application with the

fuzzy TOPSIS method in measuring the development indicators of core competencies include hard skills and

soft skills and learning factors that impact the students learning achievement. The result of this expert system

can help students and lecturers independently to detect how the level of development of student core

complications, detecting internal learning factors and external influences, and measuring how they influence

student learning achievements. The results obtained indicate the level of use of fuzzy TOPSIS reaches an

accuracy is excellent.

1 INTRODUCTION

A college is a strategic unit of education in

preparing outstanding human resources and

excellence. Global competition demands that colleges

should be able to deliver their students to increase

competition globally. The college is required to

respond to the development speed of science and

technology and the rapid development of information

technology, which can affect the student's thinking

style more advanced. The development of Islamic

religious colleges in Indonesia is speedy. It is not

apart from the demands of the era of globalization and

the work that requires qualified workforce consisting

of graduate graduates, masters graduates even the

level of competency of doctors who are experts in

their field. It is due to fulfill the demands of

employment competence following the working

standards in the era of globalization that refers to

international working standards. The needs of

workers in the period of globalization not only require

graduates who are competent as a working

requirement, but they are also required to possess

interpersonal skills competence, information

technology skills, problem-solving skills, and other

skills. There are still many college graduates who are

not ready to work according to the demands and

development of continuously changing qualifications

and competencies and in line with the pace of

development of science and technology as expected

by stakeholders. This illustrates that there are still

many colleges, including the state Islamic religious

college, which has the quality of graduates yet not

following the demands of the community of users and

has not been able to competitiveness in the global

community.

The quality of higher education graduates,

including the state Islamic religious college, can be

measured from how the level of achievement of

student learning and the level of the absorption of

graduates in employment following their field of

expertise. The competency level of graduates can

Sutiah, ., Supriyono, ., Zuhriyah, I. and Ariﬁn, Z.

Implementation of Fuzzy TOPSIS on Expert Systems for Core Competency Detection.

DOI: 10.5220/0009321900120017

In Proceedings of the International Conferences on Information System and Technology (CONRIST 2019), pages 12-17

ISBN: 978-989-758-453-4

affect the quality of Islamic college graduates. The

College of Islam must prepare graduates who can

possess core competencies and produce quality

graduates who excel. Excellent human resources can

succeed in the ideals of the nation and state

development. The performance of academic skills,

special skills, and personality attitudes can be affect

students learning achievements. Personality is

essential to be known to everyone so that each

individual can develop and improve their strengths. A

person who has difficulty in developing himself is

likely not knowing all the weaknesses and faults he

controls. Many students do not achieve a learning

accomplishment even fail not because it is not

intelligent, but fail in managing themselves or do not

know how to recognize himself.

The success of achieving high learning

achievements in college lectures become the ideals

and expectations of each student. Significant learning

achievements impact the level of comfort in doing

decent work and can compete openly in global

competition. However, each student faces various

obstacles, challenges, and difficulties so that each

student is not necessarily able to face challenges and

succeed in obtaining the achievement. The facilities

and environment in the learning environment are very

influential in student learning achievements. High

motivation becomes an internal factor in encouraging

students to achieve good performances. While the

climate is conducive and also adequate, facilities as

an external factor can impact the learning system that

can affect students learning achievement in colleges.

The following research discusses the

implementation of the fuzzy TOPSIS method applied

to expert systems in detecting the core competencies

of students. The analysis of student learning experts

can help students and lecturers in decision making to

support the improvement of student learning

achievement. The following research can produce an

expert system application with the Fuzzy TOPSIS

method. The main issues are raised by detecting core

competencies and learning factors that influence the

level of student learning achievement.

The implementation of the fuzzy TOPSIS can

solve complicated problems. The expert systems are

a representation of the performance from the human

brain that seeks to simulate the way human brains

work to replace social work. For that, in the

introduction of patterns to predict, classifying can be

solved using expert systems. Therefore, this

development has adopted the method of the Fuzzy

TOPSIS method, which aims to detect the perception

of students and lecturers in the event of core

complications, internal and external factors in

identifying student learning achievements.

Other methods of use include Fuzzy AHP and

Fuzzy SAW. Fuzzy TOPSIS has an advantage over

Fuzzy AHP, which requires a somewhat complicated

calculation process. The technique also has linguistic

variables. Other researchers also use fuzzy TOPSIS

to other issues. Based on the above, it is a primary

developed method of Fuzzy TOPSIS based Neural

Network on an expert system of detection of core

competencies and learning factors that affect the

achievement of student learning achievement.

Optimization of Fuzzy TOPSIS on core competency,

learning elements, and student learning achievements

designed and created by integrating core

competencies development levels, learning factors,

and student learning achievement from student

perceptions and lecturers in the curriculum learning

process based on the Indonesian National

Qualification Framework. The development of this

application system can help lecturers and students to

avoid failures in learning and directing and help

students can achieve the highest learning

achievement.

2 LITERATURE REVIEW

The theory of the fuzzy set is a mathematical

framework used to present uncertainty, obscurity,

imprecision, lack of information, and partial truth.

The fuzzy logic is an appropriate step in mapping an

input space into an output space. The technique for

Order Performance by fuzzy TOPSIS is a well-known

method in dealing with any Multiple Criteria

Decision Making (MCDM) problem (Ahmad and

Mohamad 2017). Implementation of the Fuzzy

TOPSIS method to solve a problem in MCDM with

the concept that the most effective alternatives were

the closest to the ideal solution positive and the

farthest of the perfect negative solution. The fuzzy

method of AHP and fuzzy TOPSIS can use to set

preferences by having a level of correspondence to

the ideal solution and is primarily developed to

handle data value.

One such technique is to implement a method

most often used in the process of making Multi-

criteria (MCDM). After calculating the rankings

against the solution, as well as finding that the

technology used to offer customers against real-time

and demand knowledge ' is the best solution with a

single rating (Bhosale and Kant 2018). A membership

function is a curve that shows the mapping of data

input points into its membership value that has

Implementation of Fuzzy TOPSIS on Expert Systems for Core Competency Detection

intervals between 0 and 1. Standard function

approaches used in Fuzzy logic utilization. It is one

way to get the membership value on the Fuzzy logic.

The problem of multi-criteria decision making

(MCDM) is the need for an efficient computing

approach. In particular, in a distributed system, the

brokerage agents containing the scheduler and

dispatcher modules have limited time to decide the

scheduling and delivery of tasks over resources as

some tasks may have interdependence and Deadline.

The Fuzzy TOPSIS method is a suitable tool because

it carries a low overhead cost and because of the

robustness of its results and its sacrifice reaches

among a variety of applicable objectives. Also, our

approach used is optimal and measurable, where

applications on large-scale problems have low

overhead costs. However, this module works well in

static and batch processing environments, such as

grid computing that has predefined behaviors. The

only drawback is that it does not fit in dynamic

settings such as dynamic cloud environments due to

varying workloads. The implementation of several

scenarios indicates that the results are appropriate for

a dynamic environment by adding a predictive

method. Related research discusses the development

of Fuzzy TOPSIS using data mining techniques that

will result in better decisions in MCDM issues.

An example is by utilizing data mining techniques

to extract data history, recognizing workload

behavior, and then triggering an increase in Fuzzy

TOPSIS (Shirvani et al. 2017). The problem of multi-

criteria decision making (MCDM) is the need for an

efficient computing approach. In particular, in a

distributed system, the brokerage agents containing

the scheduler and dispatcher modules have limited

time to decide the scheduling and delivery of tasks

over resources as some tasks may have

interdependence and Deadline. The Fuzzy TOPSIS

method is a suitable tool because it carries a low

overhead cost and because of the robustness of its

results and its sacrifice reaches among a variety of

applicable objectives. Also, our approach used is

optimal and measurable, where applications on large-

scale problems have low overhead costs. However,

this module works well in static and batch processing

environments, such as grid computing, that has

predefined functions. The only drawback is that it

does not fit in dynamic settings such as dynamic

cloud environments due to varying workloads. The

implementation of several scenarios indicates that the

results are appropriate for a dynamic environment by

adding a predictive method. Related research

discusses the development of Fuzzy TOPSIS using

data mining techniques that will result in better

decisions in MCDM issues. An example is by

utilizing data mining techniques to extract data

history, recognizing workload performance, and then

triggering an increase in Fuzzy TOPSIS (Shirvani et

al. 2017).

An example of a Fuzzy TOPSIS implementation

is a variable weight linguistic and can be assessed

very low, low, medium, high, very high, and so on.

Fuzzy numbers can also represent linguistic values.

The fuzzy hybrid analytics process and fuzzy

technique for order performance with similarities to

the Fuzzy TOPSIS method to prioritize and rank from

the solutions offered. Fuzzy AHP used to determine

preference weights, and Fuzzy TOPSIS used to rank

the solution (Sirisawat and Kiatcharoenpol 2018).

MADM regarding the proposed issue is not will be

obtained an optimum solution. But the result of this

alignment can be used to systematically evaluate and

reduce the risk of the quality selection of poor service.

In related studies discussed the utilization of

MCDM methods such as on the fuzzy use of VIKOR

and fuzzy ELECTRE to examine alternate locations

and compare results along with the findings gained

(Senvar, Otay, and Bolturk 2016). The sharp value

illustrates individual assessments in the conventional

approach of TOPSIS. It is not always possible to set

a crisp value for personal preference. In such cases,

linguistic forecasts are more likely than definite

benefits, although there may be some uncertainty

related to linguistic judgment but can be solved using

a blurred approach (Singh and Sarkar 2019).

The best alternative not only has the shortest

distance from the ideal positive solution produced but

also has the most extended range from the perfect

negative solution. It is a concept of the Fuzzy TOPSIS

method. Decision-making to solve practical decision

problems with the modeling concepts of multiple

attributes. That's because the idea is simple and

understandable; the calculations are efficient. They

can measure the relative performance of the decision

in a relatively pure form of mathematics. The Fuzzy

TOPSIS implementation procedure has specific steps.

Among them makes the decision matrix normalized.

Also, making a weighted decision matrix normalized,

determining the ideal matrix of positive solutions and

a perfect matrix of harmful solutions, determines the

distance between each other value with an perfect

positive matrix solution and matrix solution Ideal,

specify a preference value for each alternative

(Ahmed et al. 2016).

In determining the weight of each criterion set,

the pressure will be stored in the system database.

Next is determining the alternative rules decision of

any given alternative. Decision rules the alternative

CONRIST 2019 - International Conferences on Information System and Technology

arrangement will be used to calculate the value of an

option provided by each respondent. After calculating

the alternative value decisions, then the next stage is

determining the fuzzy membership used to determine

the fuzzy matrix decision in the future. After that, turn

into the normalized decision matrix. An attendant is

multiplying the predefined criteria weights with the

normalized decision matrix, thus generating a model

weighted normalized decisions. Once obtained the

decision matrix, the most weighted normalization can

be determined as the ideal positive solution and the

perfect solution negative. The final step is to calculate

the criteria distance with the perfect solution in the

fuzzy TOPSIS, the relative importance of alternative

or alternate ratings concerning attributes can be

several fuzzy. Therefore the calculations are done in

fuzzy environmental, and fuzzy operators are used

(Sadi-Nezhad and Damghani 2011).

The main differences between the fuzzy

TOPSIS approach can be summarized in selecting the

normalization method of the decision matrix,

specifying the Fuzzy Positive Ideal (FPIS), and Fuzzy

Solutions Ideal Negative solution (FNIS), calculation

of distances between fuzzy numbers, and the

defuzzification method applied. Most of all, these

research works are presented procedures using the

defuzzification method at the beginning or middle

step of the fuzzy TOPSIS algorithm. This method

converts the fuzzy numbers to the corresponding crisp

values that will cause some rounding errors as well as

possible and significant interference in the final

rankings of various alternatives. TOPSIS requires the

performance rating of each Ai alternative on each Cj

criterion that is normalized.

𝑅











∑











(1)

with i=1, 2, … , m; and j = 1, 2, … , n

3 METHODOLOGY

The methodology of the research is a series of

activities undertaken by systematic and orderly

researchers to achieve the objectives of the study

shown in Figure 1. The research methodology

consists of Identification Problem, literature review,

data collection, requirement specification, design

system, testing, result analysis, recommendations.

Figure 1: Research methodology.

Figure 1 describes several processes to complete

subsequent studies. The first step in this study is to

study and analyze the condition of the Islamic

Education Study program at Islamic State University

of Maulana Malik Ibrahim. The research object

discusses the level of core competency and

achievement in an Islamic study program at Islamic

State University of Maulana Malik Ibrahim Malang.

The results of the analysis can identify and formulate

problems faced by the Department of Islamic

Education at the State Islamic University of Maulana

Malik Ibrahim Malang as a research object.

Furthermore, the study and literature studies in

the study of the field conducted observations more

deeply to the company that used to be a research

object, especially the part to be focused on the

research. As for the literary study, the stage will be

conducted, learning about the literature that supports

the implementation of this research. This step aims to

deepen and understand the theory or method to be

used in the solve the problem. The literature study

was conducted by searching for the literature relating

to the issues faced. The reference used can solve the

problem as a solid foundation in the study. In this

study, the researchers studied was the Related to the

Decision Support System (DSS) and the Fuzzy

TOPSIS method. The next step is data collection.

Identification Problem

Literatur Review

Data Collection

uirement S

ecification

Desi

stems

Testin

Results anal

sis

Recommendations

Implementation of Fuzzy TOPSIS on Expert Systems for Core Competency Detection

Data and information According to the fact the field

is very influential in the research.

4 IMPLEMENTATION OF

FUZZY TOPSIS

The study used questionnaires distributed to 70

respondents among Islamic religious education

students at the Islamic State University of Maulana

Malik Ibrahim Malang as an object of study.

Table 1: Criteria fuzzy TOPSIS

ode Criteria Fuzzy

weights

uzzy

eights

uzzy

eights

Soft Skill

0,57 0,77 0,93

A2 Hard Skill 0,53 0,67 0,87

A3 Reli

ious 0,43 0,60 0,80

Knowledge

0,50 0,60 0,80

Quran

Memorization 0,37 0,47 0,67

Hadits

Memorization 0,40 0,60 0,80

In TOPSIS fuzzy criteria consisting of Soft

skills, Hard skills, Religious, Knowledge, Quran

Memorization, hadith Memorization shows good

results.

Figure 2: Fuzzy decisions

Figure 3: Normalized decision matrix

The result of normalization of fuzzy TOPSIS method

calculations with its six fuzzy criteria TOPSIS shows

good results. Figure 4 shows a weighted normalized

decision matrix.

Figure 4: Weighted normalized decision matrix

5 CONCLUSIONS

The following research uses six criteria consist of

Soft skills, Hard skills, Religious, Knowledge, Quran

Memorization, Hadith Memorization. Testing the

data by using these parameters indicates that each

parameter has a dependency function. The

implementation of the fuzzy TOPSIS method on an

expert system to detect core competencies in the

Department of Islamic Religious Education, Faculty

of Tarbiyah and the Teacher at state Islamic

University Maulana Malik Ibrahim Malang produced

optimal results.

The following research uses the fuzzy TOPSIS

on an expert system to detect core competencies at the

Islamic State University of Maulana Malik Ibrahim

Malang. The obtained results show an

implementation of the fuzzy TOPSIS achieving good

accuracy results.

CONRIST 2019 - International Conferences on Information System and Technology

REFERENCES

Ahmad, Sharifah Aniza Sayed, and Daud Mohamad. 2017.

“A Comparative Analysis between Fuzzy Topsis and

Simplified Fuzzy Topsis.” AIP Conference

Proceedings 1870.

Ahmed, Shameem et al. 2016. “A Comparative Decision-

Making Model for Sustainable End-of-Life Vehicle

Management Alternative Selection Using AHP and

Extent Analysis Method on Fuzzy AHP.” International

Journal of Sustainable Development and World

Ecology 23(1): 83–97.

Bhosale, V. A., and R. Kant. 2018. “Examining the

Solutions to Overcome the SCKFBS Using Fuzzy AHP

and Fuzzy TOPSIS Method.” IEEE International

Conference on Industrial Engineering and Engineering

Management 2017-Decem: 403–7.

Sadi-Nezhad, Soheil, and Kaveh Khalili Damghani. 2011.

“A Modified Topsis Technique in Presence of

Uncertainty and Its Application to Assessment of

Transportation Systems.” International Journal of

Management Science and Engineering Management

6(1): 3–13.

Senvar, Ozlem, Irem Otay, and Eda Bolturk. 2016.

“Hospital Site Selection via Hesitant Fuzzy TOPSIS.”

IFAC-PapersOnLine 49(12): 1140–45.

http://dx.doi.org/10.1016/j.ifacol.2016.07.656.

Shirvani, Mirsaeid Hosseini, Negin Amirsoleimani,

Saeideh Salimpour, and Ahmed Azab. 2017. “Multi-

Criteria Task Scheduling in Distributed Systems Based

on Fuzzy TOPSIS.” Canadian Conference on

Electrical and Computer Engineering (Figure 1): 1–4.

Singh, P. K., and P. Sarkar. 2019. “A Framework Based on

Fuzzy AHP-TOPSIS for Prioritizing Solutions to

Overcome the Barriers in the Implementation of

Ecodesign Practices in SMEs.” International Journal of

Sustainable Development and World Ecology 26(6):

506–21.

https://doi.org/10.1080/13504509.2019.1605547.

Sirisawat, Pornwasin, and Tossapol Kiatcharoenpol. 2018.

“Fuzzy AHP-TOPSIS Approaches to Prioritizing

Solutions for Reverse Logistics Barriers.” Computers

and Industrial Engineering 117(January): 303–18.

https://doi.org/10.1016/j.cie.2018.01.015.

Implementation of Fuzzy TOPSIS on Expert Systems for Core Competency Detection